Modeling and optimization of a reluctance accelerator using DOE-based response surface methodology

Abstract

This paper introduces experiment-based modeling and optimization of a reduced-scale Electromagnetic launcher (EML) using the Design of experiments (DOE) technique. Response surface models describing the velocity and kinetic energy of the launched projectile were developed using the Box-Behnken method with design variable transforms, and an Analysis of variance (ANOVA) was conducted to refine the models by removing statistically insignificant terms. A bi-objective optimization problem with the maximum velocity and maximum kinetic energy as objects was considered, and a Pareto front was obtained using the generated response surfaces as the solution of the problem. Verification tests on the optimal design points were conducted to demonstrate the validity of the developed models.