Bayesian Physics-Informed Neural Networks for Robust System Identification of Power Systems

TL;DR

This paper introduces Bayesian Physics-Informed Neural Networks (BPINNs) for power system system identification, combining robustness to noise with confidence measures, and demonstrates their superiority over existing methods like PINNs and SINDy.

Contribution

It is the first application of BPINNs in power systems, integrating Bayesian modeling with physics-informed neural networks for robust system parameter estimation.

Findings

BPINNs outperform SINDy in noisy conditions.

BPINNs and PINNs are robust to all noise levels.

BPINNs provide confidence measures for system identification.

Abstract

This paper introduces for the first time, to the best of our knowledge, the Bayesian Physics-Informed Neural Networks for applications in power systems. Bayesian Physics-Informed Neural Networks (BPINNs) combine the advantages of Physics-Informed Neural Networks (PINNs), being robust to noise and missing data, with Bayesian modeling, delivering a confidence measure for their output. Such a confidence measure can be very valuable for the operation of safety critical systems, such as power systems, as it offers a degree of trustworthiness for the neural network output. This paper applies the BPINNs for robust identification of the system inertia and damping, using a single machine infinite bus system as the guiding example. The goal of this paper is to introduce the concept and explore the strengths and weaknesses of BPINNs compared to existing methods. We compare BPINNs with the PINNs and the recently popular method for system identification, SINDy. We find that BPINNs and PINNs are robust against all noise levels, delivering estimates of the system inertia and damping with significantly lower error compared to SINDy, especially as the noise levels increases.