Statistical Identification of Chemotactic Parameter in Bacterial Foraging Algorithm

Abstract

Optimization is a complex process whose success depends on the formulation of the objective function (OF) and the selection of parameters in the optimization algorithm. This paper presents a statistical approach to select the values of such parameters in the optimization algorithms. It uses statistical tests to determine the stability of the algorithm and identify the best value of the parameter. The approach determines the chemotactic parameter in the Bacterial-Foraging Optimization Algorithm (BFOA). The advantage of the proposed approach is that it does not require (i) complete knowledge of the subject, (ii) mathematical analysis to determine the parameter, and (iii) it is extendible to n-dimensional systems. Also, a logical approach, based on the parameter sensitivity and error dynamics, to select the OF for PI controller-tuning problem in Shunt Active Power Filter (SAPF) is presented. The results obtained using the proposed approach are verified analytically.