Oscillation Monitoring with the PI Server for Large Power Systems

Transcription

1 Oscillation Monitoring with the PI Server for Large Power Systems Mani V. Venkatasubramanian Washington State University Washington State University

2 Project Objectives Oscillation Monitoring System for WECC and Entergy 2 Monitoring hundreds of PMUs simultaneously: Helps pinpoint likely source of oscillations. Improves estimation accuracy. Damping Monitor Engine ambient data analysis Event Analysis Engine detection and analysis of ringdowns and oscillations Online and off-line engines with easy interfaces for the PI Server

3 Oscillation Monitoring System Online IEEE C Real-time PMU data stream Data stream OpenPDC OMS OpenPDC OMS SQL server OMS results PI Interface Adapter PI Server OMS action adapters built into OpenPDC 64 bit version 2.1 Washington State University 3

4 Oscillation Monitoring System Off-line Comtrade, CSV Historical PMU data PI Adapter OpenPDC DMO EAO OMS Visualization OMS results CSV files PI Data Server Stand alone fast oscillation analysis programs for analyzing large-scale PMU data directly from the PI Server. Washington State University 4

5 Voltage (V) Results from Two Engines 5.15 x 10 5 Bus Voltage Magnitude at Cumberland Event Analysis Ambient Noise Analysis 1.2 Hz at +1.8% damping. Local Mode. 1.2 Hz at +1.5% damping. Local Mode. Washington State University Time (s) 5

6 Complementary Engines Event Analysis Engine (EAE) Multiple algorithms Prony, Matrix Pencil, HTLS, ERA, MFRA, METRA. Aimed at events resulting in sudden changes in damping Damping Monitor Engine (DME) Ambient noise based. Continuous. Provides early warning. Fast Frequency Domain Decomposition (FFDD), Washington State University DFDO, Recursive Adaptive Stochastic Subspace 6

7 Damping Monitor Engine Ambient noise based. Continuous. Provides early. Time-domain algorithms: Stochastic Subspace Identification (SSI-Covariance) Recursive Adaptive Stochastic Subspace Identification (RASSI) Distributed Recursive Stochastic Subspace Identification (DRSSI) Fast Stochastic Subspace Identification (FSSI) Frequency-domain algorithms: Fast Frequency Domain Decomposition (FFDD) Distributed Frequency Domain Optimization Washington (DFDO) State University 7

8 Fast Frequency Domain Decomposition (FFDD) Power spectrum estimation by FFT and Multi- Taper Method Apply SVD on the power spectrum Apply inverse FFT on largest singular values Extract pole frequency and damping ratio from exponential form by ringdown analysis Can process signals simultaneously. Washington State University 8

9 Turbo Oscillation Monitoring Can process truly large number of signals simultaneously online. Offline mode: Can get a quick overview of system modal properties. Mode trends. An hour of data from 200 PMU signals can be analyzed in less than 2 minutes on a desktop Built using PI AFSDK. Easy Oscillation Analysis of Large-scale PI Server for PMU data Washington State University 9

10 0.38 Hz WECC mode (poorly damped) Washington State University 10

11 June 13th PSD Singular Values from WECC data Washington State University 11

12 Mode Shapes on June 13, Hz at Near Zero Damping Ratio (7.30 am to 8.00 am) 0.4 Hz at Near 8% Damping Ratio (10 am to 11 am) Washington State University 12

13 June 13th 0.37 Hz oscillations at Generator Washington State University 13

14 Generator MW Oscillations Hydro operated in rough zone when wind power output high. Vortex effect in Francis turbine when water flow level is low 5 to 25 MW oscillations observed at 0.37 Hz Can potentially lead to resonance with system inter-area modes Mode shape analysis critical Multi-dimensional analysis crucial Washington State University 14

15 Event Analysis Engine Algorithms for oscillation monitoring Prony s Method, Matrix Pencil Method, Hankel Total Least Square (HTLS), Eigenvalue Realization Algorithm (ERA) Each algorithm has its own advantages and disadvantages Speed Noise performance Consistent estimations needed for reliable estimation 15

16 Event Analysis Engine WECC Event 16

17 PI Interfaces 17

18 PI Interfaces 18

19 PI Interfaces 19

20 PI Interfaces 20

21 Damping Monitor Offline Demo 21

22 Event Analysis Offline Demo 22

23 Oscillation analysis of hundreds of PMU signals fast and easy. Event detection, mode trends, mode shape trends Washington State University 23 Summary PMUs enabling technology for online oscillation analysis PI Server efficient storage of large-scale PMU data WSU oscillation monitoring engines provide efficient analysis of PI Server historical data Fast and easy extraction of historical PI Server data using custom interfaces

24 Mani V. Venkatasubramanian Professor Washington State University, Pullman WA Copyright 2015 OSIsoft, LLC 24

25 Questions Please wait for the microphone before asking your questions State your name & company Copyright 2015 OSIsoft, LLC 25

26 Please contact for information on WSU oscillation monitoring software for the PI Server. Copyright 2015 OSIsoft, LLC