Optimization of Induction Motor Design using Five Artificial Intelligence Techniques

Abstract

This work optimizes material cost of a 10HP, 4 Pole, 3- phase, 50Hz squirrel cage inductor motor design using five different Artificial Intelligence (AI) techniques with a view to determining the best technique. Material cost optimization selects the optimal values of the design variables which give the least cost while satisfying some constraints. Seven design variables and six constraints were used in the optimization process.  The optimization was implemented using MATLAB software. The  results show that  the best optimized material   cost of N83, 300 was achieved with the Genetic Algorithm technique,  while Chicken Swarm Optimization, Cuckoo Search Algorithm, Artificial Bee Colony and  Bat Algorithm gave  N83,763, N84,375, N85,414 and N88,528,  respectively.  Comparisons of these results show a difference of N5,228 between Genetic Algorithm which gives the best result and Bat Algorithm which performed the least. This demonstrates the efficiency of GA over others being compared. This outcome can serve as a guide for induction motor designers and manufacturers that use AI techniques and whose goal is to roll out cheaper products and make more profit.