A Modified PSO Algorithm for Parameters Identification of the Double-Dispersion Cole Model

Abstract

Many practical systems, such as thermal system, economic system and electric power system, can be more accurately described by the fractional-order system rather than integer-order system. Therefore, it is an important topic to study the fractional-order system and estimate its parameters. The problem of parameter estimation is essentially a multi-dimensional parameter optimization problem. In this paper, according to the average value of position information, an improved Tent mapping and a piecewise mutation probability, a modified particle swarm optimization (MPSO) algorithm is presented to solve the parameter estimation problem. The performance of MPSO is tested with eight benchmark functions, which proves the effectiveness of the algorithm. Based on the double-dispersion Cole model, the proposed MPSO algorithm is used to estimate the parameters for the generated simulated datasets. Experimental results show that the MPSO algorithm for parameters identification of the Cole model is an effective and promising method with high accuracy and good robustness.