Assessment of the Center of Inertia and Regional Inertia with Load Contribution via a Fully Data-Driven Method
This paper proposes a new comprehensive and fully data-driven methodology to estimate the center of inertia (COI) and the regional inertia, considering the displacement of the COI due to disturbances and load inertial contributions. The strategy uses the typicality-based data analysis (TDA) technique to detect the right pilot-bus that represents the COI. In the TDA, a compound of correlation and cosine similarities is implemented to approximate the actual distribution of the data and find the point (bus) closest to the mean which is elected as the pilot-bus. Then, the frequency response at the pilot-bus and the active power deviations are embedded into an autoregressive moving average exogenous input (ARMAX)-based approach to determine the regional inertia represented by an equivalent machine, whose inertia constant corresponds to the inertial contribution in the Region. % the This is done in order to perform the inertial estimation of the region using the frequency response of the pilot-bus and the active power of the region interconnections to model the region as and equivalent machine whose inertial constant is estimated using ARMAX identification technique. The methodology is tested using the IEEE 68-bus benchmark test system and an adapted version with aggregated dynamical loads, corroborating the method effectiveness.