Identification of Important Locational, Physical and Economic Dimensions in Power System Transient Stability Margin Estimation

Increasing renewable generation can lead to significant spatial and temporal changes the rotor angle stability boundary, such that critical contingencies may drastically change. Additionally, inherent variability of renewables increases number operational scenarios require assessment. This paper presents a methodology whereby series location-specific Decision Tree Regressors are trained, using power system variables estimate Critical Clearing Time (CCT) on locational basis throughout network. Permutation feature importance is used reveal most important for CCT estimation at each location (capturing aspects related physical characteristics, parameters as well economic dispatch). Consequently, duration fault also be made – along with identification explicitly impact fault. Results IEEE 39-bus network show accurate CCTs, mean absolute percentage error 1.19% average. Moreover, minimum 0.49%. An analysis provided, demonstrating how method assist in design targeted interventions improve margin specific locations.