Comparison of the parameter estimation methods of surge arresters using modified particle swarm optimization algorithm

The measurements show that the residual voltage of the metal oxide surge arrester (MOSA) has a dynamic characteristic. The accurate modeling and simulation of this dynamic characteristic is very important for arrester allocation, systems reliability, and insulation coordination studies. Several models with acceptable accuracy have been suggested to describe this behavior. It should be noted that the determination of nonlinear elements of MOSAs is very important for all models. In this article, a comparison among the different methods of the determination of the surge arrester voltagecurrent characteristics is represented. A new method, which is the combination of particle swarm optimization (PSO) and ant colony optimization methods, is proposed to estimate the parameters of MOSA models. In the proposed algorithm, to overcome the drawback of the PSO algorithm (convergence to local optima), to improve the global search capability, and to prevent the convergence to local minima, ant colony optimization algorithm is combined to PSO algorithm. The proposed method is called modified PSO. The transient models of MOSA have been simulated by using ATP-EMTP. The results of simulations have been applied to the program, which is based on the modified PSO method and can determine the fitness and parameters of different models. The validity and the accuracy of the estimated parameters are assessed by comparing the predicted residual voltage with the experimental results. Copyright (c) 2011 John Wiley & Sons, Ltd.