System Frequency Response Model Generation Using Weighted Recursive Least Squares

With the rapid development of renewable energy and the increase in the volatility of load demand, the frequency stability of power system is facing serious challenges. As an important indicator of power system stability, frequency response is significantly affected by inertia characteristics, damping coefficients, and external perturbations. The existing system frequency response (SFR) models seldom consider the influence of thermal state variables, which makes it difficult to accurately analyze the PFR performance of a unit. Considering the thermal state, a model generation scheme for the power system PFR is proposed. First, a SFR modeling method based on controlled autoregressive moving average with exogenous variable (ARMAX) is presented to capture the dynamic characteristics of the frequency response. Then, the weighted recursive least squares (WRLS) algorithm is used for online parameter estimation to improve the model’s adaptability to system parameter changes. Finally, the simulation analysis and error evaluation show that the proposed scheme can clearly reflect the PFR performance of the power system with thermal power units.