Online Detection and Quantification of Transient Instability using Lyapunov Exponents from PMU Data

Lyapunov Exponents (LE) have been shown to be an effective tool in predicting short-term voltage instability in power systems by determining if their voltage trajectories indicate that a system is stable or unstable. This paper provides an improved LE algorithm for use in real-time streaming applications for short-term voltage instability detection. Additionally, this paper proposes three variants to the traditional LE calculation, which estimate the final value of the LE and accurately quantify the stability of a stable power system. The proposed modifications to the LE calculation are applied to voltages sampled at Phasor Measurement Unit (PMU) data rates. The streaming algorithm presented is used to activate a Battery Energy Storage System (BESS) to improve power system stability. Results are validated by using the IEEE 39 Bus model, a Duke Energy system model, and historical events recorded by Duke Energy PMUs.