Dynamic State Estimation for Multi-Machine Power System Using Least Mean Square Algorithm

This paper presents a dynamic state estimation method based on the least mean square algorithm to increase adaptability to measurement data and reduce runtime. The proposed method can be used to estimate the rotor angle and speed of synchronous generators in real time from the terminal voltage and phase data measured using the phase measurement unit. The proposed method is compared with the extended Kalman filter based method in a WSCC 9-bus system for three cases. The simulation results show that the RMSE of the estimation result in the proposed method decreased by at least 30% compared to the existing method, and the runtime also decreased from 0.6536 ms to 0.2439 ms. Therefore, the proposed method can be applied to power systems that are larger and contain many components. To the best of our knowledge, the proposed method is the first attempt to apply the LMS algorithm to power system dynamic state estimation.