Input Signal Selection for Measurement-Based Power System ARX Dynamic Model Response Estimation

This paper proposes a measurement-based approach to optimize the inputs of Auto-Regressive with eXogenous input (ARX) model identification in large power systems. Correlation Coefficient Index (CCI) is defined in this paper and Correlation Coefficient Map (CCM) is developed for the US Eastern Interconnection (EI) to show the correlation between any two power system output measurement signals visually. This approach is verified with EI system simulation data and applied to Frequency Disturbance Recorder (FDR) measurement data to estimate system dynamic response. The verification result shows that the number of ARX model inputs can be decreased and the estimation accuracy can be ensured by using the proposed approach.