CDEC-SING Recomendaciones de KEMA relacionadas con la

Transcription

25 de Febrero, 2011
CDEC-SING
Dirección de Operación (DO)
Santiago, Chile
Attn.
Ing. Daniel Salazar
Director DO
Asunto:
Recomendaciones de KEMA relacionadas con la Auditoría de Cumplimiento de
la Norma Técnica de Seguridad y Calidad de Servicio y Recomendaciones
Adicionales sobre Mejores Prácticas para la Implementación de los Sistemas
de Información de Tiempo Real y el Sistema de Monitoreo del CDEC-SINC.
Apreciado Ing. Salazar
En conformidad con su solicitud durante las reuniones previas con Uds., y teniendo en cuenta los
resultados obtenidos durante la auditoría de cumplimiento de la Norma Técnica de Seguridad y
Calidad de Servicio (NTSyCS), Capítulo Nº 4 “Exigencias Mínimas para Sistemas de Información y
Comunicación”, KEMA ha elaborado las siguientes recomendaciones al CDEC-SING. Estas
Recomendaciones se dividen en dos partes, a saber:
1. Recomendaciones relacionadas con la Auditoría de cumplimiento de la NTSyC
2. Recomendaciones Adicionales sobre Mejores Prácticas para la implementación de los
Sistemas de Información en Tiempo Real y el Sistema de Monitoreo
Estas recomendaciones están basadas en la experiencia de KEMA por su participación en
proyectos de auditoría y en organismos de coordinación de la operación similares al CDEC-SING a
nivel mundial.
I–
Recomendaciones de KEMA relacionadas con la Auditoría de Cumplimiento
de la Norma Técnica de Seguridad y Calidad de Servicio
Recomendaciones Generales a la DO del CDEC-SING

KEMA recomienda realizar un análisis y revisión de los procedimientos actuales de
operación, dada su función y responsabilidad como coordinador Independiente de la
Operación del Sistema (Independent System Operator- ISO) ó de la Operación del Sistema
de Transmisión (Transmission System Operator - TSO) relacionados con el Sistema
Interconectado Chileno del Norte Grande. Este análisis de procedimientos tendrá por objeto
permitir que el CDEC-SING sea reconocido como un ISO ó TSO de calidad y talla mundial y
que pueda ser tenido en cuenta, por los diferentes actores del sector, como un referente
KEMA, Inc., 4377 County Line Road, Chalfont, PA 18914 U.S.A.
Tel: +1 215.997.4500 Fax: +1 215.997.3818 info.consulting@kema.com www.kema.com
25 de Febrero, 2011
Página 2
dentro del Sector Eléctrico de Chile. Esta revisión se podrá basar en una metodología de
Análisis de Brechas (Gaps), comparado (benchmarking) con las mejores prácticas
realizadas por otros ISOs ó TSOs a nivel mundial.
Es importante resaltar que sería recomendable hacer las gestiones para que el CDECSING sea miembro del TSO Comparison (www.tso-comparison.com), el cual es un grupo
internacional de comparación de las mejores prácticas de Operación en Sistemas de
Transmisión.
El TSO es un grupo de Operadores de Redes de Transmisión eléctrica con miembros de
Asia, Europa, Sudáfrica, Suramérica y Norteamérica. Su misión es intercambiar información
sobre sus prácticas operativas actuales y futuras con el propósito de hacer una evaluación
comparativa. Cada año es llevada a cabo una encuesta para determinar los requisitos de
personal, las mejores prácticas y/o medidas de desempeño en áreas como las operaciones
del Sistema de Transmisión, incluyendo programación y despacho de la generación,
operación del mercado eléctrico, planeación de las operaciones, tecnología de información,
capacitación etc.
Los resultados de la encuesta anual son una base importante para la comparación del
desempeño y de las mejoras en las prácticas operativas. En algunos países, las
autoridades reguladoras requieren este tipo de información. La información es
estrictamente confidencial entre los miembros, pero los miembros están autorizados a
divulgar algunos datos acordados.
Los miembros del Grupo discuten sus experiencias cada año en un Workshop organizado
por una de las empresas miembro. Actualmente, el TSO está compuesto por 30 empresas
que se califican como Operadores de Transmisión. El TSO es administrado por un Comité
Directivo compuesto por seis miembros elegidos y está asesorado por KEMA.

KEMA recomienda implementar y hacer seguimiento a los procedimientos con mayor
precisión. En comparación con las "mejores prácticas" de los TSO’s en otras partes del
mundo, el CDEC-SING parece estar aplicando los procedimientos de una manera menos
formal. Cada vez más los TSO’s se enfrentan a requisitos más estrictos por parte de sus
reguladores y clientes, lo que significa que deben ser también más capaces de mostrar que
están trabajando de acuerdo con procedimientos apropiados.

KEMA recomienda designar un encargado de verificar los cumplimientos e informar
oportunamente a la Superintendencia sobre todos los asuntos. Esta persona debe tener la
autoridad (formal) para solicitar información de todos los departamentos del CDEC-SING.

KEMA recomienda proponer a la CNE la revisión de la Norma Técnica actual con el fin de
precisar métodos y procedimientos de cumplimiento de tal forma que evite ambigüedades e
interpretaciones sobre los mismos.
25 de Febrero, 2011
Página 3
Sistema de Información en Tiempo Real
Recomendaciones para la DO del CDEC-SING

KEMA recomienda realizar una auditoría de todas las instalaciones con que cuentan los
Coordinados tanto a nivel de equipos de medición (medidores, transformadores de medida)
como de los medios de comunicación utilizados para la transmisión de los datos, y verificar
que el número ID de éstos equipos en campo corresponda exactamente con el nombre
dado en el Sistema SCADA del CDC, de tal manera que no haya incoherencias en las
comunicaciones de los eventos.

KEMA recomienda realizar una auditoría exhaustiva de las señales que son transmitidas
por los Coordinados para efectos de la operación del Sistema Interconectado, teniendo en
cuenta la importancia de ellas para una adecuada coordinación del sistema.

KEMA recomienda realizar una nueva auditoría exhaustiva de los Numerales I.M y I.N
relativos a la medición de la sincronización horaria y de los retardos de la información
enviada al CDC por los Coordinados, en base a una selección, que debe cubrir al menos el
30% de las empresas y, en el mejor de los casos, todos los Coordinados. Los datos de
tiempo real de los Coordinados deben estar sincronizados con la base de tiempo del SI a fin
de realizar un procesamiento preciso y correlacionado de los diversos datos (por ejemplo
de los estados de los dispositivos de maniobra y las alarmas relacionadas con la actuación
de dispositivos de protección) que se encuentran distribuidos por toda la Red así como un
registro de la secuencia de eventos con resolución de milisegundos para realizar un
efectivo análisis y diagnóstico ex-post de las perturbaciones del Sistema Eléctrico.
Asimismo, es muy importante que los retardos del envío de los datos al Sistema SCADA se
mantengan dentro de los límites exigidos a fin de que los tiempos de respuesta de las
variables del Sistema Eléctrico sean compatibles con los tiempos de acción y reacción de la
inteligencia humana pero no más allá de ellos que puedan provocar stress en los
operadores del CDC. Adicionalmente, estos datos van a ser procesados por funciones de
aplicación como el Análisis de Red (Network Analysis) por lo que retardos superiores al
exigido pueden representar time squews incompatibles con el buen funcionamiento de
programas como el Estimador de Estado, por ejemplo, pudiendo provocar impactos en el
monitoreo y análisis de la seguridad del Sistema Eléctrico. Como la DO del CDEC-SING
tiene planes de poner en servicio a corto plazo mejoras en el Sistema SCADA/EMS actual,
entre las cuales se encuentra la implantación de funciones avanzadas como el Análisis de
Red, que dicho sea de paso se encuentran disponibles en el actual Sistema; KEMA
recomienda que los incumplimientos detectados en la presente Auditoría por los
Coordinados sean solucionados a corto plazo a fin de conseguir efectivamente las mejoras
requeridas en la coordinación y supervisión de la operación del SITR.

Cabe mencionar que si bien la medición de la disponibilidad de la información recibida en el
CDC no fue parte de la auditoría, durante las mediciones realizadas para los numerales I.M
e I.N, KEMA constató una alta tasa de pérdida de mensajes en la red con eventos (cambios
de estado) que estaban siendo simulados por los Auditores. De todos los mensajes
25 de Febrero, 2011
Página 4
recibidos con estos eventos simulados, el Coordinado que evidenció mejores resultados
reportó al CDC un máximo de 79% de los eventos, aunque en terreno se registraron todos
los eventos. Durante estas mediciones se verificó también que todas las comunicaciones
(vía protocolos ICCP y DNP) habían sido correctamente configuradas con reporte
espontáneo, Sin embargo, como en la mayor parte de las mediciones se obtuvo retardos
bastante variables, los Auditores sospechan una posible falta de envío de información con
confirmación (acknowledge) por parte de los Coordinados. KEMA recomienda entonces que
la DO verifique si todos los Coordinados están enviando la información con servicios de
confirmación (acknowledge) para justamente evitar que se pierdan los mensajes con
información en casos de colisión o bajo nivel de señal en el mensaje.
Recomendaciones Genéricas para los Coordinados del CDEC-SING
A continuación se incluyen las siguientes recomendaciones genéricas para todos los Coordinados
relacionadas con la sincronización horaria y los retardos de la información enviada al CDC
(Numerales I.M y I.N respectivamente):

Las Unidades Terminales Remotas (UTRs) de los Coordinados deben tener conectados sus
respectivos aparatos GPS, evitando el uso de la hora del aparato GPS del Centro de
Control del Coordinado vía protocolo NTP, cuya precisión de sincronización usualmente
está fuera del rango exigido por la NT.

Los Coordinados deben verificar que durante el envío de información al SCADA del CDC
sus UTRs deben estar configuradas en la forma de envío espontánea (unsolicited
response).

Los Coordinados deben verificar que el envío de la información al SCADA del CDC por
parte de sus UTRs se esté realizando con servicio de confirmación (acknowledge), de
manera que no se estén perdiendo mensajes por congestión.

En caso de que algún Coordinado presente tiempos de vida del dato superiores a los
establecidos en la NT, deben revisar los parámetros de configuración para el envío de
información.

Aunque la sincronización de tiempo puede realizarse por medio de una conexión vía puerto
RS-232, normalmente la precisión de esta conexión está dada por la destreza del
programador para considerar los retardos de tiempo del frame de sincronismo, por lo que
KEMA recomienda en estos casos al Coordinado verificar el algoritmo que sincroniza la
RTU, con el proveedor y a la vez, la secuencia que se aplica para etiquetar los eventos.
Recomendaciones Específicas para los Coordinados Seleccionados del CDEC-SING
Las recomendaciones específicas para los Coordinados seleccionados como muestra para las
mediciones de sincronización horaria y de los retardos de la información enviada al CDC
(Numerales I.M y I.N respectivamente) son las siguientes:
25 de Febrero, 2011
Página 5


Para el Coordinado E-CL:
o
La Unidad Terminal Remota (UTR) de la subestación Antofagasta sobre la cual se
realizaron las mediciones no tenía conectado el GPS de la unidad, sino que usaba la
hora del GPS del Centro de Control del Coordinado, la cual era adquirida vía
protocolo NTP. Como la precisión de esta sincronización está fuera del rango
exigido por la NT, se recomienda al Coordinado E-CL conectar el GPS de esta UTR.
o
Se recomienda al Coordinado E-CL verificar que el envío de la información al
SCADA del CDC se esté realizando con servicio de confirmación (acknowledge), de
Para el Coordinado Minera Escondida:
o

Los parámetros de tiempo de los equipos de la Subestación Coloso estaban
configurados adecuadamente, y los tiempos de refresco de la información también.
KEMA recomienda al Coordinado Minera Escondida, sin embargo, verificar si el
reintento o confirmación del envío de información está considerado, dado que dentro
de las pruebas hubo 76 eventos, de un total de 360, que no fueron refrescados en el
CDC.
Para el Coordinado Gasatacama:
o
Los eventos medidos de la Subestación de la Central Gastacama presentaron una
diferencia de tiempo relevante con respecto al GPS en el CDC. Aunque la
sincronización de tiempo puede realizarse por medio de una conexión vía puerto
RS-232, normalmente la precisión de esta conexión está dada por la destreza del
programador para considerar los retardos de tiempo del frame de sincronismo, por lo
que KEMA recomienda al Coordinado Gastacama verificar el algoritmo que
sincroniza la RTU con el proveedor, y a la vez, la secuencia que se aplica para
etiquetar los eventos.
o
Como al momento de realizar las pruebas, la RTU no estaba enviando mensajes en
forma espontanea (unsolicited response), hubo que configurar esta acción, con lo
cual se pudo obtener la información de los cambios en el SCADA del CDC. KEMA
recomienda al Coordinado Gastacama verificar que el envío de la información al
SCADA del CDC se esté realizando con servicio de confirmación (acknowledge), de
Comunicaciones de Voz Operativas

Numeral II.A.:
o
KEMA recomienda verificar periódicamente que los sistemas de comunicación y
grabación de los 9 Coordinados que tienen Centro de Control y se comunican
directamente con los demás Coordinados funcione correctamente.
25 de Febrero, 2011
Página 6


Numeral II.B:
o
El registro de tiempo de las comunicaciones grabadas no está sincronizado con la
base de tiempo del CDC por lo que no cumple con la exigencia. KEMA recomienda
al CDEC-SING que verifique las razones por las que el registro de tiempo de las
comunicaciones grabadas no está sincronizado con la base de tiempo del SCADA
del CDC y tomar en breve las acciones correctivas correspondientes.
o
Adicionalmente, KEMA recomienda sincronizar las comunicaciones registradas por
los grabadores de voz de los 9 Coordinados que disponen de Centro de Control con
la base de tiempo de su Centro de Control y la del SCADA del CDC.
o
KEMA recomienda estandarizar el tipo de hora que despliega el grabador de voz
para que esté de acuerdo con la base de tiempo del SCADA del CDC. Lo anterior
también aplica a los 9 Coordinados que disponen de Centro de Control.
Numeral II.C.:
o

Numeral II.E.:
o

KEMA recomienda modificar el Procedimiento Tareas y Responsabilidades del CDC,
Artículo 12, a fin de homologar los períodos mínimos de conservación del archivo de
las comunicaciones del canal de voz por 6 meses tanto para las comunicaciones
entre el CDC y los CC de los Coordinados, como de éstos con los COs de los
demás Coordinados.
KEMA recomienda al CDEC-SING verificar permanentemente que para los 9
Coordinados que tienen Centro de Control y se comunican directamente con los
demás Coordinados, si un evento o incidente ocurrido en el SI está siendo analizado
o investigado por la DO o la Superintendencia, respectivamente, y el registro de
comunicaciones de voz se torne una evidencia necesaria para los anteriores
procesos, el citado registro se conserve hasta que dichos procesos hayan concluido
o exista pronunciamiento definitivo al respecto.
Numeral II.G.:
o
KEMA recomienda registrar las pruebas regulares que son realizadas al teléfono
satelital del Sitio de Respaldo del CDC.
o
KEMA recomienda resolver lo más pronto posible las fallas de comunicación que se
han presentado mediante el uso de los teléfonos satelitales de los Coordinados, ya
que durante una emergencia es fundamental que tanto éstos como los del CDC
funcionen correctamente y los Coordinados estén atentos a su llamado.
25 de Febrero, 2011
Página 7
II –
Recomendaciones de KEMA Adicionales sobre Mejores Prácticas para la
implementación de los Sistemas de Información en Tiempo Real y el Sistema
de Monitoreo
Sistema de Información en Tiempo Real

KEMA recomienda a la DO del CDEC-SING contar con un sistema Supervisory Control &
Data Acquisition (SCADA) / Energy Management System (EMS) / Operator Training
Simulator (OTS) con todas estas aplicaciones avanzadas integradas en una única
plataforma y contando con el personal de soporte necesario para el mantenimiento y
expansión de la base de datos de estas aplicaciones y el modelo del Sistema Eléctrico.
El Sistema EMS deberá incorporar funciones para Análisis de la Seguridad del Sistema
Interconectado (Network Analiysis) que incluya aplicaciones de Estimador de Estado (State
Estimator), Análisis de Contingencias (Contingency Analysis) y otras relacionadas, en
tiempo real y programadas automáticamente. De esta manera, el operador será avisado
oportuna y consistentemente sobre la ocurrencia de contingencias críticas. Estas
aplicaciones son estándares dentro de las "mejores prácticas" para un TSO y son
ampliamente utilizadas por los Operadores de Sistema de Transmisión Eléctrica líderes en
el mundo. La frecuencia típica de los cálculos automáticos del Estimador de Estado y del
Análisis de Contingencias es de 2 a 12 veces por hora (es decir, de cada 5 a 30 minutos)
y/o en caso de un gran cambio en el sistema (como por ejemplo la apertura de un
interruptor, desconexión de planta/carga, cambio de la generación/carga superando un
número determinado de MW). KEMA asegura que la apropiada aplicación de estas
funciones facilitará el trabajo del operador y mejorará la calidad de la operación del sistema,
especialmente cuando los sistemas crecen y la carga atendida es mayor.
El Simulador para el Entrenamiento de Operadores (OTS) tornará al CDEC-SING como un
organismo proactivo en este tipo de herramientas que podrán ser utilizadas por las otras
empresas del Norte Grande de Chile, mejorando consecuentemente la calidad de la
operación del Sistema Interconectado.

KEMA recomienda a la DO del CDEC-SING, además, investigar si el Control Automático de
Generación (AGC, por sus siglas en Inglés), puede ser utilizado para balancear la carga y la
generación por instrucción de varias unidades de generación a aumentar y disminuir sus
cargas de una forma continua. Aunque no todos los TSO’s líderes a nivel mundial utilizan
actualmente un sistema central de Control de Carga-Frecuencia (LFC, por sus siglas en
Inglés) o sistema AGC, muchos de ellos están investigando si el LFC puede mejorar la
calidad de su frecuencia y/o reducir el costo de balancear la carga así como reducir la carga
de trabajo de sus operadores. La razón de esta investigación es que los TSOs enfrentan
actualmente escenarios operativos con calidad de frecuencia reducida y mayor número de
instrucciones requeridas para mantener la frecuencia dentro de los estándares aplicables.
Esto se debe al incremento en la cantidad de generación intermitente (principalmente
eólica), implementación de mercados horarios, y el incremento del número de
interconexiones en Sistemas de Corriente Continua de Alta Tensión (High Voltage Direct
25 de Febrero, 2011
Página 8
Current- HVDC) (con la posibilidad de pasos de carga muy rápidos). KEMA no está
afirmando que un LFC/AGC proveerá una mejor calidad de frecuencia y que traerá más
beneficios que costos al CDEC-SING, pero un estudio podría concluir que usar el LFC/AGC
permitirá tener una mejor calidad de frecuencia y ahorrar en los costos de balance de carga
y generación.
Sistema de Monitoreo
La siguientes recomendaciones de KEMA a la DO del CDEC-SING aplican para la elaboración de
los Procedimientos DO relativos al “Sistema de Monitoreo”, considerando la estructura general
observada en otros procedimientos de la DO:



El documento deberá contener los mecanismos y métodos de supervisión y seguimiento del
cumplimiento para lo establecido en los criterios definidos. Esta función de supervisión y
seguimiento del cumplimiento estará orientada a:
o
Monitorear la aplicación y cumplimiento del procedimiento
o
Identificar conductas que constituyan incumplimientos
o
Detectar problemáticas operativas que impidan o dificulten la plena aplicación del
Procedimiento
o
Identificar deficiencias o vacíos en el Procedimiento que deban ser subsanadas
o
Formular las acciones correctivas necesarias para superar cualquiera de las
situaciones anteriores
Los mecanismos y métodos apropiados para ejercer la supervisión y seguimiento
cumplimiento deberán comprender como mínimo los siguientes:
del
o
Establecer medidas de cumplimiento, como índices y resultados de pruebas
utilizados para evaluar el cumplimiento
o
Determinar la manera en que se verificarán las medidas e índices y los
responsables de las verificaciones
o
Determinar las periodicidades con las cuales se realizarán las verificaciones
o
Preparar reportes periódicos con los resultados de las actividades de supervisión y
seguimiento del cumplimiento. Dichos reportes incluirán la descripción de las
situaciones encontradas (hallazgos), los resultados de los indicadores de
supervisión frente a las metas establecidas, análisis de los problemas identificados y
las acciones correctivas propuestas o ejecutadas
Incorporar un procedimiento de investigación de incidentes. Este procedimiento definirá:
o
El tipo de incidentes que serán investigados, por ejemplo:

Incidentes con pérdida de carga mayor que determinado número de MW;
25 de Febrero, 2011
Página 9
o
o

Incidentes con una duración de pérdida de carga mayor que determinado
número de minutos;

Excursiones de frecuencia fuera de la banda especificada en la NT;

Excusiones de Voltaje fuera de la banda especificada en la NT.
Las obligaciones de todas las partes involucradas, incluyendo:

Obligación de suministrar información;

Obligación para realizar parte de las investigaciones;

Obligación de cooperar;

Obligación de la DO de investigar e informar todos los incidentes;

Obligación de discutir, acordar e implementar las lecciones aprendidas.
La metodología de investigación, incluyendo:

Actividades;

Fuentes de información a usar y la manera en que la información será
obtenida de estas fuentes, incluyendo SCADA, grabadores de voz,
registradores de fallas, relés de protección ó Dispositivos Electrónicos
Inteligentes (IEDs) y unidades de medición fasorial o sincro-fasores (PMUs).

Cronograma para la investigación, informes, discusión e implementación de
lecciones aprendidas.
KEMA también recomienda al CDEC-SING específicamente lo siguiente para el Sistema de
Monitoreo:

Realizar un benchmarking para conocer los avances más recientes a nivel mundial en el
tema de la integración de datos provenientes de PMUs, de relés de protección ó IEDs, de
redes de oscilografía digital de estos IEDs o de registradores digitales de fallas (DFR), y de
datos del SCADA, y la aplicabilidad de estos desarrollos para el CDEC-SING

Investigar los usos más efectivos que le dan las empresas eléctricas a nivel mundial a las
aplicaciones de las PMUs y la verificación de su aplicabilidad en el CDEC-SING

Investigar la arquitectura más conveniente para crear un sistema que permita tener
funcionalidades de Mediciones de Área Extendida utilizando Sincro-fasores (WAMS).
Para mayor información a la DO del CDEC-SING, KEMA ha recopilado una serie de documentos y
artículos técnicos que son incorporados como Anexos a la presente carta, a saber:

Anexo 1- Resumen del Sistema de Monitoreo de la Comisión Federal de Electricidad de
México

Anexo 2- Artículo Técnico de KEMA “PMUs y su Impacto Potencial en las Operaciones de
Centros de Control en Tiempo Real”, presentado en el Panel de Energía de Verano del
IEEE en Estados Unidos 2010.
25 de Febrero, 2011
Página 10


Anexo 3- Tutorial de KEMA en Tecnología y Aplicaciones de Unidades de Medición Fasorial
(PMUs), presentado a la Conferencia Internacional de Aplicaciones de Medición de SincroFasores, Brasil 2006. Temario:
o
Conceptos Generales y Definiciones
o
Necesidades de la Industria
o
Beneficios Esperados y Brechas
o
Proyectos y Experiencia de la Industria
o
Arquitectura del Sistemas
o
Desafíos
o
Estandarización, Pruebas y Certificación
Anexo 4- Selección de Artículos Técnicos del CIGRE París 2010, incluyendo:
o
La Aplicación de Monitoreo de Área Extendida (WAMS) al Sistema de Transmisión
de Gran Bretaña para Facilitar la Integración a Gran Escala de Generación
Renovable – National Grid / Scottish Poweer Transmission Ltd. / Psymetrix Ltd.,UK
o
Desarrollo de un Estándar Chino en la Estación Principal de WAMS para Mejora
Adicional de la Capacidad de Monitoreo Dinámico en Tiempo Real – State Grid
Electric Power Research Institute / North China Power Engineering Co..Ltd. / Beijing
Sifang Automation Co.Ltd.-/ China Electric Power Research Institute, China
o
Evaluación de Desempeño del Sistema de Monitoreo de Área Extendida (WAMS)
Coreano bajo Condiciones Operativas de Campo de la Red Eléctrica de Corea –
KDN Co. Ltd. / LSIS Korea / Univ. KERI
Esperamos que estas recomendaciones permitan al CDEC-SING mejorar continuamente en el
desempeño de sus funciones y responsabilidades y realizar un plan de acción al respecto.
Estaremos atentos a responder cualquier inquietud que tengan respecto a los puntos aquí tratados.
Finalmente, aprovechamos para agradecer al CDEC-SING habernos permitido participar en este
importante proyecto.
Atentamente,
David G. Cáceres
Gerente del Proyecto
KEMA, Inc.
25 de Febrero, 2011
Página 11
ANEXO 1
Resumen del Sistema de Monitoreo
de la Comisión Federal de Electricidad de México
SISTEMA DE MONITOREO DE CFE DE MÉXICO
Para la Comisión Federal de Electricidad (CFE) de México, KEMA efectuó el “Estudio y
Desarrollo de Normas de Seguridad y Confiabilidad del Sistema Eléctrico Nacional”,
como parte del cual se desarrolló un sistema de supervisión y registro para monitorear
lo siguiente:
Indicadores
Los siguientes indicadores miden aspectos técnicos directamente relacionados con la
calidad de la operación y desempeño del sistema y la calidad de varios procedimientos
aplicados por el personal operativo:
N°
Indicador
1
Porcentaje de Tiempo en Estado de Operación Normal
2
Porcentaje de Tiempo en Estado de Operación de Alerta
3
Porcentaje de Tiempo en Estado de Operación de Emergencia
4
Número de Cambios de Estado de Operación
5
Indicador de Desempeño de Confiabilidad
6
Indicador de Discrepancia de Suficiencia
7
Índice de Calidad de Frecuencia
8
Número Acumulado de Excursiones de Frecuencia
9
Estándar de Control de Desempeño 1
10
Estándar de Control de Desempeño 2
11
Desviación Típica () de la Frecuencia
12
Variación Acumulada de las Desviaciones de Frecuencia
13
Índice de Calidad de Voltaje
14
Indicador de la Duración de Tensión Precaria
15
Indicador de la Duración de Tensión Crítica
16
Índice de Control de Enlaces
17
Energía en Riesgo
18
Violaciones de Límites de Elementos y Límites de Transferencia en Enlaces
Críticos
19
Control de la Vegetación
20
Energía no Suministrada
21
Índice de Severidad de Interrupciones
22
Tiempo de Interrupción por Usuario
23
Desempeño Operativo
24
Índice de Licencias Autorizadas en Tiempo
25
Índice de Solicitud de Licencias Atendidas
26
Índice de Solicitud de Licencias Autorizadas con Análisis
27
Ejecución del Despacho Económico
28
Tiempo Promedio de Restablecimiento de Líneas de Transmisión
29
Tiempo Promedio de Restablecimiento de Carga
30
Horas de Entrenamiento por Operador
31
Indicador Global de Seguridad de Infraestructura
Criterios de Desempeño
Las normas Mexicanas en sus secciones 7 - Control de Generación y Frecuencia y 8 Operación Normal de la Red Eléctrica, establecen recomendaciones de tipo
cuantitativo respecto a criterios de operación y requisitos técnicos de desempeño del
sistema eléctrico. Por este motivo se llevaron a cabo verificaciones y simulaciones de
dichos valores para evaluar el efecto de la aplicación de los criterios recomendados y
el cumplimiento de los requisitos de desempeño con el equipamiento del sistema.
Igualmente se evaluó el impacto de las recomendaciones propuestas sobre los límites
y márgenes de seguridad empleados en la operación del sistema.
Control de Generación y Frecuencia
Respecto al control de generación y regulación de frecuencia, se validó la factibilidad
de criterios y procedimientos en relación con regulación primaria, regulación
secundaria, regulación terciaria y control de tiempo, y medidas en condiciones de
emergencia.
Criterios Operativos en Estado Normal y Contingencia
Respecto a los criterios operativos definidos para la operación de la red eléctrica
(transmisión y subtransmisión) se realizaron validaciones para determinar en condición
normal, en eventos resultantes en la pérdida de un único elemento y en eventos
resultantes en la pérdida de dos circuitos o múltiples unidades, si se cumplía con
ciertos criterios operativos en estado normal y de contingencia, requisitos de
desempeño, y rangos aplicables de voltaje y frecuencia.
25 de Febrero, 2011
Página 12
ANEXO 2
Artículo Técnico de KEMA “PMUs y su Impacto Potencial
en las Operaciones de Centros de Control en Tiempo Real”,
presentado en el Panel de Energía de Verano del IEEE en Estados Unidos 2010.
1
PMUs and their Potential Impact on
Real-Time Control Center Operations
D. Carty, P.E., Senior Principal Consultant, KEMA Inc. and Member, IEEE; and M. Atanacio, P.E.,
Principal Consultant, KEMA Inc. and Senior Member, IEEE
Abstract-- The application of phasor technology is one of the
most important next steps in the enhancement of power systems
for smart grid technology. Phasor Measurement Unit (PMU)
data is here and firmly established as vital to forensic studies.
The electric utility industry is now moving at an accelerated pace
with new initiatives to bring PMU data into the real-time fold.
This panel paper highlights six areas where PMU data could
impact real-time control center operations.
I. INTRODUCTION
T
he expanded use of synchro-phasor measurements (from
here on phasors) is arguably one of the most important
developments in the shift from current power systems
technology to the envisioned power systems of the future to be
enabled by smart grid technology [1]. Phasor Measurement
Unit (PMU) data is here and firmly established as vital to
forensic studies. The electric utility industry is now moving at
an accelerated pace with new initiatives to bring PMU data
into the real-time fold [2]. This panel paper highlights six
areas where PMU data has the potential to impact real-time
control center operations: (1) Wide-Area Situational
Awareness, (2) State Estimation, (3) Real-Time Voltage and
Transient Stability Analysis, (4) Contingency Analysis, (5)
Special Protection Schemes Design, and (6) Dynamic
Remedial Action Plans Design.
II. PHASOR MEASUREMENT UNIT DATA AND REAL-TIME
OPERATIONS
The installation and implementation of Phasor
Measurement Unit (PMU) data into power systems
measurement networks has grabbed the limelight in recent
months due in large part to funding opportunities arising from
the U.S. Department of Energy Smart Grid Investment Grant
(SGIG) Program [3].
The forensic use of Phasor measurements for postdisturbance analysis of bulk electric system (BES) events has
been at the forefront of the industry for the last few years [4].
Previously unknown and/or inaccessible measurements in the
field are now becoming available through PMUs. These
include frequency, voltage & current magnitudes, voltage &
current phasor angles, and voltage/current phase angle
differences. For instance, phasor data can now “see” system
behavior at lower voltage levels, such as open phases that did
not cause a phase-to-ground fault, which were previously
impossible to detect remotely. But this additional new
visibility is only half the picture. Measurements may be
sampled at rates of 30 times a second or more [5]. They can
also be time-stamped and synchronized with very high
accuracy and millisecond resolution across a wide geographic
area, as is the case for the North American Interconnections
[6]. The data has been very valuable as input to an array of
advanced engineering studies such as voltage & transient
stability, root cause analysis, and post-disturbance analysis.
The Eastern Interconnection Phasor Project (EIPP) has been
instrumental in enabling utilities to take a leading role in the
installation of PMUs and deployment of phasor infrastructure
[7].
The electric utility industry is now moving forward with
new initiatives to bring PMU data into the real-time domain.
Many utilities are performing studies and pilot projects to
determine the value of PMU data in real-time applications and
decision-making. The effort is strongly backed by
policymakers at Department of Energy, Department of
Commerce, National Institute of Standards and Technology
(NIST) and major industry groups such as IEEE, North
American Synchro-Phasor Initiative (NASPI), and the North
American Electric Reliability Corporation (NERC). The
question of late circulating within the industry is “How can
PMU data be used in Real-time Operations?”
A. Wide-Area Situational Awareness
An area receiving a lot of attention lately is Wide-Area
Situational Awareness [8]. The term “Wide-Area” holds
special significance. It conveys the ability to analyze events
that go beyond the borders of a single utility, Independent
System Operator or Regional Transmission Operator’s electric
system into neighboring control areas and regional areas.
Analysis includes the ability to monitor phase angle
differences across intra- and inter-ties. Oscillation patterns can
also be captured and monitored. Real-time identification of
oscillation patterns can be an indicator of incipient stability
problems and provide a more reliable basis for automatic
adjustment of dynamic limits. Equally important is the
application of graphical representations of the information
collected and analyzed to display results to system operators
and engineers. System operators and industry groups alike
need to continue to work on innovative methods to transform
the millions of individual pieces of information into summary
graphical depictions of the data that can display complex
concepts in a visually pleasing way, to be easily consumed by
2
system engineers and operators making the short-tern
planning and command & control decisions. For example,
adaptation of contouring and nomogram [9] information
representation techniques should be explored to facilitate the
interpretation of pre-contingency analysis results as well as
producing near-real-time leading indicators for high risk
conditions.
B. State Estimation
State Estimation has been a common application in control
center operations for over 35 years. It is the first in a sequence
of major network applications that runs to create a consistent,
cohesive representation of the “current” state of the power
system. It executes as frequently as every minute to produce a
starting-point model for applications performing sophisticated
security studies. Much effort has been invested over the years
to improve the quantity and quality of measurement data input
to State Estimation modules. Phasor angles and phase angle
differences are additional data from PMUs that could be used
by the State Estimator [10]. The new data could help improve
network observability as well as the detection of bad data.
Further work is needed to determine how the basic State
Estimator algorithm can best be adapted to use these
measurements.
C. Voltage and Transient Stability
Real-time sequence Voltage and Transient Stability
Analysis applications are also not new to the utility control
center arena. What is new is the ability for these applications
to use phasor angles and phase angle differences as input. This
could allow stability assessments to be performed that would
enable operation closer to the system’s stability limits. The
results could also be used to improve the calculation of
generic transmission limits and transmission constraints to
better ensure stable operation of the power system.
D. Contingency Analysis
Contingency Analysis evaluates “what if” situations by
evaluating multiple contingency scenarios and identifying the
most severe contingencies for operator action. With the
addition of PMU data and some improvements in the basic
algorithm, the application could provide more reliable
recommendations on actions to be taken for pre-contingencies
(before they occur) and post-contingencies (after they occur).
Exploration of how this additional information can help
change the study and prediction of future operating states and
reliability conditions from deterministic to probabilistic [11]
also merits further study.
E. Special Protection Schemes
Special Protection Schemes (SPS) [12] are automatic,
closed-loop field actions that are performed to mitigate the
effects of a forced outage that might otherwise have severe
consequences to the power system. Some utilities operate
advanced application functions that can automatically adjust
the arming of SPSs in response to changing real-time
conditions. The addition of PMU measurements could expand
the use and improve the reliability of these applications to
provide better input information for the arming/disarming
scenarios as real-time conditions change.
F. Dynamic Remedial Action Plans
Dynamic Remedial Action Plans (DRAP) [13] is a function
that executes in conjunction with the real-time security
sequence, to answer the question “Can generation be
redispatched within the time frame of one security constrained
economic dispatch execution to resolve a security violation?”
As with the SPS, the addition of PMU measurements could
improve the reliability of the real-time sequence applications
to determine if the operator has time to adjust a unit output to
resolve a security violation, after it occurs. Both automated
and integrated SPSs and DRAPs could become more prevalent
as the precursors for self-healing grids as PMUs and other
Smart Grid technologies mature over the next decade.
G. Conclusion
PMU data is here and firmly established as vital to forensic
studies. As more PMUs are added and networked, RTO/ISOs
and utilities will have the capability to more closely monitor
the stability of the bulk electric system and coordinate
remedial actions with their neighbors. New more reliable data
updated in time frames close to real-time will be available
enabling the more accurate prediction of future operating
conditions. For this vision to achieve its full potential,
significant changes can be expected to the true and tried
algorithms of many of the advanced network applications that
have been at the disposal of the current generation of
operations power engineers and system operators.
III. REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
Phasor Technology Research Roadmap For The Grid Of The Future,
Eastern Interconnection Phasor Project Executive Steering Group,
February 7, 2006, p. 8
North American SynchroPhasor Initiative; http://www.naspi.org/images/
naspi_map_20090922.jpg; TVA Opens Data Collection Software for
Industry Use, Oct. 7, 2009, http://www.tva.gov/news/releases/
octdec09/data_collection_software.htm
NIST Framework and Roadmap for Smart Grid Interoperability
Standards Release 1.0 (Draft), Office of the National Coordinator for
Smart Grid Interoperability, September, 2009, p. 59 and DOE Recovery
Act - Smart Grid Investment Grant Program, http://www07.grants.gov/
search/search.do;jsessionid=CGgcLDjfnW9GTqHtpJbcK1hyvk1cF4SW
Tvly9pgxlbFySjf9KGYT!-1179711943?oppId=46833&mode=VIEW
The Challenges Of Testing Phasor Measurement Unit (Pmu) With A
Disturbance Fault Recorder (DFR); Krish Narendra, Zhiying Zhang,
John Lane, Ed Khan, Jim Wood; March 2007, p.1; WECC Synchro
Synchro-Phasors from Wide Area Monitoring Systems to Wide Area
Situational Awareness and Controls; Dmitry Kosterev; July 2009, p. 4,
6.
What is a “phasor” anyway?; TRO Glossary; PMU Basic Specification;
NASPI; http://www.naspi.org/resources/pstt/martin_1_define_standard_
pmu_20080522.pdf
North American SynchroPhasor Initiative; http://www.naspi.org/images/
naspi_map_20090922.jpg; TVA Opens Data Collection Software for
Industry Use, Oct. 7, 2009, http://www.tva.gov/news/releases/octdec09/
data_collection_software.htm
3
[8]
[9]
[10]
[11]
[12]
[13]
Estudio y Desarrollo de Normas de Seguridad y Confiabilidad del
Sistema Eléctrico Nacional (Study and Development of Reliability and
Security Norms for the National Electric System of CFE), KEMA Inc.,
November 2008, p. 99
Metrics PMU Impact on Power System State Estimation, Sam Brattini,
John Finney, EIPP Meeting October 14, 2005.
Reliability @ RiskSM A New Paradigm for Assessing Reliability, Ralph
Masiello, John Spare, Al Roark, and Sam Brattini, Published in:
Electricity Journal October 2004.
The Role of Remedial Action Schemes in Renewable Generation
Integration; J. Wen, P.Arons, E. Liu, and Smart Remedial Action
Scheme, S. Wang, G. Rodriguez, IEEE / PES Innovative Smart Grid
Technologies Conference, Gaithersburg, MD, January 19, 2010
Texas Nodal Energy Management System Requirements Specification
For Network Security and Stability Analysis, Version 0.90, December
2006 http://nodal.ercot.com/docs/pd/ems/pd/nsasa/TN.EMS.61C01.Net
workSecurity andStabilityAnalysisReqSpec.doc
IV. BIOGRAPHIES
Manuel Atanacio, P.E. / MBA / BSEE
Manuel Atanacio is a Principal Consultant and Practice Area Coordinator
with KEMA Inc. Mr. Atanacio has 24 years of experience working with
Electric Power Transmission Systems, Bulk Power Generation, Distribution
Systems, Energy Management Systems, SCADA Systems and Market
Operations Systems.
Since joining KEMA, Mr. Atanacio has worked on energy policy and
deregulation related assignments for several US ISOs and TransCos as well as
International clients. Prior to joining KEMA, he worked for 12 years for the
Puerto Rico Electric Power Authority (PREPA) were he held several positions
related to bulk power generation and transmission operations including
Superintendent of Electric System Operations.
Mr. Atanacio holds a Professional Engineer License from the State
Department of the Commonwealth of Puerto Rico and is a member in good
standing of the Professional Engineers and Surveyors of Puerto Rico, Florida
Chapter. He is a Senior Member of IEEE and an inductee of the Beta Gamma
Sigma International Business Honor Society. He earned his Bachelors in
Science in Electrical Engineering from the University of Puerto Rico Mayaguez and a Masters in Business Administration from the Roy E.
Crummer School of Business at Rollins College in Winter Park Florida.
David W. Carty, P.E / MBA / BSEE
David Carty is a Senior Principal Consultant with KEMA Inc. Throughout
his 28 years of electric utility consultancy, Mr. Carty has applied his extensive
knowledge of control center operations and applications to the development of
instructor-led training and reliability compliance programs. His most recent
assignment has been on location implementing network data defect and data
quality procedures in support of the client’s transition to a Nodal Market. He
has also collaborated with a major utility on the development of a “marketlike” environment to promote price visibility of energy and ancillary services
products to the company’s generation, trading, and market operation
organizations.
Mr. Carty is a Professional Engineer in the State of Pennsylvania and a
Member of IEEE. He holds a Bachelor of Science in Electrical Engineering
from the University of Virginia and a Masters of Business Administration
from Temple University.
25 de Febrero, 2011
Página 13
ANEXO 3
Tutorial de KEMA en Tecnología y Aplicaciones de Unidades de Medición Fasorial
(PMUs), presentado a la Conferencia Internacional de
Aplicaciones de Medición de Sincro-Fasores, Brasil 2006.
Temario:
o
Conceptos Generales y Definiciones
o
Necesidades de la Industria
o
Beneficios Esperados y Brechas
o
Proyectos y Experiencia de la Industria
o
Arquitectura del Sistemas
o
Desafíos
o
Estandarización, Pruebas y Certificación
Outline
•
•
•
•
•
•
•
General Concepts and Definitions
Industry Needs
Expected Benefits and Gaps
Industry Projects and Experience
System Architecture
Challenges
Standardization, Tests, and Certification
General Concepts and Definitions
June 5 – 7, 2006
Experience you can trust.
Overview: Synchronized Measurements
• A PMU at a substation measures voltage and current phasors
– Very precise synchronization, with µs accuracy is becoming
standard
– Compute MW/MVAR and frequency
• Measurements are reported at a rate of 20-60 times a second
– Well-suited to track grid dynamics in real time (SCADA/EMS
refresh rate is seconds to minutes)
• Each utility has its own Phasor Data Concentrator (PDC) to
aggregate and align data from various PMUs based on the time
tag
• Measurements from each utility’s PDC is sent to the Central
Facility (e.g. TVA’s SuperPDC) where the data are
synchronized across utilities
Synchronizing Signals Hundreds Miles Apart
PMU 2
1
2
PMU
Phasors on the
same diagram
Indirect
PMU 1
Status Displays
PMU1
Voltage
Trend
Meter 1
Meter 2
Meter 3
Meter 4
Meter 5
PMU2
Meter 1
Meter 2
Meter 3
Meter 4
Meter 5
Formatting Options?
Formatting Options?
Comparison Between SCADA and PMUs
Source: CFE
PMU Measurements
SCADA Measurements
Time
Synchrophasor Definition
• Synchrophasor – Precision Time-tagged Positive
θ
Imaginary
Sequence Phasor measured at different locations
• Phasor Measurement Unit (PMU) – A transducer that
converts three-phase analog signal of voltage or current
into Synchrophasors
θ
Real
t=0
Overview: Synchronized Measurements
GPS
receiver
Analog
Inputs
Phase-locked
oscillator
Anti-aliasing
filters
Phasor X =
16-bit
A/D conv
√2
Modems
Phasor
microprocessor
--Σ
xk(coskθ - j sinkθ)
N
Original algorithm
• Recursive algorithm calculate the
fundamental frequency
component as the phasor
• Assumed the fundamental
frequency is fixed at 60 Hz
• Angles are affected at offnominal frequencies
• This problem has been corrected
in modern PMUs using
frequency tracking algorithms
Measurement Synchronization
• 24 Satellites
• 12 Hour Orbit Time
• Visibility: 5 to 8 Units from Any
Point at Any Time
• Signal: Position, Velocity, Time
• Precise Positioning Service
(PPS):
– 22 meter Horizontal accuracy
– 27.7 meter vertical accuracy
– 100 nanosecond time accuracy
• Performance is 95% Reliable
Synchronization Sources
• Pulses
• Radio
• GOES
• GPS
Industry Needs
June 5 – 7, 2006
Benefits as seen by DOE & FERC
From: Dept. of Energy & FERC, Feb. 2006,
“Steps to Establish a Transmission Monitoring System
for Transmission Owners and Operators within the Eastern
and Western Interconnections”
Wide Area Monitoring, Protection, and Control
(WAMPAC): Industry Needs
• System Vulnerability
– Response to emergencies
(blackouts being the extreme case)
• Emergency operations
•
•
Prevent disturbance propagation:
Planned islanding with well coordinated under-frequency load
shedding scheme; Avoid tripping generators & lines too early; etc.
Faster system restoration (e.g. reclosing the tie line)
Compliance monitoring and reduction in post-mortem
troubleshooting time and effort
– Asset management/Aging infrastructure
• Capacity deferments
• Increase transmission capacity and power reserve management
• Condition assessment
Wide Area Monitoring, Protection, and Control
(WAMPAC): Industry Needs
WECC System Oscillations under stressed
conditions – August 4, 2000
• System Operations
and Planning
– State est. improvements
– Model validation
– Benchmarking, etc.
• Market Operations
– Congestion management
08/04/00 Event at 12:55 Pacific Time (08/04/00 at 19:55 GMT )
105
99
93
Vincent 500kV
88
Mohave 500kV
Devers 500kV
82
Grand Coulee 500kV
76
70
12:55:19.00
12:55:33.93
12:55:48.87
12:56:03.80
12:56:18.73
12:56:33.67
12:56:48.60
Pacific Time
Angle Reference is Grand Coulee 500kV
Nominal Transfer Capability (NTC) based on thermal, voltage, or
stability limitations
Unused transfer capability and lost opportunity dispatch costs
• DG monitoring, protection, and control
WAMPAC Enablers
• Application Modules
– Angular & voltage stability monitoring and control
– Dynamic line models:
–
–
–
–
–
–
Overload monitoring and control and Fault location
Power oscillation monitoring & damping (e.g. PSS)
Critical equipment status and condition monitoring
Frequency and df/dt monitoring and protection
Monitoring machine excitation & governor systems
Adaptive relay settings and protection
Etc.
• Technology
– Integrated system-wide communication infrastructure allowing flexible and secure
–
–
–
–
data collection and transfer where and when needed
Synchronized measurements
Use of standard technology, such as IEC61850, for easier integration,
configuration, engineering, and maintenance
Advanced sensors (line thermal monitors, equipment condition assessment, etc.)
Advanced visualization tools and algorithms
Expected Benefits and Gaps
June 5 – 7, 2006
Phase Angle Monitoring and Control
• Needs:
angle and angle change between buses
– Avoid incorrect out-of-step operation
– Improved planned power system separation
• Benefits:
– Improved real-time awareness,
Relative Phase Angle
– Provide operators with real-time
0
-10
-20
-30
-40
-50
-60
-70
-80
-90
-100
-110
-120
-130
-140
-150
-160
-170
August 14, 2003 Blackout
Normal Phase Angle is approx. -25º
Phase
Angles
Diverged
Prior To
Blackout
incl. neighboring systems
Cleveland
West Michigan
– Improved out-of-step tripping and blocking
15:05:00 15:32:00 15:44:00 15:51:00 16:05:00 16:06:01 16:09:05 16:10:38
– Separate the system on most-balanced way
Time (EDT)
– Assist operator during manual reclosing of tie lines
Source: TVA
• Technology:
– Advanced algorithms using wide-area information
– Visualization tools
– Smart algorithms for instability and coherency detection, separation boundary
– PMU system
• Gap:
– Operator acceptance, incorporation in the utility/ISO process/rules
– System studies and testing
Enhanced State Estimation (SE)
• Need: Use phasors directly to enhance SE
• Benefits: Better network observability; robust performance due to more
measurements; more precise derived calculations
• Requirements:
0.30
100.00
% Angle Variance Reduction
Angle Variance
– Evolutionary solution
Apply ‘meter placement’ methods to
determine most beneficial PMU locations
0.20
60.00
0.15
40.00
0.10
20.00
0.05
• Actively pursued by major EMS vendors
0.00
0.00
– Revolutionary solution: All-PMU State Calculation0 0.5 1 1.5 2 2.5 3 3.5 4
Entropy Reduction
• State estimate available much more frequently
Shannon entropy analysis is
• Massive PMU deployment (30% - 50%) of buses
promising technique
• Foundation for “closed loop” control
• Cost-benefit analysis required for justification for each user
– “Equivalent” solution: ISO/RTO applications use PMUs for “boundary
conditions” for utility state estimators
Degrees^2
•
Add phasor measurements to existing SE
measurement set
80.00
%
•
0.25
Enhanced State Estimation (SE)
• Gap:
– Need for more PMUs to realize benefits
– Measurement-error (accuracy) analysis for combined traditional
telemetry and PMUs
•
What is redundancy with both traditional telemetry and PMU
measurements?
– As conventional SE uses app. 10s window, what is the level of
–
–
–
–
improvements with PMUs?
Time skew impact to be quantified
Bad data detection (robustness) may be affected by accuracy issues
Will positive sequence measurement help as existing telemetry uses
one or two phases?
Further develop “linear” SE application
Scope and nature of SE enhancement is
system/customer dependent
Post-Disturbance Analysis and
Compliance Monitoring
Loss of Palo Verde Units 1, 2 & 3
• Need: To reconstruct sequence of events
on June 14 , 2004
Frequency plots from different PMUs
after a disturbance has occurred
• Benefits: Savings in troubleshooting
time (several orders of magnitude)
and resources
• Technology:
– PMU with low comm. requirements
Source: SCE
– Data storage in substations
– “Smart Analyzer” to sift through vast amount of data for key info
from an assortment of data loggers (DFR, SER, Relays,
PMUs,…)
• Gap: No commercial products exist as Smart Analyzer
th
06/14/04 Event at 07:40 Pacific Time (06/14/04 at 14:40 GMT )
60.025
59.820
59.615
59.410
59.205
59.000
14:40:34.00
14:40:44.00
14:40:54.00
14:41:04.00
Pacific Time
14:41:14.00
14:41:24.00
14:41:34.00
VINC
JDAY
DEVR
MALN
BGCR
COLS
ALAM
BE23
SONG
BE50
KRMR
SYLM
DEV2
MPLV
ANTP
KEEL
VLLY
CPJK
MAGN
SUML
LUGO
SLAT
GC50
SCE1
Under-voltage load shedding
x
•
Operates when the voltage drops to
a certain pre-selected level for
a certain pre-selected time period
•
UVLS is usually set with a longer (2
– 10 s) reaction time compared to
SVC / STATCOM (0.1 – 1 s)
•
Voltage recovery to be studied
•
Use of other measurements:
–
–
•
Line and generator status
Dynamic VAR reserve at
generators
– Etc.
Deployed world-wide
Under-voltage
relay operates
#1
r
#2
Voltage
instability region
Issues with voltage as an indicator
of voltage instability:
#1: UV relay trips unnecessarily
#2: UV relay fails to trip
Voltage Instability Predictor*
•
•
•
V
Maximal power transfer ⇔
|Zapp | = |ZThev | is point of collapse
Measuring the proximity to instability
- improvement to UV LS
Corridor version: Two PMUs on the
both side of the line
–
More accurate Thevenin
equivalent
E
ZThev
Thevenin
Zapp
load
VOLTAGES
1.2
stable voltage (local measurement)
1
Point of
collapse
0.8
0.6
0.4
unstable voltage
0.2
0
1
(calculated by relay)
1.2
1.4
1.6
load parameter, λ
* K. Vu and D. Novosel, “Voltage Instability Predictor (VIP) - Method and
System for Performing Adaptive Control to Improve Voltage Stability in Power
Systems,” US Patent No. 6,219,591, April 2001.
1.8
Real-Time Congestion Management
• Need: Improve calculation of real-time path flows and increase
transfer limits for optimal market dispatch
• Benefits: Avoid large congestion costs
– Avoids unused transfer capability and lost opportunity dispatch costs
through more accurate real-time ratings
– Experience from real-time ratings will help hour-ahead, and day-ahead
limits
– Leads to better utilization of generation resources and less load
curtailment
• Technology:
– Adequate visibility of corridors with incorporation of improved basic
modules to EMS/SE: Angular stability, Voltage stability, Thermal
constraints
– PMU applications
• Gap:
– Industry and staff adoption of new rules and procedures and PMU-based
calculations
Wide-Area Power System Stabilizer (PSS)
• Need: Generator control to suppress
low-freq. oscillations in interconnected
grids
A
• Benefits: Better system damping by
feeding multi-input PSS with wide-area
signals
B
• Technology:
– Selection of signals; design and tuning of
algorithm
C
– Fall-back scheme: use local signals when
remote ones are disrupted
• Gap:
– Dedicated communications link
– Quantified benefits of WA-PSS
A: Conventional PSS.
B: Multi-input PSS; local signals.
C: Multi-input PSS; wide-area signals.
Source: Hitachi.
Dynamic Line Models
•
Need:
1.
2.
•
Dynamic rating by real-time assessment of transmission lines
thermal limits
More accurate line parameter detection for accurate faultlocation
Benefits:
1.
2.
•
Operator can determine the proper loading
Faster restoration for permanent faults and better detection of
week spots for temporary faults
Technology:
–
–
–
•
Sagometers
Temperature measurements
PMUs in substations
Gap:
–
Industry acceptance
Power-System Restoration
• Need: Use of phase-angle monitoring to assist operator during
restoration
• Benefits: Time savings
– Operator knows if it is feasible to
reclose the tie line
– Valuable tool for operator who works
under stress to reenergize grid.
• Technology:
– PMU system
• Gap:
– Operator training required
– Simulators need to provide trainee with feedback signals that
simulate direct measurements
Monitoring/Protection/Control for DG
• Need: Better monitoring /protection/
control methods
• Benefits: Determination of unintentional
islanding
• Technology:
– A pair of PMUs has been shown to
detect islanding cases where local-based
methods could not
• Gap:
– Field experience still lacking
– Cost requirements
Adaptive Protection
• Need: To use synchronized phasors to allow relays to adapt to
prevailing system conditions
• Benefits:
– Line relays: to better handle complex configurations (e.g., multi-terminal
lines, series-compensated lines)
– Adaptive Security & Dependability to avoid cascading (2 out of 3)
– Improved backup protection
• Technology:
PMU
– PMU signals
– Advanced algorithms
Zone of
Protection
• Gap:
– More field experience needed
– Acceptance by engineers
PMU
Controller
Dynamic Relay Settings
• Needs:
– Reduce complexity of implementation, maintenance, testing,
and verification of relay settings with multi-function IEDs
– Avoid that equipment protection operates incorrectly under
stressed system conditions not set and designed for
• Benefits: Ease of applying and changing settings with IEDs
– Automated review and update of relay settings as system
conditions change (e.g. load growth, new equipment
installations)
– Dynamic setting adjustments under stressed system
conditions (e.g. line overload, voltage and angular instability)
• Technology:
– Enterprise level process and tools
– WAMS high-resolution “system data” data, detect stressed
conditions and system changes
– First level alarm => Second level automated adjustments
• Gap: Industry acceptance
Industry Projects and Experience
June 5 – 7, 2006
Deployment Status
• Synchronized Measurement (SM) and Synchronized
Phasor Measurement (SPM) devices are available from
Phase
many vendors
– ABB, AMETEK, Arbiters, GE,
Macrodyne, Mehta Tech, SEL, …
• Systems are already
installed and operating
• Large scale deployment
– WECC, EIPP, ONS-Brazil, etc.
• New IEEE C37.118 standard
Angle
+30
+20
+10
+00
-10
-20
-30
Source: A. Phadke, VT
has been approved
• Many ongoing SM/SPM application researches/studies
Eastern Interconnection Phasor Project (EIPP)
Under DOE leadership, EIPP participation has been
unprecedented:
• Number of utilities:
¾ 32
• Number of research organizations:
¾ 14
• Number of vendors:
¾ 27
• DOE investment in EIPP:
• Industry investment in EIPP:
• Future DOE investment needed:
• Number of years needed:
Source: EIPP
¾ $3 million (since 2002)
¾ $15 million (5 to 1 leverage)
¾ $5 million (yearly)
¾ 5 years
Eastern Interconnection Phasor Project
(EIPP)
EIPP
EIPPPMU
PMUCompanies*
Companies*
• •Ameren
Ameren
• •AEP
AEP
• •American
AmericanTrans.
Trans.Co.
Co.
• •ConEdison
ConEdison
• •Entergy
Entergy
• •Excelon/ComEd
Excelon/ComEd
• •Excelon/PECO
Excelon/PECO
• •First
FirstEnergy
Energy
• •Hydro
Hydro11
• •LIPA
LIPA
• •Manitoba
ManitobaHydro
Hydro
• •METC
METC
• •Midwest
MidwestISO
ISO
• •NY
ISO
/
NYPA
NY ISO / NYPA
• •PPL
PPLCorp.
Corp.
• •Southern
SouthernCompany
Company
• •TVA
TVA
(proposed)
(proposed)
* Companies with PMUs Planned or In Service
34
PMUs offer Wide-Area Visibility
35
RTDMS VISUALIZATION – SAMPLE DISPLAY
compare angles
selected
Monitor :
- voltage angles
and magnitudes
- color coded
- quickly identify
low or high
voltage regions
historical tracking
and comparison
over specified time
duration
voltage angle and
magnitude tracking
at selected location
Source: EPG
Conceptual Proposal for Build-out of a WECC
Synchronized Phasor Network
Phasor-Assisted State Estimation, NYPA/EPRI
• Goal: with PMU data, State Estimation
can be solved non-iteratively delivering
much improved performance.
• Experience:
– First PMU installed in 1992; now 6+ units
in NY State
– On-line data streamed from PMUs to the
EMS computer via dedicated
communication channels
– Modified the traditional State Estimation
Source: Bruce Fardenesh, NYPA
algorithm
– Tested to confirm improvements to the
traditional SE
– Adopted phasors as integral part of the
EMS
Entergy/TVA PMU-SE Project Objectives
• Phase 1: Benefits using PMU measurements in the State Estimator
Partners: Entergy, TVA, AREVA
– Off-line case studies with captured real-time data from TVA and
ENTERGY control centers
– Use captured real-time PMU data synchronized with SCADA
– Demonstrate results
• Phase 2: Online EMS SE Demonstration
Partners: AREVA, TVA, Entergy, PG&E, and Manitoba Hydro with
expressed interest from Idaho Power, WECC, First Energy, and BPA
– Automate transfer of PMU/PDC data to EMS
– Selection of PMU data relevant to current SCADA data for SE
– Test online TVA State Estimator using PMU measurements from TVA’s
Super Phasor Data Concentrator
– Assess and quantify benefits using online performance metrics
– Implement & demonstrate at TVA control center, on a parallel (nonoperational), online SE which uses PMU data
State Estimation-PMU Data exchange-Phase 1
TVA-Super PDC
PMU
PMU
EMP2.3+
PMU Data
Processing
Measurements
Time point Tables for
all PMU
¾30 samples/second
Case #1
Processed data
PMU
PMU
Applications Input
Case #2
“5” minutes
Monitoring
Case #3
PMUData
Data
PMU
Converter
Converter
ENTIRE
Savecase
Case #4
……
….
SCADA
TVA/ENTERGY’s Real-Time
State Estimator
Time “T”
¾TVA-ICCP (60 s)
¾ENTERGY~2-4 secs
Source: TVA
Real-time
State
Estimator
Study
State
Estimator
RTNET
Savecase
SE
Statistics
GRID
GRID
(XLS)
(XLS)
Wide-Area Stability and Voltage Control System (WACS)
• On-line demonstration project
• Inputs from 8 PMUs (2005)
• Fiber optic communications
(SONET)
– Data rate: 30 packets per second
• Existing Remedial Action Scheme
(RAS) transfer trip from control
center to power plants and
substations
• Computer at control center:
Power System
Disturbances
switch capacitor/reactor banks
direct
detection
(SPS)
trip generators/loads
Power
System
Dynamics
Continuous
Feedback
Controls
(generators)
Discontinuous
Controls
response detection
∆y
– LabVIEW real-time HW and SW
– Algorithms based on: voltage
magnitudes and generator VARs
– Actions: Generator tripping and
capacitor/reactor switching
(WACS)
“Model studies predict that when WACS is fully accepted, an additional 300 MW could be routed down
the Pacific Intertie, resulting in a conservative estimate of $7.2 million per year benefit.” Source: BPA.
Wide-area PSS
• Iceland has a strong 220kV
#1
C
E
C
E
Laxá 28 MW
Krafla 60 MW
Rangárvellir
Laxárvatn
Varmahlíð
Bjarnarflag 3 MW
Mjólká
Geiradalur
Blanda 150 MW
Glerárskógar
Hryggstekkur
Hrútatunga
Vatnshamrar
Brennimelur
Korpa
Geitháls
Teigarhorn
Sultartangi 120 MW
Hrauneyjafoss 210 MW
Vatnsfell 90 MW
Sog 89 MW
Hólar
Sigalda 150 MW
Hamranes
Búrfell 270 MW
Transmission lines
220 kV
132 kV
66 kV
Prestbakki
#2
C
E
Hydro power station
Geothermal power station
Substation
Power intensive industry
150 km
grid connected to a weak
132kV ring.
• Power oscillation occurs
when ring is opened (due to
line fault).
• Two PSS designs have been
studied for Plant #1:
– Conventional PSS -- use (local)
shaft speed as input.
– Wide-area PSS -- use remote
Source: Landvirkjun (Iceland’s National Power)
Real-life recording;
Plant#1 has no PSS
signal (PMU#2’s freq.) and local
signal (PMU#1’s freq.) to
produce ∆f as input
Simulated local and
wide-area PSS at
Plant#1
WAMS as a tool during UCTE Reconnection
• Wide Area Monitoring
system provided more
confidence and security
during the reconnection of
UCTE:
– Zone 1: Green
– Zone 2: Blue
50.10 Hz
green: f in Greece
Blue: f in Switz.
49.90 Hz
+180 deg.
• Critical grid
oscillations/separations
could be detected fast
Phase Angle
-180 deg.
09:34AM
Sources: ETRANS, UCTE
Synchronized Phasor Measurement System, Brazil
Gen Capac 80GW;
Max Demand 59GW;
50,000mi of TL 230kV or above
• 1999: ISO study of a PMU-based
recording system for:
1,800 mi.
–
–
Post-disturbance analysis (inter-area oscillations)
Dynamic model evaluation
• 2003: Experiment project by university:
–
–
–
3 PMUs and 1 PDC; All locally made.
Monitoring 3-ph distribution voltage at three
universities in Southern Brazil.
Applications: Frequency monitoring, disturbance
detection, phase-angle monitoring.
• 2006: Brazilian ISO, “ONS”, prepares
for wide-area deployment:
–
“Specification of the Phasor Measurement System”
as a blueprint for how the system will be built.
–
Local utilities will buy and install PMUs and PDC
according to the blueprint’s specs.
Master PDC at the ISO control center.
Anticipated uses include: forensic analysis of grid
disturbances; validation of model parameters;
evaluation of protection-system performance.
2,030 mi.
–
–
WAMS in China
• 15% annual growth in consumption;
Generation and tie lines are being added:
– Interconnecting of six regional networks have
rendered challenges to operations
– Low-freq. oscillations; Volt/VAR problems
• Power shortage costs economy 2
Sources: CEPRI, China
BUSD/year
• Systems and Apps under development
• 10 WAMS; PMUs--80 installed, 60
planned
System Architecture
June 5 – 7, 2006
Architecture Today
• Most installations consist of one-PDC architecture
with a limited number of PMUs
• WECC and EIPP systems
– Multiple PDCs with a master data concentrator
• The master data concentrator
– Aggregate real-time PMU data and rebroadcast to other PDCs
– Provide online/archived data for non-real-time applications
– Custom developed
– Evolved from interconnecting single-PDC based systems
of the participating utilities
TVA SuperPDC Architecture (EIPP)
• System
performance
depends on the
weakest link (e.g.
low-performance
PDC connected to
SPDC will affect all
users)
• Time delay about 5
seconds
• Mainly perform data
archiving and
rebroadcast
System Architecture - Today
Example
Substation
Super Phasor Data
Concentrator (SPDC)
at TVA (EIPP)
MI
Utility Owned PDC
Data Stream Collection and
Analysis Server for control and protection.
T
N
Utility Enterprise
Service Providers
Other Substation
LANs
SCADA/EMS
Corporate WAN
Line A Relay 1
IEC 61850
& DNP 3.0
via Primary and Backup
Data Communications Services:
-Utility owned WAN, and/or
-Common carrier MPLS service
= VPN defined WAN
PMU 1
COMTRADE / IEEE
C37.118 §
Managed Optical Ethernet Switches - LAN 1
IN
SUBSTATION
Xfmr Relay 1
IEC 61850
& DNP 3.0
Bus Relay 1
IEC 61850
& DNP 3.0
Substation
Automation Host
Physical and electrical isolation of redundant protection systems
Line A Relay 2
IEC 61850
& DNP 3.0
Monitoing IEDs
Serial Comms
Protocol
PMU 2
COMTRADE / IEEE
C37.118 §
Local Historian
Local
HMI
Managed Optical Ethernet Switches - LAN 2
Xfmr Relay 2
IEC 61850
& DNP 3.0
Bus Relay 2
IEC 61850
& DNP 3.0
Routers
DFR Data Host
Wired connections for PMU
synch and 1 ms IED time stamp
synch
GPS Clock
PG&E – Improvements on Remedial Action Scheme
Controller-B
Controller-A
GPS Clock
OPC
OPC
OPC
OPC
Host Computer
IEC 61850
SOE
OSC
&
Hub
Ac
tio
Alternate
Primary
ns
Scheme B
Tele-Protection and SCADA
Network
Scheme A
St
Hub
atu
s
&
SOE
OSC
S ta
tus
ns
IEC 61850
Ac
tio
Host Computer
GPS Clock
Substation
Substation
GPS Clock
Hub
Hub
Watts
Freq.
Temp.
Thermal
Phasor
V/A Status
Control
Source: Vahid Madani, PG&E
V/A
Status
Watts, VAR, Freq., Temp.,
Thermal, Phasor
Control
System Architecture
• How to connect SMs/SPMs with Applications?
Application
management?
Data flow
management/
optimization?
Archive/access
management?
Applications
SMs/SPMs
Device
management?
System
Architecture?
Initial cost
Performance
Operating cost
Flexibility
Other costs
Ease of use
PDC Status
• Lack of mature off-the-shelf PDCs
– Custom developed PDCs
– Vendor PDCs: Not fully productized
• Limitations unknown
• Interoperability with other PMUs/PDCs
• Limits of a master PDC – max. number of PMU/PDC
data streams that it can process?
– Varies depending on types of PDC, and Data volume (# of
phasors/data and data rate) and Processing tasks
• Pros/cons of using intermediate PDCs
– Data flow, latency, bandwidth, configuration, etc.
Need for New Architecture
• Standardized system architecture design
–
–
–
–
Meet the diverse requirements of different applications
Enable data sharing Æ minimize overall cost
Use off-the-shelf products (e.g. process automation)
Be supported by available communication infrastructure
• Bandwidth, protocol, latency
– Can be easily integrated and configured
•
•
•
•
Highly scalable and flexible
Reliable and secure
Easy to install, operate, and maintain
Easy to interface with other systems
Challenges
June 5 – 7, 2006
Challenges
• Disparity among algorithms used by PMU vendors (e.g. phase
angle calculations)
60.5
• Challenges for data analysis
•
•
•
•
C37.118 is for Steady State Operation
60
59.5
Frequency(Hz)
– Disparate sampling rates
– Disparate filtering techniques
– Data compression practices
• Unaccountability of
instrumentation errors
Freq. measured by the proposed algorithm
Freq. measured by PMU A
Actual freq.
Freq. measured by PMU B
59
58.5
58
57.5
19.2
19.4
19.6
19.8
20
20.2
20.4
Time(s)
Visualization of vast amount of data
Secure and non-corrupted data through data links
Scalability: Design architecture to accommodate application
additions
• High accuracy and data bandwidth requirements
20.6
20.8
Phase measurement vs. frequency
PMU angle measurement error
100
80
Phase angle - degrees
60
40
20
0
-20
-40
-60
-80
54
56
58
60
62
64
66
Frequency - Hz
4 PMUs show difference in phase angle at different frequencies.
Example of importance of PMU testing and standards development.
Source: Ken Marin, BPA
Transducer Accuracy - ANSI
ANSI CT
Type
Load
Current
Max. Magnitude
Error pu
Max. Phase
Error (degrees)
Max. Phase
Error (µs)
Relaying
10 to 2000%
0.10
Not tested
Not tested
Metering 1.2
10%
0.024
2.08
96
100%
0.012
1.04
48
10%
0.012
1.04
48
100%
0.006
0.52
24
10%
0.006
0.52
24
100%
0.003
0.26
12
Metering 0.6
Metering 0.3
ANSI
PT TYPE
Max. Magnitude ±
Error P.U.
Max. Phase Error
(± degrees)
Max. Phase Error
(± µs)
Relaying
0.1
Not tested
Not tested
Metering 1.2
0.012
2.08
96
Metering 0.6
0.006
1.04
48
Metering 0.3
0.003
0.52
24
* T.K. Hamrita, B.S. Heck, and A. P. S. Meliopoulos; "On-Line Correction of Errors Introduced by Instrument Transformers in
Transmission-Level Steady-State Waveform Measurements", IEEE Trans. on PWDR, Oct. 2000.
Transducer Accuracy - IEC
IEC CT Type
RELAY TYPE 10P
Relay Type 5P
Metering Type
1.0 Accuracy
Metering Type
0.5 Accuracy
Metering Type
0.2 accuracy
Metering Type
0.1 Accuracy
Load
Max. Magnitude Error ±
P.U.
Max. Phase Error ± degrees
Max. Phase error ± µs
100%
0.1
Not tested
Not tested
max. limit
0.5
Not tested
Not tested
100%
0.3000
2.000
92.6
max. limit
1.0000
2.000
92.6
5%
0.0300
6.000
277.8
20%
0.0150
3.000
138.9
100%
0.0100
2.000
92.6
120%
0.0100
2.000
92.6
5%
0.0150
3.000
138.9
20%
0.0075
2.000
92.6
100%
0.0050
1.000
46.3
120%
0.0050
1.000
46.3
5%
0.0075
1.000
46.3
20%
0.0035
0.500
23.1
100%
0.0020
0.167
7.7
120%
0.0020
0.167
7.7
5%
0.0040
0.500
23.1
20%
0.0020
0.333
15.4
100%
0.0010
0.167
7.7
120%
0.0010
0.167
7.7
System Accuracy
• Input signal accuracy affected mainly by signal transducers
• Input circuits and algorithms (analog and digital filtering, DFT window, signal
processing, data concentrators, multiplexers)
• Timing reference
– GPS today can provide accuracy that is less than 1 µs or 0.022° at 60 Hz
• Fix delay Tf ~ 75 µs
• Propagation delay Tp ~ 25 µs
• Data transmission delay Td for a typical PMU (12 phasors and 10 DI, data
frame 680 bits, header frame 200 bits and configuration frame 2.8 kbits)
– 110 µs on a 33.6 Kbps telephone line channel (worst case)
– Negligible for fiber optic cable
• The total delay Tf + Tp + Td ~ 210 µs (telephone line) and ~100 µs (fiber)
Standardization, Tests, and
Certification
- IEEE Std 1344-1995 (R2001)
- IEEE Standard C37.118-2005
- EIPP/PRTT activities
June 5 – 7, 2006
How to exchange PMU data?
• PMU configuration information
– Data format definition
– Static after setup
• Synchrophasor data
– Real-time data stream
• Reporting rate
– Format
• Fixed or floating point
• Polar or rectangular
IEEE Std 1344-1995 (R2001)
• IEEE Standard for Synchrophasors for Power System
– Approved December 1995 and reaffirmed 2001 (no change)
• Main achievement
– Defined a consistent and accurate time-tagging method
– Allowed the use of both synchronized and non-synchronized
sampling
– Not locked at the nominal frequency but follows the frequency
of the signal (steady-state)
– Defined angle convention independent of window size
– Required the correction of internal phase angle delays
IEEE Std 1344-1995 (R2001)
• Main achievement (cont’)
– Defined the data format of phasors being transmitted
• Configuration frame
• Header frame
• Phasor Information frame
IEEE Std-1344 Phasor Information Data Frame
Limitations of 1344-1995 (R2001)
• Defined angle convention only at Zero-crossing
– Phasor angle requirements set at 1 PPS mark but not inside
the 1 second window
• Limited to steady-state conditions
– The standard accepts different responses for non-steadystate conditions
• Data format not fully compatible to network
communications
– COMTRADE style aimed for serial communication links
• Limited implementation by manufacturers
PMU comparative test – May 2003
PMU comparative test – May 2003
Test Conclusions on Exiting PMUs May 2003
IEEE C37.118 – The new standard
• Approved December 2005
• Main improvement over IEEE Std 1344
– Defined an “Absolute Phasor” referenced to GPS-based and
nominal frequency phasors
– Defined a better time-tagging method
• Main improvement over IEEE Std 1344 (cont’)
– Introduced TVE (Total Vector Error) for quantifying phasor
measurement errors
( X r (n) − X r ) + ( X i (n) − X i )
2
TVE ≡
X r2 + X i2
2
r
r
X Measured − X Ideal
⇒
r
X Ideal
Total Vector Error
TVE
=
VIdeal - VMeasured
VIdeal
VError
• ±5 Hz frequency range
resulting in:
– Magnitude Errors
– Angle Errors
• 10% Total Harmonic Distortion
• 10% Interfering Signal
VIdeal
VMeasured
TVE from all Sources must be < 1%
• Main improvement over IEEE Std 1344 (cont’)
– Recommended PMU steady-state performance compliance
test requirement
Error Limits for Compliance Level 0 and 1
• Main improvement over IEEE Std-1394 (cont’)
– Defined data format compatible with other standards (e.g.
IEC 61850)
IEEE C37.118 Limitations
• Recommended but not required the dynamic
performance compliance
IEEE C37.118 Limitations (cont’)
• Lack of frequency measurement accuracy requirement
makes TVE not constant in a time window
In this example a frequency mismatch produces TVE = 0 only at
the center sample window but varies for any other time window
IEEE C37.118 Does Not
• Define a common phasor reference in a power system
• Provide detailed test setup and test procedures for
steady-state performance compliance test
• Address some practical application issues
– PMU field installation and commission
– PMU connection to Phasor Data Concentrators
EIPP Performance Requirements
Task Team (PRTT)
•
Requirements and protocols for data collection, communications,
and security through guidelines and standards
Eastern Interconnection Phase Angle Reference
•
•
Document: “Definition and Implementation of a System-Wide Phase Angle
Reference for Real-Time Visualization Applications” (approved).
Implementation of Virtual Bus Angle Reference at TVA SuperPDC
Phasor Requirements for State Estimation
•
•
Document approved by PRTT
In the EIPP acceptance process
Phase Inconsistency
•
•
Address phase inconsistency issue with corrective actions included.
Document posted
EIPP Performance Requirements
Task Team (PRTT)
PMU Installation/Commissioning/Maintenance Survey
•
•
Understand current practices and provide reference for others.
Document: “Survey on PMU Installation and Maintenance” (posted).
Installation costs for one PMU
Installation Time
PMU Acceptance Checklist for Connecting to SuperPDC
•
•
Facilitate connecting PMUs to SuperPDC (current critical path of EIPP)
Document developed
PRTT Top 3 Items
Guide for calibration standards and testing procedures (including
dynamic) to assure performance and interoperability
•
•
Standardize testing facility/signals/cases/criteria Î NERC Standard
Draft guide under review - Target complete date December 2006
Synchrophasor Accuracy Characterization
•
•
Characterize phasor accuracy in the instrumentation channel including
PTs/CTs, instrumentation, communication channels, and PMUs
Draft document under review - Target complete date December 2006
PMU Installation/Commissioning/Maintenance Guide
•
•
Start with survey results, provide guidelines for PMU
installation/commissioning/maintenance
Staged methods
•
Part I: PMU acceptance test
May 2006
•
Part II: PMU Installation procedures
December 2006
•
Part III: PMU maintenance procedures
May 2007
•
Part IV: PMU commissioning procedures
October 2007
Guide for Calibration Standards and
Testing Procedures
• Scope
– Performance and Interoperability of PMUs
– Covers static tests as described in IEEE C37.118
– Covers dynamic tests beyond C37.118
– System tests
• Purpose
– To provide clear guidelines for conformance tests and certification
•
•
•
•
Test equipment
Test requirement (steady-state and dynamic)
Test setup and test procedures
Data frame conformance verification
– Laboratory and Utility Environments
– Compatibility with PDCs and System Requirements
• To become a NERC standard
Status of Calibration System
• System performed frequency, amplitude, and phase tests
on PMUs
• Preliminary Calibrations Show that the System will meet
– Less than 0.01 % magnitude error
– Less than 0.2 µs time error
– Less than 0.013 % TVE
• Plans
– Program and Test PMU for:
• Harmonic Distortion
sensitivity
• Inter-harmonic Sensitivity
• Frequency Ramps
– Develop Additional Dynamic Tests
Document transducer errors
Summary
• IEEE C37.118 has provided a good foundation for
Synchrophasor applications
• There are still some pressing issues that C37.118 did
not address
• EIPP PRTT is currently working on these issues to fill in
the gap
• Results of EIPP PRTT activities are critical for the
successful applications of the Synchrophasors and
PMUs
Conclusions and Next Steps
•
Advances in sensing, communication, computing, visualization, and
algorithmic techniques for Wide Area Monitoring, Protection, and Control
Systems provide cost effective solutions to reduce costs, improve system
performance, and minimize risks
•
Need for WAMPAC application and deployment roadmap based on
“business case” analysis to support utilities, regulators, and vendors
•
Leverage benefits through integration of applications
•
Early adopters lead the industry – Need for wider deployment
•
Needs for education, training, and process and culture change
–
Ownership within a utility and how to share benefits among groups
•
System-wide implementation and common architecture
•
Uniform requirements and protocols for data collection, communications,
and security achieved through guidelines/standards
•
Sharing experience and best practices (e.g. EIPP)
25 de Febrero, 2011
Página 14
Anexo 4
Selección de Artículos Técnicos del CIGRE París 2010, incluyendo:
o
La Aplicación de Monitoreo de Área Extendida (WAMS) al Sistema de Transmisión
de Gran Bretaña para Facilitar la Integración a Gran Escala de Generación
Renovable – National Grid / Scottish Poweer Transmission Ltd. / Psymetrix Ltd.,UK
o
Desarrollo de un Estándar Chino en la Estación Principal de WAMS para Mejora
Adicional de la Capacidad de Monitoreo Dinámico en Tiempo Real – State Grid
Electric Power Research Institute / North China Power Engineering Co..Ltd. / Beijing
Sifang Automation Co.Ltd.-/ China Electric Power Research Institute, China
o
Evaluación de Desempeño del Sistema de Monitoreo de Área Extendida (WAMS)
Coreano bajo Condiciones Operativas de Campo de la Red Eléctrica de Corea –
KDN Co. Ltd. / LSIS Korea / Univ. KERI, Korea
http : //www.cigre.org
21, rue d’Artois, F-75008 PARIS
C2_112_2010
CIGRE 2010
The application of wide area monitoring to the GB transmission system to
facilitate large-scale integration of renewable generation
A M Carter1, M Perry1, C H Bayfield2, T Cumming2, R Folkes3, D H Wilson3
(National Grid plc1, ScottishPower Transmission Ltd2, Psymetrix Ltd3)
United Kingdom
Introduction
The three Transmission Owners (TOs) in GB have embarked on an ambitious programme of
network reinforcements to accommodate the UK Government renewable energy targets.
These reinforcements are cited in a report published by the Department of Energy & Climate
Change, Energy Network Strategy Group, [1]
Acknowledging the strong environmental lobby to curtail the development of new overhead
lines, the reinforcement proposals seek to maximise the use of the existing network using
series & shunt reactive compensation, voltage uprating, dynamic rating and Wide Area
Monitoring (WAM) etc. In addition, further infrastructure reinforcements will include the
deployment of embedded onshore and offshore HVDC links and Reactive Compensation.
The integration of embedded HVDC and Series Compensation on a heavily meshed
transmission system, with potentially 30GW of wind generation by 2020, presents new
technical challenges for the planning and operation of the GB transmission system. The
known potential for undesirable oscillatory behaviour when these technologies are deployed,
requires the detailed study of system dynamics covering a wide range of planning scenarios to
establish the optimum control system architecture.
The development of a WAM system (WAMS) is seen as a pre-requisite to the deployment of
the above technologies and this paper describes a road map to realise the benefits of this
technology in planning and operational time-scales, identifying short, medium and long term
benefits.
Overview Principle of Operation
The concept of using fast, synchronised measurements of power system variables was first
introduced in the 1980s [2], but has only recently emerged as industry-standard practice for
monitoring power systems following the release of the IEEE standard C37.118 in 2005 [3].
The components of a typical WAMS are illustrated in Figure 1. Data is collected, time
stamped and converted to phasor format at the substation, and then streamed through a widearea network to a central location. At the central location, it is processed and the information
and alarms are presented through client applications. Information is also passed from the
WAMS to third-party applications, in particular the Energy Management System (EMS),
which is the operator’s primary tool for observing and controlling the state of the system.
1
Figure 1: Typical Wide Area Monitoring Scheme
There are a number of key features of the technology that enable very significant advances in
observing and controlling grid behaviour.
Phasor representation: The measured values are represented as phasors. The acquisition
devices (Phasor Measurement Units, or PMUs) translate voltage and current waveforms from
signals sampled at several kHz to an accurately synchronised phasor representation of the
magnitude and angle of the fundamental 50Hz or 60Hz component. This leads to a very
concise representation of the measurement, as the most important characteristics of the
waveforms are captured in only two values – magnitude and angle, sampled at up to 50/60
times per second.
The synchronised phasor representation of the measurements is particularly useful, as this
form of information is used widely in power system analysis and state estimation. Previously,
estimation of the state of the system was done by deriving phasors from measurements of
active and reactive power; now they can be measured directly.
Accurate time stamping: Accurate time synchronisation is fundamental to wide-area
monitoring. PMUs use a GPS-synchronised time source to timestamp data to microsecond
accuracy. Data is streamed into a central location, time-aligned and then processed.
It is significant that the approach to alarming and presentation of power system disturbance is
based on a synchronised wide-area view. A major problem with the traditional operational
observability of disturbances is that events are detected locally and fed to the EMS as an event
list without strict time-alignment, priority or indication of dependencies. The WAMS
approach to alarming and presentation of disturbances is fundamentally different because the
whole system can be observed simultaneously.
Streaming data: WAMS data is streamed from the source, and can therefore be used for realtime applications. Before the emergence of this technology, fast data was only available from
disturbance recorders for post-event analysis, and only slow-scan data was available in realtime. The streaming capability enables users to observe the system in real-time with a level of
detail that includes system dynamics.
2
Dynamic response: Data is produced by phasor measurements at up to one sample per cycle.
This data rate captures the dynamics of the system, including many of the characteristic
phenomena that propagate through the system. This includes, for example, the characteristic
ringdown response to disturbances as well as frequency control, inter-area and local mode
oscillations. However, there are some issues that are not currently observed consistently using
phasor data from commercial PMUs, including sub-synchronous resonance, higher frequency
HVDC control interactions and harmonics. These issues are discussed further in a later
section.
PMU Installation and Network / Server requirements
ScottishPower (SP) has over 120 locations with System Monitoring equipment with Phasor
Measurement capability. At these sites, SP have installed Integrated Disturbance Monitors
(IDM ) and Local Storage Units (LSU) that are IP connected to a Substation Operational LAN
that is then connected to the Corporate Network through two back-to-back firewalls for
security. The LSU is configured to send IEEE C37.118 data streams and they are stored on a
central server at the Grid System Operations Control Centre. High stability crystals have been
installed in the IDM recorders to minimise phase angle drift between 1pps time references and
gives a more accurate measurement. A server has been installed that is capable of receiving
up to 400 voltage and current phasor data streams simultaneously and storage for up to 6
months. Archive and data back up facilities have been built in as well as support for the
application.
National Grid has fault recording and System Monitoring equipment installed throughout
England and Wales with many connected via a Wide Area Network or Corporate Telephone
Network to enable the data to be retrieved for post event analysis. A number of these existing
devices are being upgraded to provide PMU functionality with the data being streamed back
to the Electricity National Control Centre. There is potential for combining the PMU data
from Scotland and England and Wales to give the National Electricity Transmission System
Operator a real time overview of the PMU data across the whole of Great Britain.
Validation of PMU & Communication Architecture
For such a widely deployed installation of PMU hardware across the power network it is
important to test all aspects of the system for accuracy, stability and delay. Initial tests have
been conducted on the test bench to validate that the PMU is digitising and recording signals
correctly. This was done using a test set fitted with a GPS Synchroniser to enable phasors of
known amplitude and phase to be generated and injected into the IDM recorders.
The next test was to do the same tests at a sub-station and inject the same signals to prove that
the on site recorders performed the same as those on the test bench. Once this was
successfully completed the Ethernet network was checked to measure the time delays
(latency) and data drop out over the network to ensure no data was lost during the test.
To check the overhead line impedance measuring algorithm, voltage and current phasors were
injected into two calibrated and GPS synchronised PMU’s on the test bench. The phasors
were scaled to make the calculation of R, X and B simple. The phasor data was extracted
from the WAM system to calculate the R, X and B values from the known phasors and
3
compared to a range of typical parameters. Further analysis is proposed to address external
errors introduced by CT’s and VT’s.
Benefits of WAM processing
Oscillatory Stability
Power systems exhibit complex dynamic behaviour, largely due to the interaction between
spinning masses interconnected through magnetic linkages to a common electrical network.
The influence of active controllers such as governors and voltage regulators is critical to the
stability of the power system. In GB, there is a characteristic 0.5Hz inter-area natural
oscillation mode across the transmission boundary between Scotland and England. In the
1980s, this led to occurrences of sustained power oscillations of up to 1500MW pk-pk shown
in Figure 2. The issue was resolved through the installation and tuning of Power Systems
Stabilisers (PSS). Continuous monitoring of the oscillatory modes and damping was
introduced to provide early warning of any further system oscillation events should they
occur.
Power (MW)
800
Harker
Linmill
600
400
200
0
0
100
200
300
400
Time (seconds)
Figure 2 : Example of an unstable 0.5Hz inter area oscillation in GB in 1980 (two of the
four circuits are shown)
This approach to stability monitoring has since been adopted by several system operators
around the world. The approach is now implemented on a WAMS platform, rather than the
original custom-built hardware, providing integration with other stability and WAMS
applications, and also further dynamics analysis capabilities. The ability to identify the areas
in which generators are swinging in coherent phase and in opposing phase is important for
understanding the nature of the dynamic behaviour and defining actions to improve the
stability of the system. Figure 3 shows a typical WAM system display of oscillatory modes.
4
Figure 3 : System-wide view of amplitude and phase of a mode of oscillation
There is a concern that the major changes in the transmission infrastructure and loading
pattern in GB with large-scale integration of renewable energy sources will significantly
change the dynamic characteristics of the system. New dynamic phenomena are likely to
emerge as the inertia of the entire system changes and the power flow between the extremities
of the network is significantly increased. In view of these changes, it is considered prudent to
deploy more extensive coverage of the system to capture a more detailed baseline of dynamic
behaviour and monitor the impact of changes.
Network Modelling, State Estimation and Line Parameter Estimation
The contribution that phasor measurement can make to improving state estimation is well
known and documented [2,4]. The state estimator takes inputs of active and reactive power
measurements and derives the state of the network in terms of phasors. All of the EMS realtime processes, such as load-flow analysis, contingency analysis, voltage stability assessment,
dynamic security analysis and market tools run from the state estimation solution, so its
accuracy and stability is important.
This process requires complete observability and minimises errors by estimating the best fit of
redundant measurements, taking account of the expected accuracy of the measurements. By
contrast, phasor measurements obtain the state vector directly. In future, it is conceivable that
state estimation is derived entirely from phasor measurements, which would lead to a very
fast, robust and accurate solution. However, at present there is no system with sufficient
penetration of PMUs to achieve this, and EMS vendors now offer hybrid solutions that
integrate PMU measurements to achieve a more accurate and robust solution. We anticipate
that Scotland could be a leading example of high density of phasor measurements, using a
large installed base of PMU-capable disturbance recorders.
The accuracy of the state estimator solution depends on the network model, as well as the
SCADA and PMU measurements. The network model depends on line parameter values that
are defined from the length and geometry of the line, and may not be very accurately
5
determined. Line parameters are not static, and vary with weather conditions and loading. In
particular, the shunt susceptance and series resistance are weather related.
Energy Management System
WAMS / PMUs
Dynamics
Analysis
Line Parameter
Measurement
Security
Enhancement
Network Model
Phasor > EMS
Dynamic Security
Assessment
Downsample, time-align,
bad data filter
State
Estimator
Contingency
Analysis
SCADA/Phasor
Hybrid
SCADA / RTUs
Market
Applications
Etc...
Figure 4 : Inter-relation between WAMS, State Estimation and EMS Processes
The value of the -model line parameters can be identified using phasor measurements of
voltage and current at the sending and receiving ends of a transmission line, as follows, where
the V and I values are complex numbers:
Series resistance:
Series reactance:
Shunt susceptance:
A key aspect of the practical use of phasor measurements to identify transmission line
parameters is handling the errors. When the line is lightly loaded, the angle differences are
small and the relative error is large, so the confidence in the results is low. However, the
relative errors decrease at higher line loading. The value of identifying transmission line
parameters is greater at higher loading, as this is the condition in which an accurate
knowledge of the real capability of the line is useful. It is intended that the line parameter
measurements are associated with a confidence measure, and provided a sufficiently high
accuracy is obtained, the values can be provided to the network model within the EMS for
contingency analysis and real-time voltage stability assessment.
In some practical tests of transmission line parameter estimation, a significant imbalance has
been observed in the transmission parameters. The significance of the imbalance in line
impedance is of interest, and its effect on the analysis of the stability of the system is worth
investigating.
Islanding Detection and Re-synchronisation Aid
WAMS provides a very direct indication of the presence of electrical islands, and can also
show the stability of the island, and whether the islands can be reconnected. Islanding is
indicated by a difference in frequency across the network, and by freely rotating voltage
phasors. This direct approach contrasts with a conventional EMS approach depending on
6
correct topology measurements and subsequent analysis to determine the islanded state.
Furthermore, the PMU approach clearly shows out-of-step conditions, where breakers have
not opened.
Tailoring WAM for the future
Detection and Presentation of Higher Frequency Wide Area Phenomena
As discussed above, the phasor measurement technology available today provides data at a
rate up to 1 sample per cycle. In order to avoid aliasing and attenuation, the highest frequency
that can be observed with 50Hz sample rate is seldom more than 12Hz, and can be
significantly less depending on the PMU model and configuration.
There are some phenomena that can potentially affect a wide area of the network at
frequencies above the measurement range of current PMUs. This includes Sub-Synchronous
Resonance (SSR) and interaction between HVDC controllers. SSR is a resonance between the
natural frequency of a series capacitor in an inductive line, and the torsional modes of a
generator shaft [5]. If this occurs, it can typically be in the range of 10-30Hz, and therefore
outside of range of existing PMUs. Likewise, interaction between HVDC controllers can be at
a high frequency. It is also possible that a similar phenomenon to SSR may occur with wind
turbine gearboxes that have natural frequencies in the range of 10-30Hz. This issue could
interact in a series-compensated network in the same way as the torsional oscillations in the
classical SSR, or potentially interact with other controllers in the system.
These issues are relevant in the GB context. Series capacitance is being added to the ScotlandEngland transmission boundary to increase the AC interchange capacity. The ScotlandEngland interconnection is to be reinforced by subsea HVDC cables, off the east and west
coasts. Furthermore, much of the new wind generation will come from large offshore
windfarms and be connected through HVDC.
The infrastructure of monitoring and communications is well suited to addressing the issues
of higher frequency components. However, there is development required relating to the
measurement processes. Two approaches will be considered and compared in further
development work. Firstly, it may be possible to increase the sample rate of certain PMUs to
half-cycle measurements, and process these measurements centrally, as in the current
practices for WAMS applications. However, there are some inherent limitations to the
measurement bandwidth of PMUs. A second approach is to use the high-speed waveform
recordings that are acquired by the PMU and analyse these locally for high frequency
phenomena. The results of the local analysis could then be streamed to a central location
using the infrastructure and protocol that is standard practice for WAMS technology.
Identification of Sources of Damping Issues
When oscillation monitoring is deployed in a system, it is commonly found that there are
several modes of oscillation found that are not replicated accurately in the model. This
implies that there is
a) equipment connected to the grid that is not functioning normally, or
b) an aspect of the grid dynamic behaviour that is not captured in the analysis processes
to determine the constraints of the system relating to dynamic behaviour.
7
It is important to be able to identify the most significant contributing factors to issues that
emerge that are not clearly understood.
At time of writing this paper it is proposed to trial two processes that address the problem of
identifying the location or source of system oscillations [6,7,8]. The first uses only phasor
measurements to identify the positive or negative contribution to damping that arises from
within an area whose boundaries are fully monitored with PMUs. The second process uses a
statistical approach to identify the sensitivity of the derived dynamic behaviour to the
conditions in the grid observed in slower-scan SCADA recordings. Thus, in the first
approach, it is determined whether the region as a whole has a degrading effect on the mode,
and if so, the second approach is used within the region to identify specific contributions from
plant.
The capability to identify the main contributions to the sources of oscillations that will enable
operators, analysts and planners to identify emerging dynamics issues in the grid and manage
them before they emerge as a significant threat to system stability. Furthermore, it will
provide operators with immediate guidance on dispatch actions that can be taken, if required,
to stabilise the oscillations.
Future Control Needs
The changes in the GB generation profile will result in much more variable power flow than
historic experience. The 2020 scenario of 30GW of wind power capacity connected, with
approximately 10 GW in Scotland, 15 GW off the east coast of England and 5 GW to the
west, presents a challenge for conventional control approaches. Wind variability can produce
large rapid changes in the generation pattern and consequently the flows around the network.
Furthermore, it is necessary to ensure that the transmission system stability is maintained
following single- or credible multiple contingencies. The challenge will be to ensure that the
maximum amount of renewable energy can be accommodated whilst ensuring system
stability. Special Protection schemes allow the network to be utilised more fully as the action
required to maintain system stability following the critical contingency is automated.
Typically this involves disconnecting a large generating unit. The use of special protection
schemes to ensure transmission system stability is more challenging with a much greater
proportion of energy from intermittent sources as they are typically smaller in size and it is
less certain as to their power output at a given time.
It is envisaged that by 2020, north-to-south power flow in the corridor between Scotland and
England will be strongly dependent on the wind strength. As an example of the potential
network stability problem, one can consider the failure of two of the circuits in the
interconnection between Scotland and England. In the past, it was possible to set up a special
protection scheme with generator tripping to prevent overload or instability in the corridor,
but in future, it is much more difficult to design a conventional generation shedding scheme
that would reduce the power flow by a predictable amount when the level will depend on
where the wind is blowing.
It is therefore thought that the concept of wide-area protection schemes, in which there is
centralised identification of disturbances, and intelligent selection of a response would be of
value. The concept of WAMS can therefore be extended to cover this consequential
protection and control of the network (WAMPAC). WAMS can be used to identify the
8
location, nature (loss of load, generation or line) and extent of a disturbance, and then a
centralised control action can be determined and then enacted to restore the stability of the
system. While it may not be possible for WAMPAC to act quickly enough to counteract the
first event, it is feasible for a true wide-area protection scheme with centralised logic to act in
response to a first contingency and increase its resilience to subsequent events, ie in the one to
ten minute timeframe. Initially, WAMS is likely to give advice to Control Engineers on what
actions to take, but ultimately this could be automated once its advice has been proved to be
correct.
At present, where Special Protection schemes are not used, the system operator must
manually restore the system to an N-1/N-D secure state within a specified time period. The
load that can carried by overhead transmission lines are related to the sag of the line caused by
the current heating the conductor. Therefore lines often have different time related ratings,
with high currents being able to be carried for short periods of time. Using a more
sophisticated scheme would enable the time to respond to a critical contingency to be reduced
and therefore enable the higher shorter term ratings to be used.
A centralised security scheme could also greatly improve the likelihood of maintaining stable
islands if a system separation did occur. Islands can only be maintained if there is a
sufficiently close balance of generation and load in the island for frequency stability. A wide
area scheme can be used to identify islanded conditions very quickly and could interact with
the energy balancing system to balance the islands to within a frequency margin that local
controllers can control. Furthermore, it is thought that it may be possible to determine when
an islanding event is inevitable (or would provide the greatest likelihood of continued
operation). In this case, the wide-area scheme could be used to determine where the system
should be split.
WAM systems are a key enabler to the Smart Grid concept and future sustainable energy
systems. Fundamentally, the Smart Grid involves much greater capability to observe and
control the grid, involving many more grid customers. Intelligent use of storage (for example
using electric vehicles) is also envisaged. While at present, there are only limited
opportunities to control load, it is thought that this aspect must become much more important
with a large component of intermittent energy sources, but this tends to increase the
complexity of the system. Angle measurements in different regions of the grid provide
intuitive location-specific signal of stress across the regional boundaries in the grid, and could
be used for relatively fast control of load and storage.
Conclusion
It is clear that wide area monitoring is a critical component in the GB system for
accommodating very ambitious targets for connection of renewable energy. The operation of
the transmission system will change significantly as a result of the integration of new energy
sources and it is important to ensure that the stability and security of the grid is maintained as
intermittent generation increases. The development of a WAM system is important to monitor
the changes in network performance as series compensation and embedded HVDC are
deployed in the system.
The use of the existing base of PMU-capable disturbance recorders affords the GB system the
opportunity to achieve a high penetration of synchrophasor measurements relatively quickly.
This is useful for capturing the dynamic behaviour of the system, and for trialing advances in
9
WAMS technology and applications. Looking forward, it is expected that the use of WAMS
in the GB system will progress to automated control applications that will assist the system
operator in continuing to deliver high reliability in a much more complex operating
environment.
References:
[1] Our Electricity Transmission Network: A Vision for 2020, Electricity Networks Strategy
Group, March 2009
http://www.ensg.gov.uk/assets/ensg_transmission_pwg_full_report_final_issue_1.pdf
[2]
Phadke A, G,
Thorp J, S, “Synchronised Phasor Measurements and their
Application”, Summer 2008, ISBN, 978-0-387-76535-8
[3]
IEEE Standard C37.118 –2005: “IEEE Standard for Synchrophasors for Power
Systems.
[4]
Avila – Rosales R. et al: “Recent experience with a hybrid SCADA/PMU on-line state
estimator”, IEEE PES General Meeting, Calgary, Canada 2009
[5]
Machowski J, Bialek J, Bumby J.R, “Power System Dynamics, Stability and Control”,
nd
2 edition, Wiley 2008, ISBN 978-0-470-72558-0
[6]
Wilson D.H., Hay K., MacLaren R. F. B., Hawkins D. J., Dunn A., Middleton A. J.,
Carter A., Hung W.: “Control Centre Applications of Integrated WAMS-based Dynamics
Monitoring and Energy Management Systems”, Cigre Session, Paris 2008, C2-105
[7]
McNabb P.J., Bochkina N., Wilson D.H., Bialek J.: “Oscillation Source Location
using a Novel Data Mining Technique”, IEEE T&D Conference, New Orleans, April 2010
[8]
Wilson D. H., Hay K., McNabb P.J., Bialek J., Lubosny Z., Gustavsson N.,
Gudmansson R.: “Identifying sources of damping issues in the Icelandic power system”,
PSCC Glasgow, UK, 2008
10
C2_209_2010
CIGRE 2010
Development of a Chinese standard on WAMS main station for further
enhancement of real-time dynamics monitoring capability
Y J FANG1, D N ZHANG2, D YANG3, Y XU4, T S XU1
1
State Grid Electric Power Research Institute
2
North China Power Engineering (Beijing) Co., Ltd.
3
Beijing Sifang Automation Co., Ltd
4
China Electric Power Research Institute
China
SUMMARY
In spite of wide-scale deployment of phasor measurement units (PMUs) based wide-area measurement
systems (WAMSs) in China, a survey report made by the State Grid Dispatching and Communication
Center pointed out that WAMS is still in a process of development and exploration. Its application
functions are still not yet systematic and mature. Its level of operational management and practical
utilization is relatively low. Its system maintenance and on-duty monitoring are far from wellestablished compared with that of SCADA. Its processing speed upon data abnormality and degree of
attention to data quality cannot yet meet the requirements of real-time operation. Its utilization of
functions and data is mainly on post-mortem analysis. There is still a gap between its accuracy of lowfrequency oscillation early warning together with the sufficiency of the relevant information and
professional requirements of power system operation. There is still a need for improving its reliability
and efficiency of data flow in the whole process of acquisition, communication, processing and
archiving. In order to further enhance the real-time dynamics monitoring capability through regulatory
compliance, a Chinese standard on WAMS main station “Specification for Main Stations of Real-time
Dynamics Monitoring Systems for Power Systems” is under development. The standard specifies that
the main station is of distributed structure. Different application functions can be distributed
accordingly to different computer nodes with key applications having a redundant hardware
configuration. All computers in the system should be interconnected directly through a redundantly
configured network. Application functions should include monitoring and analysis of low-frequency
oscillations, power system dynamics identification, system model and parameter verification,
assessment of generator primary frequency regulation etc. The standard specifies performance indexes
of WAMS main station regarding main station load ratio, delay and accuracy of data acquisition, MMI
response time, and accuracy of on-line low-frequency oscillation detection. To improve WAMS’s
level of achievement in practical applications, other efforts are also underway.
KEYWORDS
WAMS - Main Station - Standard - Real-time Dynamics Monitoring - Application Functions Performance Index
fangyongjie@sgepri.com
INTRODUCTION
In China optimal utilization of resources has advanced the ever-increasing scale of power system
interconnection and at the same time increases the degree of difficulty in keeping secure and stable
operation of such a large power system. Since 1995, power companies have undertaken a plan of
actions aimed to improve the operational security through the enhancement of monitoring facilities
and now there is wide-scale deployment of phasor measurement units (PMUs) based wide-area
measurement systems (WAMSs) [1].
In 1995 the first set of WAMS was installed in China Southern Power Grid and by the end of 2008
there were 29 WAMS main stations in China. Currently the maximum number of PMUs connecting to
one single main station is around 300 and the highest data transmission rate is 100 Hz with the
transmission protocol conforming to a Chinese standard based on the IEEE Synchrophasor Standard
1344-1995 (R2001). Application functions of a main station include dynamic information monitoring,
disturbance identification, fault analysis, low frequency analysis, and market auxiliary service
monitoring, model verification and so on. Dynamic databases are used for high-speed storage and
retrieval of massive data. In particular fault recording functions are integrated into PMUs and recorded
fault transient data can be retrieved by the main station. Data from a main station can be saved as BPA
format files for system oscillation analysis using small disturbance stability analysis software. In
addition, load model verification has been conducted using the PMU records of large disturbance field
tests in Northeast Grid in 2004 and 2005. Other PMU applications under development include peak
wind power regulation, oscillation damping control and so on.
This paper first gives an overview of the operating experience, proven benefits and currently
encountered problems of WAMS according to a survey report made by the State Grid Dispatching and
Communication Center. As one of the planned enhancement measures and with already considerable
expertise of WAMS engineering applications and its related subjects on a nation-wide scale, a Chinese
standard on WAMS main station “Specification for Main Stations of Real-time Dynamics Monitoring
Systems for Power Systems” is under development, as a follow-up of the standard “Technical
Specification of Power System Real-time Dynamics Monitoring System” issued in 2005. This paper
then describes the main contents of this new standard in the order of system architecture, application
functions and performance requirements. The State Grid Dispatching and Communication Center also
have proposed other efforts which are devoted to address the application issues such as data quality,
system operation stability, maintenance and management support, interconnection of WAMSs, unified
MMI with EMS and so on. These suggested actions are then presented.
AN OVERVIEW OF THE OPERATING EXPERIENCE
At present, nearly 90% of the 500kV substations have PMUs installed and a selective installation in
220kV substations organised by each provincial power company is also underway. Network integrity
of PMU data transmission has been greatly improved and construction of WAMS main stations has
been popular among power control centres at and above provincial levels. WAMS has been playing an
indispensable part in power engineering test, grid operation monitoring and disturbance analysis, and
is gradually becoming an invaluable tool for verifying the accuracy of power system stability
calculation results. The WAMS main station at the North China Grid Power Dispatching Center is
shown in Figure 1.
1
Real-Time Data Server
Historical Data Server
Advanced
Network Security
+ Data Archive Array
Application Server
Isolation Device
Maintenance
Communication Front-End
Workstation
Server
Communication Network
WEB Server
Dispatch
Monitoring
Off-line
Analysis
Workstation
Workstation
Internet Gateway
Communication Network
PMUs at
Substations
WAMS Main Station at
State Grid Dispatching and
Communication Center
WAMS Main Station at
Shandong Electric Power
Dispatching Center
Figure 1 The WAMS main station at the North China Grid Power Dispatching Center
In the view of system operating effects, functions that can be put into practical operation include
historical data storage and off-line analysis, low frequency oscillation statistics and analysis, tie-line
power dynamics monitoring, primary frequency regulation monitoring and assessment, and fault or
disturbance identification. There is strong demand for low frequency oscillation detection and
alarming but there is still deficiency in the active alarming capability and alarming information
integrity. In most cases the application is limited to post mortem analysis of known oscillation events
and expert experience is frequently relied upon in order to pinpoint coherency groups and oscillation
causes. Verification of network model and parameters still stay at a level of manually conducted
qualitative comparison. Identification of load model and parameters is still under exploration.
From the perspective of R&D direction, further application functions will include integrated low
frequency oscillation alarming based on the combination of on-line small disturbance analysis and
WAMS dynamic measurement data analysis, tile-line low frequency power osccillation suppression
based on the combination of WAMS data and AGC control, and analysis and control of a wider
diversity of stability problems including transient, voltage, thermal and frequency stability [2].
Operating experience has demonstrated that WAMS provides the most direct and effective technology
of monitoring, analyzing and understanding low frequency oscillations. In November 2008, low
frequency oscillation of a relatively large scale occurred in the Central China Grid and WAMSs in
both the State Grid Dispatching and Communication Center and the Central China Grid Dispatching
and Communication Center captured system dynamics in time. A post-disturbance analysis of systemwide PMU data showed that oscillation mode was clearly determined by frequency recordings and
oscillation center was accurately located by line power recordings, both being consistent with small
disturbance analysis results using on-line operational data. The final verdict of the event was reached
in less than one day thanks to this new technology, as contrasted with the one-month lasting
investigation process of the oscillation event in October 2005.
Figure 2 shows an oscillation of Hebei power system on 3 June 2008 recorded by the main station at
the Hebei Grid Power Dispatching Center. The low frequency oscillation was between Hebei southern
power grid and Beijing-Tianjin-Tangshan power grid with the dominant oscillation frequency being
0.24 Hz. The oscillation was featured by a fast decay as a result of relatively strong grid structure.
2
(a) Active power recording of Fangshan-Baoding tie line I
(b) Amplitude-Frequency characteristics from short-term Fourier analysis
(c) Prony algorithm based curve-fitting (red) of the original recording (green)
Figure 2 Oscillation recording and analysis result of Hebei Power Company WAMS
Around 15 WAMS projects are featured by a unified EMS/WAMS platform and functions of 4 main
stations have been extended beyond the conventional WAMS functions to cover on-line DSA related
functions from enhancement of state estimation by PMU data to on-line stability analysis, decision
support and even adaptive transient stability control [3].
In spite of the rapid development of PMU based WAMS in recent years, their practical application is
still in a stage of cultivation. Operational management has not yet been regulated, a complete
framework of technical specifications has not been set up and there still exist some problems in system
design, construction, management and utilisation, particularly in the following aspects.
 Data quality needs to be improved.
As WAMS data are tagged with time stamps, have a high acquisition frequency and provide phasor
measurements, high requirements are raised for data acquisition, communication and processing and
there are more factors influencing data quality in WAMS than in traditional SCADA. Operating
experience of the commissioned systems shows that data quality problems are often related to causes
such as time synchronization abnormality as a result of loss of GPS signal or time synchronization
module faults, improper frequency deviation compensation in phasor calculation algorithms, loss of
data points due to high speed or abnormality of communication, PMU device failure, software
abnormality of main station data processing programs, and TV/TA circuit problems.
 System robustness needs to be improved.
3
With the increase of the quantity of interconnected PMU substations, capacity of data communication
and processing increases, network performance decreases, and processing demand of software
programs increases, which can probably result in heavy load for WAMS front end communication
processors, application servers and data storage servers, increasing the risk of system’s inability of
normal operation. In addition, some systems are of single server and single network structure, leading
to lower system reliability.
 Technical support and management in operational maintenence need to be strengthened.
Some WAMSs lack functions of self-monitoring and automatic alarming for communication
interruption and process abnormality. Some WAMSs have not been maintained and managed in a way
as required by a real-time system.
 Application functions need to be developed.
Currently the commissioned functions mainly focus on analysis and early warning of low frequency
oscillations, monitoring and assessment of primary frequency regulation, and identification and
analysis of faults or disturbances. Post mortem analysis results are relatively reliable but on-line
functions need to be improved. The pressure of developing and commissioning new application
functions increases as the perfection of data quality and quantity due to scale enlargement of PMU
locations, accuracy and frequency improvement of data acquisition, as well as network integrity
enhancement.
 WAMS interconnection needs to be implemented.
In analysis and early warning of low frequency oscillations, PMU data of a wide coverage of power
systems are neccessary but WAMSs of various power companies retrieve data mainly from PMU
substations and there is a lack of data exchange and sharing through WAMS interconnection.
 MMI needs to be integrated with that of EMS.
In some WAMSs, MMI is not integrated with that of EMS, resulting in operators’ inconvinience or
even inability of using the system. Therefore coordinated analysis using WAMS and EMS data cannot
be conducted, which is not in favour of the development of real-time application functions.
 Technical specifications of field test of PMU devices need to be developed.
Dedicated technical specifications of network access test for PMU devices are not available and thus
their performance under various operating conditions cannot be guaranteed. Also a standard for field
test is required.
MAIN CONTENTS OF THE NEW STANDARD
The key points of the standard are summarized as follows, including system design requirements,
system functions and performance requirements.
System Design Requirements
 System architecture
The main station system of a Real-Time Dynamics Monitoring System is preferably of distributed
structure consisting of hardware such as data acquisition computer servers, real-time application
servers, historical data storage servers, graphics monitoring workstations, etc., and the corresponding
support software and application software.
 Hardware and software
4
A dual-network structure should be used and a dual-server or multi-server configuration is preferable
for hot standby capability so as to meet the requirements of reliability, maintainability, and
extendibility. System software, database software and application software should be integrated, which
should be characterized by openness, high reliability, high security and maturity.
 Data communication
It is preferable to use electric power dispatching data network. Point-to-point connection with backup
channels is preferably used for communication between the main station and substations and
communication protocol should follow Q/GDW 131－2006, Technical Specification of Power System
Real-time Dynamics Monitoring System. Network communication is preferably used for inter-utility
or inter-system data exchange. Secondary Power System Security Protection Requirements should be
followed for network security defense.
System Functions
 Data acquisition and monitoring
Data acquisition should be able to collect time-stamped data the types of which include phasors,
analogue signals, status (digital) signals, status change triggering message, measurement out-of-limit
triggering message, dynamic data files and transient data files.
It is preferable to receive and pre-process at a rate of 25Hz or 50Hz the real-time data message
transmitted from substations and then form a snap shot data of system dynamics. The system should
be capable of detect, count and retransmit corrupted data due to communication failure. When more
than one front-end communication servers are used, processing load should be balanced.
Data processing functions should include time stamp alignment, data plausibility check, corrupted data
management, event classification and frequently used calculation support.
For data storage and management, it should be able to set the storage object and cycle, the storage
capacity should allow for at least 14 days of historical data, and the stored data accuracy should be
consistent with that of the data transmitted from substations. Data screening and compression
functions are preferable. When the power system is subject to disturbances, dynamic data record
length should cover the whole disturbance process, record density should not be lower than data
transmission density of substations, and data storage should be long term.
Dynamics monitoring should include display of system frequency, voltage, current, power and phasor
by tables, curves, meters etc. and their distribution and trend by graphics, and alarming related to
communication channel abnormality, measurement threshold-crossing etc. by colour changing,
blinking, aural warning and so on.
 Analysis Functions
On-line monitoring, analysis and real-time alarming of low frequency oscillations should be
implemented. The analysis should be based on effective time domain signal extraction algorithms and
alarming information should include the frequency and amplitude of each relative measurement, and
the dominant oscillation mode and the frequency and amplitude of each mode in case of multi-mode
oscillations. For the most severe oscillation mode, a unified analysis combining mode parameters and
oscillation information from several substations should be conducted to give oscillation phase
relationship among substations, relevant substations and exchange interface of oscillating power.
Off-line analysis of low frequency oscillations should be implemented for data from historical data
storage. Prony’s method should be available and analysis results should include the amplitude,
5
frequency, phase and damping ratio of each oscillation mode. The function for comparing analysis
results with original data should also be available.
The main station should be able to identify power system disturbances such as unscheduled outage of
generators and line tripping etc. A search tool should be provided for searching disturbance events
according to time of occurrence, event type and faulted equipment etc. Off-line disturbance analysis
function should be available and a curve drawing tool should be provided for playback of the
disturbance process recorded in dynamics data file.
For assessment of primary frequency regulation of generators, the main station should take real-time
measurements of generator frequency and active power, and calculate and compile statistics of
accuracy rate and energy contribution of the generator’s primary regulation operation.
It is preferable to have model and parameter verification function for power system components such
as transmission lines, transformers and loads etc.
Performance requirements
 System response time
The time for data acquisition in substations, data transmission to the main station and data display
should be no more than 3 seconds. Response time of 90% of the MMI should be within 3 seconds with
that of other MMI being no more than 5 seconds. Data refresh cycle of the MMI should be adjustable
between 1s-10s. Automatic switch-over time between main and backup computers should be no more
than 30 seconds.
 Load rate
Under normal operating conditions of power systems, the average load rate within 5 minutes of a
computer server CPU or a MMI workstation CPU should be no more than 30%, and that of the main
station LAN should be no more than 15%. Under power system disturbances, the average load rate
within 10 seconds of a computer server CPU or a MMI workstation CPU should be no more than 70%,
and that of the main station LAN should be no more than 20%.
 Data processing
Error between historical data through query and local storage data in substations should be less than
0.001o for phase angles and 0.5% for other measurements. The minimum capacity of data storage is 14
days. The time for query of each measurement’s one-hour data should be no more than 5 seconds.
 On-line low frequency oscillation monitoring
The calculation error of oscillation frequency between 0.1～1Hz should be no more than 0.02Hz and
that between 1～2.5Hz should be no more than 0.05Hz. The calculation error of phase relationship of
active power oscillations of participating generators should be no more than 10o. The success rate of
event capture should be no less than 99.8% and data should be stored for at least one year.
 System disturbance identification
The correctness rate of identification should be no less than 95% and the identification time should be
no more than 5 seconds. Data should be stored for at least one year.
 Assessment of primary frequency regulation of generators
6
The correctness rate of assessment should be no less than 95% and data of one time analysis results
should be stored for at least one year.
OTHER APPLICATION ISSUES
In addition to the development of a standard of WAMS main station in order to improve its level of
achievement in practical applications, other efforts proposed by the State Grid Dispatching and
Communication Center include: (1) the development of a grid connection code and field test code of
PMU device so as to improve both the quality of network access products and the standard of
engineering commissioning; (2) improvement of the self-monitoring function of WAMS, clarification
of maintenance responsibility and enhancement of an operational management and assessment
mechanism according to requirements of the real-time system; (3) research on coordinated application
of AGC control and WAMS in order to gradually implement wide-area damping of low frequency
oscillations through high-quality AGC control; (4) development of an integrated scheme of substation
measurement, fault recording and PMU so as to lower the cost of data acquisition and increase data
utilization efficiency at the substation level.
CONCLUSIONS
Operating experience has demonstrated that WAMS can greatly enhance the ability to obtain a realtime view of the system state over a wide area [4]. Practical applications are still relatively
undeveloped but great progress is being made. Each link of the chain needs to be further strengthened
and coordination is the key to long-term success of WAMS’s role in providing greater power system
reliability. Development of the standard of WAMS main station is such an effort for regulating the
construction and utilization of WAMS at provincial power company level and facilitating the interutility integration at nation-wide level. Together with the development of measurement and
communication technology to support application requirements, it is believed that the initial promises
of real-time dynamics monitoring could be fulfilled and will continue to show its potential of
contributing to the solution of the grid-wide problems. With the emerging applications of phasor
measurements in monitoring and control, WAMS will play a more and more important role in the
development of UHV interconnected smart grid in China.
BIBLIOGRAPHY
[1] A.G. Phadke, Hecotr Volskis, Rui Menezes de Morraes, etc. “The Wide World of Wide-Area
Measurement” (IEEE Power and Energy, Vol. 6, No.5, September/October 2008, pp 52-65).
[2] Y. Xue. “Some viewpoints and experiences on WAMS and WACS” (IEEE-PES 2008 General
Meeting Pittsburgh, PA,USA, 2008).
[3] Fei Shengying, Xue Yusheng, Ma Sulong etc. “Application of Electric Power Alarming and
Coordinated Control System in Jiangsu Power Grid” (CIGRE Symposium on Operation and
Development of Power Systems in the New Context, Guilin, China, Oct 28-30 2009).
[4] CIGRE Task Force WG C4.601. “Wide Area Monitoring and Control for Transmission
Capability Enhancement” (Brochure 330, August 2007).
7
C2_103_2010
CIGRE 2010
Performance evaluation for K-WAMS (Korean wide area monitoring system)
under field operating condition of Korea power grid
S.T. KIM* J.Y. KIM
KDN Co.,Ltd
S.H. JANG
LSIS
S.W.HAN B.LEE
Korea Univ.
KOREA
Y.H. MOON T.H. KIM
KERI
SUMMARY
The spread of Synchro-Phasor Unit such as PMU(Phasor Measurement Unit) of each country
in power grid represents an evolutionary change in power system measurement, monitoring,
control and protection, as phasor measurements hold the promise of providing a fast dynamic
picture of power grid status. In particular, synchro-phasor becomes the heart of smart gird
transmission from Smart Grid beginning to make its appearance in these days. Recently
deployed ‘Korea Wide Area Monitoring System (K-WAMS) β version’ in Korea power grid
with analyzing and assessing the local and wide area power system condition is introduced
and the evaluation of this system is dealt with a various aspects in this paper. The Korean
Wide Area Monitoring System (K-WAMS) in the first phase of system proving is composed of
eight i-PIU’s(Intelligent Power system Information Unit) such like PMU, and central system
i-PIS(Intelligent Power system Information System). As in the case of the stressed modern
power systems with marginal stability, metropolitan voltage instability and small disturbance
stability problem are the most concerned in the Korea power grid, and K-WAMS treats those
problems. In the metropolitan voltage instability view point, i-PIU’s are installed at three
heavily loaded points in generation areas of six routs which, three 345kV substations(Asan,
Chungyang, SinJaechun), and at the important point of SPS(Special Protection Scheme) view
point, one 345kV substation(DongSeoul) with leased line as a communication network for
just suitable data transferring. Moreover to monitor the characteristics of HVDC and wind
farm, i-PIUs were installed in Jeju-island. In this paper, the performance evaluation of
deployed K-WAMS β version in Korea power grid is introduced and focuses primarily on
the application using i-PIU measurements for grid dynamic analysis.
KEYWORDS
Synchro-Phasor - WAMS - Voltage Stability - Small Signal Stability - Real Time Stability Assessment
- PMU
jesteka@kdn.com, kimjy@kdn.com, suhjang@lsis.biz
INTRODUCTION
The development of Wide Area Measurement and Monitoring System(WAMS) technology,
combined with Phasor Measurement Unit(PMU) devices, is offering new, as valuable solutions for
power grid analysis, monitoring and assessment. After the initial applications, limited to off-line
studies, essentially for modeling and event reconstruction purposes, synchronized phasor
measurements have become a reality in the EMS room of utilities worldwide. Operators can track
system dynamics, in real time and with a degree of accuracy and detail that was not possible with
conventional SCADA/EMS. This allows a deeper and more straightforward understanding of system
conditions, and a consistent support in deciding and performing control actions and maneuvering.
During the past two decades, tremendous efforts have been made by power grid engineers as well as
researcher to improve power grid stability. Real time wide area monitoring, protection and control
systems based on synchronized phasor measurement has been recognized as a better technique that
can provide real time information on the dynamic behavior for any power grid, and could lead to
efficient solutions in handling the cascaded outage through coordinated and optimized stabilizing
actions. As of today, various approached of wide area monitoring, protection, and control system has
been implemented by power utility worldwide to monitor and maintain their system reliability. This
paper introduces the research and development project to build a reliable and accurate WAMS(KWAMS, Korea Wide Area Monitoring System) with wide area voltage instability and power system
oscillations monitoring application. K-WAMS is currently under-going the system trial and evaluation
phase on KEPCO power grid.
KOREA POWER GRID
The Korean power grid can be characterized as follows : First, the metropolitan region which has more
than 40% of total load demand and only 20% of total generation capability inherently requires a large
volume of power transfer from other regions. And most generating units with low generation cost are
in non-metropolitan areas. With the aim of economic operation, generators in the non-metropolitan
region mainly take charge of the base load and then generators in the metropolitan region is operated
to meet the demand increase in peak time. This also makes power transfer increase toward the
metropolitan region. Therefore, these interface lines, six routs, from the non-metropolitan regions to
the metropolitan region are heavily loaded. Moreover, transfer limit needs to be constrained by voltage
stability limit to operate the system in a secured manner considering severe outages. However, the
trend of heavy power transfer will continue and become more and more profound because the
construction of new facilities is difficult due to environment and the strong public opposition in the
metropolitan region. Secondly, in power systems that are composed of generators and other various
machines, the machines are operated in synchronization at a constant frequency. Generators in
synchronization supply rated power to load. In power systems, generators and loads are connected in
parallel through a network. If one generator is deviated at the rated speed, power change that occurs
from this is delivered to the other generator. At the same time, rotor speed of each generator changes
and other controllers such as turbine and exciter take appropriate control actions so that the generator
can return to the rated speed. Most generators have the damper winding to damp electric oscillation.
Also, a governor system for an automatic control of turbine is included to maintain constant
generator speed, and so is the excitation system to retain a fixed level of generator terminal voltage.
These controllers are indispensable not only to provide quality power but also to supply power in a
stable manner. However, due to inadequate controller setting or network condition, low frequency
oscillation may take place. Especially when a fault occurs, the high-speed excitation system to
prevents any damage to synchronizing torque and to improve transient stability tends to weaken
damping characteristics of low frequency oscillation that occur in power systems. Korean Power grid
has some low frequency oscillations (0.7Hz ~1.0Hz), and this low frequency oscillation has become a
serious problem which has an influence to limit the capability of power transfer in transmission
network and causes wide area blackout. The map of 765kV & 345kV transmission network and major
1
generating plants in the Korea power system and the problems of Korean Power Grid are shown in
Figure 1 and Figure 2.
Fig 1 Korea Power Grid
Fig 2 Characteristics of Power Grid
K-WAMS ARCHITECTURE
A. K-WAMS Platform Architecture
The Architecture of the K-WAMS platform at the control centre, i.e. of the central server of the KWAMS, consists of several processing tasks carried out in parallel on the server computer. Each task
is in real time mode, to assure high performances with the shortest possible execution times. The
whole system is designed to accomplish operations with different time cycles. To prevent a dangerous
time drift on K-WAMS operation, it is required that all processing tasks strictly comply with the
planned execution times. Additionally it is required that the various tasks be accurately synchronized
with each other, in order to avoid errors that could compromise the operation of the K-WAMS server.
The basic activity carried out by the K-WAMS platform is the acquisition and storage into shared
memory of the data package sent by the i-PIUs such as PMU to the control centre.
The K-WAMS(Korea Wide Area Measurement and Monitoring System) is the first phase of the total
planned R&D works on WAMPAC(Wide Area Monitoring, Protection, And Control). The basic
architecture of developed system comprises of the following hardware components:





i-PIU(Intelligent Power system Information Unit)
CSU & FEP Communication links
i-PIS(Intelligent Power system Information System)
i-PIE(Intelligent Power system Information Evaluator)
HCI(Human-Computer Interface)
Figure 3 shows the basic architecture of K-WAMS. Basically, i-PIU receive the GPS time signal at
each substation. They are installed at CT, PT points to measure the positive sequence data of voltage
and current via field data. Operator of substation can monitor i-PIU condition and data using MMS
communication line of IEC61850. The measured data are transferred at the rate of 60[sample/sec] via
T1 KEPCO communication line to Centre System based on IEEE C37.118 Synchro-Phasor Standard
with GPS time tag. Figure 4 shows the i-PIU made by LSIS with IEEE C37.118 and IEC61850
standard such like PMU.
2
Fig 3 System Architecture of K-WAMS
Fig 4 i-PIU (PMU)
Figure 5 shows the single type and multi type CSU which treat the communication network
integration. The communication way of K-WAMS is the B8ZS which is KEPCO communication
coding method. Single type CSU plays a main role to exchange communication type.
Fig 5 Single and Multi Type CSU
FEP(Front-End Processor) which interpret the IEEE C37.118 protocol and concentrate the real time
field data such like PDC is shown in Figure 6.
Fig 6 Front-End Processor
3
B. Monitoring Functions
The K-WAMS β version has dedicated displays at the NCC of KEPCO. The monitoring functions
visualised on-line to the operators include plots and charts of quantities directly provided by i-PIUs
such as voltage magnitude and phase angle, phase angle displacement between nodes, system
frequency, active and reactive power. Other information are also available, consisting of stability
indices and quantities obtained by processing the i-PIU data. In particular, with reference to Figure 6,
the following features can be highlighted:
 Indicators of phase angle differences between specific lines, customisable by the user
 Phase angle of nodes
 Voltage and frequency value in the map board
 Trend data monitoring of PQVF by time axis
 Arrow indicating the direction of active/reactive power flow between specific nodes, and the
angle difference between them
In particular, the following monitoring functions are currently implemented:
 Voltage instability indicator(under testing)
 Oscillation analysis(under testing)
 Event detection function(under testing)
 Z-Locus of distance relay(in operation)
Fig 7 Display of the K-WAMS HCI
In normal operation, a useful support to
operators is provided by the time plots of
quantities such as angles(in particular, angle
difference between nodes) and active/reactive
power flows, as elementary indicators of system
stress.
FIELD TRIAL SETUP AND TESTING
Fig 8 T/L Monitoring Viewer
As for monitoring the power system grid voltage
instability and power oscillation problem, the
KEPCO Dispatcher Center has selected four
345kV main substations:
4
Six i-PIU were installed in Korea Power Grid. The first
target of developed system is the voltage instability of
metropolitan area. The three i-PIUs are placed at
important substation of generator area in six routs to allow
observation of the Korean power grid transmission system
under any operational conditions. The second target is
wide/local area low frequency oscillation monitoring with
Asan and Sinjaechun data(345kV). The third target is
SPS(Special Protection Scheme) monitoring which is
installed at Dong-Seoul Substation. Figure 9 depicts the
position of i-PIU. The measured data are transferred via
dedicated communication channel(T1) of KEPCO, CSU
and FEP, to a K-WAMS. Based on the trial, it’s found that
K-WAMS is communicating with all i-PIU through
KEPCO communication network. Figure 10 shows phase
angle differences in each substation. i-PIU which installed
at DongSeoul, Asan, Chungyang, and Sinjaechun is
Fig 9 Placement of i-PIU
giving the good angle difference when compared with
power system studies result using PSS/e. The angle difference measured between Asan-DongSeoul is
16.01° while the PSS/e studies result is 15°.
Fig 10 Phase angle differences
A. Casee 1
Figure 11 shows the area of generators drop at Busan area in Jul.20th.2009. When about 2,000 MW
combined cycle generator drop occur, up to 59.7[Hz] frequency
drop during about eight minutes. Figure 12 shows frequency
trend data.
Fig 11 Busan C/C Drop
Fig 12 Freq. of Case1
5
B. Case 2
Figure 14 shows the area of generators drop at Boryung area in Jul.30th.2009. When about 50 MW
steam turbine generator drop occur, up to 59.85[Hz] frequency
drop during about one minutes. Figure 15 shows frequency trend
data
Fig 13 Boryong S/T Drop
Fig 14 Freq. of Case2
C. Case 3
When the generator drop occurs, low frequency oscillation is
detected in Wide area mode and local area mode. Figure 15
shows local mode and wide area mode on generator drop. Local
mode is detected at Asan substation near by Boryung generator
group, and the mode of the other i-PIU data is wide area mode.
Asan has 0.6 Hz of low frequency oscillation and the others
have 0.9 Hz of low frequency oscillation. As the trend of low
frequency oscillation is watched, Korea Power Grid depicts the
good characteristics of damping.
D. Case 4
Figure 16 shows voltage rising at DongSeoul bus on shunt Fig 15 Local and Wide Area Mode
reactor open at Sungdong Substation. Figure 17 shows the reactive power change on shunt reactor
open of related substation
Fig 16 Shunt Reactor Open
Fig 17 Voltage and Reactive Power
6
E. Case 5
When the winter comes, the winter load increases more and more currently. At the middle of
December in 2009, after one of the nuclear power plant, YoungGwang #5, #6, in the west-southern
frequency data and figure 21 shows the voltage recovery. After YG N/P #5, #6 were dropped, the
voltage was recovered as all P/P in all over the country immediately operated.
Fig 18 Freq Drop and Voltage Recovery
CONCLUSION
The paper presented the architecture and main functionalities of the K-WAMS installed by KEPCO
power grid. The K-WAMS platform, available to operators in the control room, already provides a
valuable support to operation. Real Time plots and chart of both system quantities such as phase angle
differences, PQVF time based trend plots and the output of monitoring functions such as wide area
voltage instability index and low frequency oscillation detector, allow operators to better track system
stress and dynamic phenomena, and evaluate manoeuvre viability. The K-WAMS is currently under
further development, with the implementation of new real time monitoring functions and the
improvement of existing ones according to the feedback coming from field experience.
BIBLIOGRAPHY
[1] Yan Dengjun, et al., “Real time Power Angle Measurement of a Synchronous Generator
Based on GPS Clock Signal and Tachometer”, Automation of Electric Power Systems,
2002, 26(8),38-40. In China
[2] I.Kamwa, “PMU-based Vulnerability Assessment Using Wide Area Serverity Indices
and Tracking Modal Analysis” PSCE2006, IEEE
[3] Yan Dengjun, “Wide Area Protection and Control System with WAMS Based”, IEEE
PWRS SEP, 2006
[4] A.B.Leirbukt, “Wide Area Monitoring Experiences in Norway”, PSCE2006, IEEE
[5] Joo Cheon Bae, et al., “The Introduction of Special Protection System(SPS) for
Increasing Interface Flow Limits in the Korean Power System” CIGRE 2006 C-210
[6] Hongjun Li, “Implement of On-Line Transient Stability Control Pre-decision in Wide
Are Measurement System in Jiangsu Power Network”, IEEE PWRS APR. 2005
[7] Mats Larsson, Joachim Bertsch, “Monitoring and Operation of Transmission Corridors,”
Power Tech Conference Proceedings, 2003 IEEE Bologna, 23-26 June 2003
[8] Mats Larsson, Christian Rehtanz, and Joachim Bertsch, “Real Time Voltage Stability
Assessment of Transmission Corridors”, 2003 IFAC
[9] Substation Automation Handbook, ABB
[10] IEEE C37.118 Standard.
7

CDEC-SING Recomendaciones de KEMA relacionadas con la

Transcription

Similar documents

Diptico de la Comarca de Guadix

St. Therese

NEW Build at Guardamar del Segura Los Estaños Beach

used equipment - Russo Power Equipment

Useful advises to help you protect your Child From domestic

Life changes, the household changes Useful advises to help you

St. Jerome Church - St. Jerome Catholic Church

St. Jerome Church - St. Jerome Catholic Church

After Party Flyer

Canyon Club Newsletter