Event Classification in Distribution Networks Using a Quotient Gradient System

The classification of events or sudden changes in power networks versus normal abrupt changes or switching actions is essential to take appropriate maintenance actions that guarantee the quality of power delivery. This issue has increased in importance and complexity with the proliferation of volatile resources that introduce variability, uncertainty, and intermittency in network behavior, observed as variations in voltage and current phasors. This paper proposes using a quotient gradient system (QGS) to train a two-stage partially recurrent neural network to improve event classification rate in power distribution networks using high-fidelity data from micro-phasor measurement units (µPMUs). QGS is a systematic approach to finding solutions of constraint satisfaction problems. We transform the µPMUs data from the power network into a constraint satisfaction problem and use QGS to train a neural network by solving the resulting optimization problem. Simulation results show that the proposed supervised classification method can reliably distinguish between different events in power distribution networks. Comparison with other neural network classifiers shows that QGS trained networks provide significantly better classification. Sensitivity analysis is performed concerning the number of µPMUs, reporting rates, noise level and early versus late data stream fusion frameworks.