Multidisciplinary design optimization of a gear train transmission

Abstract

This article presents a method for multidisciplinary design optimization of a one-stage gear train transmission for an industrial application. The formulation and implementation that enable the integrated design of the gearbox elements (gears, shafts, and bearings) are detailed. The analytical formulation problem is based on four disciplines: product reliability, customer preference, product cost, and structure. The proposed integrated design process takes into account constraints imposed by quality standards. The optimization of the gear train transmission is performed according to a multidisciplinary feasible architecture and uses a population-based evolutionary algorithm (non-dominated sorting genetic algorithm II) to generate Pareto-optimal fronts. Finally, a detailed case study is presented to illustrate the effectiveness of the proposed approach.