Multi-objective optimization of current-assisted splitting spinning of small module tooth-shaped part based on the combination of BP neural network and NSGA-II algorithm

Small module tooth-shaped parts (SMTSPs) with characteristics of hollow, thin wall-thickness made of difficult-to-deformed metals, are one of the most precision transmission components, which are traditionally manufactured by tooth hobbing or tooth shaping. Current-assisted splitting spinning (CASS) has been introduced as a method to achieve integrated manufacturing of SMTSPs. A coupled electrical-thermal-mechanical finite element analysis (FEA) model was established based on the ABAQUS software, the deformation characteristics of the small module tooth and the mechanism of tooth filling under current-assisted splitting spinning were investigated. A BP neural network (BPNN) was used to establish the mapping relationship between process parameters of CASS and forming quality evaluation metrics, and the Non-dominated Sorting Genetic Algorithm (NSGA-II) multi-objective genetic optimization algorithm was employed to optimize the forming process parameters. The results show that the material at the tooth tip along the radial direction is in the state of tensile stress along radial and compressive stresses along tangential and axial directions, which promotes the radial flowing of the material and is beneficial the tooth filling of SMTSPs; the tooth saturation increases obviously under pulse current comparing without pulse current; the BPNN combined with the NSGA-II algorithm can reliably optimize the process parameters of the CASS, improving the forming quality of SMTSPs; experiments verified the feasibility of the process and the accuracy of the predictive model based on the optimization results.