Optimization of heat treatment deformation control process parameters for face-hobbed hypoid gear using FEA-PSO-BP method

• Published in: Journal of manufacturing processes Vol. 117; pp. 40 - 58
• Main Authors: Sun, Beier, Liu, Huaming, Tang, Jinyuan, Rong, Shifeng, Liu, Yashuai, Jiang, Weizhe
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.05.2024
• Subjects: Computer simulation; Deformation prediction; Face-hobbed hypoid gear; Heat treatment deformation; Multi-field coupling; Process parameter optimization; Deformation prediction; Multi-field coupling; Computer simulation; Heat treatment deformation; Face-hobbed hypoid gear; Process parameter optimization
• ISSN: 1526-6125; 2212-4616
• DOI: 10.1016/j.jmapro.2024.02.025

The face-hobbed hypoid gears are manufactured from low-carbon alloy steel and possess a complex geometric profile. Typically, carburizing and quenching are employed to enhance the surface hardness of the gear teeth. However, deformation unavoidably occurs during the processes of carburizing, quenching, and tempering. This deformation leads to a reduction in the contact precision of the gear pair and an increase in noise during meshing. To effectively control the thermal deformation of face-hobbed hypoid gear and ensure their accuracy, a computational model for carburizing, quenching, and low-temperature tempering processes of face-hobbed hypoid gear was established using the finite element method (FEM) in this study. The computational model simulated the temperature field, phase field, and stress-strain field during the heat treatment process, and the accuracy of hardness, phase transformation structure, and deformation in the model was confirmed by experimental verification. The investigation, conducted through single-factor experiments, aimed to study the impact of process parameters such as carburizing potential, quenching temperature, oil temperature, and tempering temperature on heat treatment deformation. The study results demonstrated that the quenching temperature has a relatively significant influence on the tooth form error. To control and reduce the thermal deformation error on the tooth surface, a back propagation (BP) neural network was constructed using single factor experiments and orthogonal experiments. The optimal process parameters were obtained through particle swarm optimization (PSO) algorithm. Under the optimal process parameters, the average tooth form error of the working surface was reduced by 2.3 μm (14 %), and the range was reduced by 6.2 μm (13 %). The FEA-PSO-BP method for optimizing process parameters to reduce thermal deformation in heat treatment offers a set of parameters that minimize the thermal deformation. This method can be applied to optimize heat treatment process parameters for other types of gears and has significant technical content and broad engineering application value. •A multi-field coupled FEA model of heat treatment for face-hobbed hypoid gear is established based on experiments.•The variation patterns of alloy carbon content, hardness gradient, and microstructure during the heat treatment process were analyzed.•A single-factor experiment was established to investigate the coupled relationship between heat treatment deformation and process parameters.•Utilizing neural networks to predict tooth form errors and determine optimal process parameters.