Generator Model Validation and Calibration using Synchrophasor Data

Abstract

Model validation and calibration of generating units are critical for using the most updated models in planning and operation of power grid. Typically, validation and calibration is done manually in couple of years at substations, which are not effective due to the test expenses and the requirement to take generator out of service for several days. However, high resolution data from Phasor Measurement Unit (PMU)s possible through automation in ongoing smart grid activities can help in validation and calibration process. Variations of signal-playback form the power grid dynamic simulator and comparing the results with PMU measurements for different events is one of the possible approach for model validation. Calibration requires optimizing generator parameters to fit with response obtained by field PMU measurements. This is a challenging problem to solve given multiple parameters to tune in generators and given limited PMU measurements. This paper automates the process of model validation and parameter calibration by using a commercial dynamic simulator, PSS/E and developing all the interfacing with Ensemble of Kalman Filter tool. Additionally, the advantage, and limitations of existing optimization methods for calibration, specially Ensemble Kalman Filter (EnKF) are discussed. A sensitivity test is devised to diminish the subset of to-be-calibrated parameters and our framework in PSS/E is proposed with simulation results. Advantage and limitations of the approach are discussed and guidelines for future works are presented.