Optimisation Design of a Taguchi-Based Real-Code Genetic Algorithm for Thermal Reducing of Air-Core Linear Brushless Permanent Magnet Motor

Abstract

This study combined the Taguchi method with the genetic algorithm (GA) to analyse the optimal design parameters of the thermal distribution in an air-core linear brushless permanent magnet motor (ALBPMM). First, this study adopted an L18(21×37) orthogonal array to determine the significant factors, including active currents, the length of magnets, pole distance of magnets, air-gap length, and thickness and width of coils. Then, the study uses response surface methodology (RSM) to construct the predictive model. Finally, the optimal combinations of design parameters that involve using real-code GA were obtained and verified by finite element modelling. The simulation results show that the thermal distribution in the optimal design of parameters is 41% more effective than that of any models in which the parameters are not optimised. Therefore, the proposed approach can be used as the basis for designing and predicting the temperature effects of the ALBPMM..