A novel multi-objective optimization approach of machining parameters with small sample problem in gear hobbing

Low carbon gear hobbing is an environmentally friendly way to machine massive workpieces. The appropriate process parameter decision-making is of great significance to improve processing quality, reduce the machining time, production cost, and carbon emission in gear manufacturing. This paper first proposes a support vector machine/ant lion optimizer/gear hobbing (SVM/ALO/GH) integrated approach to do the multi-objective optimization of machining parameters for solving small sample problem of batch production. The first population of process parameters is generated by the multi-class SVM method. Pareto improvement and ALO algorithm are employed to obtain the optimal process parameters. Finally, the case study is presented to give a clear picture of the application of the optimization approach. The results uncover that the proposed SVM/ALO/GH method has better performance than the improved back propagation neural network/differential evolution (IBPNN/DE) algorithm over the small sample problem.