Estimation of generator excitation-system parameters by tabu search

The paper presents a tabu search method for estimating the parameters of a generator excitation-system model. To apply the method, it is necessary to have a measured time domain response of the generator and excitation system for a specific perturbation. The method systematically compares the measured response to simulated responses using different sets of excitation system parameters. This comparison continues until it has been determined that a set of parameters has been identified that gives the best comparison between the measured response and the simulated response. An algorithm for the tabu search has been developed, which provides quick convergence and a satisfactory solution. This algorithm has been applied for a generator and excitation system that has been field-tested. The paper shows that a set of parameters has been found by the tabu search method, which provides excellent comparison of the simulated response to the measured response.