A Novel Approach for Estimation of Synchronous Machine Automatic Voltage Regulation System Parameters

Abstract

In this paper, a novel approach for the estimation of parameters of the synchronous machine automatic voltage regulation (AVR) system is proposed. The proposed approach is based on the simple test procedure– the measurement of the reference voltage value and the generator's voltage response when there is step change in the reference voltage value. The parameters estimation procedure is conducted using novel Simulated Annealing– Cooperation Search Algorithm (SA-CSA), while the Heffron–Phillips third-order synchronous machine model is applied. Unlike any other available method for the estimation of synchronous machine parameters from the literature, the proposed test procedure is very simple, easy to conduct, harmless for the machine, and does not require the disconnection of the machine from the grid. Simulation verification is carried out using simulation model of excitation system of 40 MVA synchronous machine from hydropower plant “Perucica”. Experimental verification of the proposed estimation approach is provided by carrying out the described test procedure on the real 120 MVA synchronous machine from hydropower plant “Piva”. The applicability and accuracy of the proposed approach are proven by comparing the parameters determined using proposed method with corresponding parameters from the machine catalogue (manufacturer data). Also, the superiority of the proposed algorithm is demonstrated by comparison with other frequently used and recent metaheuristic algorithms.