Multi-Objective Design of An Induction Motor Using A New Method For Finding The Optimal Solution By The Pareto Front Based On Harmony Search

Abstract

This article introduces a novel method for choosing the optimal solution from the Pareto front generated by the multi-objective harmony search (MOHS) algorithm, specifically aimed at optimizing an induction motor. We formulate both single- and multi-objective optimization problems that minimize active mass while maximizing efficiency and rated torque. To effectively identify and manage the Pareto-optimal front, we utilize non-dominated elitist fast sorting and crowding distance. Comprehensive simulations using MATLAB software were conducted on a 18.5 kW, 4- pole, 50 Hz squirrel-cage induction motor serving as our test system. Furthermore, in-depth comparisons were made between our proposed method and established approaches like the fuzzy membership approach and the geometric mean, all vying to select the best solution from the Pareto set generated by the Harmony search algorithm. This rigorous evaluation highlighted the superior efficiency and robustness of our method, demonstrating significant improvements compared to the motor’s initial values: 0.4377% for rated torque, 0.1766% for efficiency, and a remarkable 7.1048% for active mass.