Optimization techniques for parameter estimation of dynamic load models

In this paper, the performance of different optimization techniques is evaluated for the estimation of dynamic load model parameters. Two categories of methods are considered: the nonlinear least-squares-based and the population-based. The accuracy of the optimization methods is assessed by applying Monte Carlo simulations using artificially created distorted data. In order to represent different real-world conditions, the performance of the methods is evaluated considering different levels of signal-to-noise ratio. The findings of this paper indicate that similar results are obtained by the different techniques examined and verify the validity of the optimization methods for parameter estimation of dynamic load models.