Learning a Generator Model from Terminal Bus Data
This work addresses the need for fast predictive emulators in power systems, but it is incremental as it builds on existing ML techniques for a specific domain.
The paper tackled the problem of reconstructing generator models from terminal bus data using machine learning, developing both a standard VAR model and a novel customized LSTM model, with trade-offs in reconstruction ability and computational demand analyzed.
In this work we investigate approaches to reconstruct generator models from measurements available at the generator terminal bus using machine learning (ML) techniques. The goal is to develop an emulator which is trained online and is capable of fast predictive computations. The training is illustrated on synthetic data generated based on available open-source dynamical generator model. Two ML techniques were developed and tested: (a) standard vector auto-regressive (VAR) model; and (b) novel customized long short-term memory (LSTM) deep learning model. Trade-offs in reconstruction ability between computationally light but linear AR model and powerful but computationally demanding LSTM model are established and analyzed.