Robust-Oriented Optimization of Switched Reluctance Motors Considering Manufacturing Fluctuation

Abstract

For electric drive systems, high robustness is the key performance in the design and optimization stage. Switched reluctance motor (SRM) is considered as an ideal candidate for industrial applications like electric vehicles (EVs), due to its superior thermal and mechanical performances. This article proposes an effective robust-oriented design optimization method for SRMs considering manufacturing tolerances for better finding the optimal robust design solution. A segmented-rotor SRM is taken as a design example, and the finite-element model (FEM) results are verified by the experiments. In the implementation, first, the optimization problem including the optimization objectives and constraints is defined. Sensitivity analysis is performed for the selection of strong-sensitive parameters, and the tolerance-affected parameter is considered in the optimization to ensure that the obtained optimal solution exhibits high robustness. Then, for the traditional Taguchi method, an orthogonal array (OA) is established based on the preset of design levels, and the optimal solution is determined by selecting the highest signal/noise ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$S/N$ </tex-math></inline-formula> ) ratios. An improved sequential Taguchi optimization method (STOM) is proposed by combining the process in the traditional Taguchi method with the space reduction method inside the iteration process. The optimal solution can be found with the reduction of design space until meeting the preset condition. Finally, a comprehensive comparison of the initial, Taguchi, and STOM designs is presented for the SRM. Compared with the initial and Taguchi designs, the proposed STOM can obtain a robust design solution with more centralized performance distributions.