基于RBF神经网络的齿轮轴热锻成形工艺优化  

Hot Forging Process Optimization of Gear Shaft Based on RBF Neural Network

作　　者：巢淑娟[1] 魏利亚 CHAO Shujuan;WEI Liya(Xuchang Vocational Technical College,Xuchang 461000,China;College of Materials Science and Engineering,Taiyuan University of Technology,Taiyuan 030024,China)

机构地区：[1]许昌职业技术学院,河南许昌461000 [2]太原理工大学材料科学与工程学院,山西太原030024

出　　处：《热加工工艺》2023年第9期103-105,110,共4页Hot Working Technology

基　　金：河南省科技攻关重点项目(12210210175)。

摘　　要：建立了齿轮轴热锻成形工艺优化的RBF(径向基函数)神经网络模型,对模型计算流程、收敛特性及拟合结果进行分析。基于RBF神经网络对齿轮轴热锻成形工艺进行优化。结果表明:基于RBF神经网络模型优化后的齿轮轴热锻成形件屈服强度由425 MPa提升到456 MPa,最大成形力由565 kN降低到508 kN,屈服强度提升率为7.3%,最大成形力降低率为10.1%;齿轮轴热锻成形最佳生产工艺参数为模具预热温度300℃、坯料加热温度1150℃、摩擦系数0.3、热锻速度40 mm/s。The RBF(radial basis function)neural network model for the optimization of gear shaft hot forging forming process was established,and the model calculation process,convergence characteristics and fitting results were analyzed.Based on RBF neural network,the hot forging forming process of the gear shaft was optimized.The results show that the yield strength of the gear shaft hot forged parts optimized based on the RBF neural network model is increased from 425 MPa to 456 MPa,the maximum forming force is reduced from 565 kN to 508 kN,the yield strength increase rate is 7.3%,and the maximum forming force reduction rate is 10.1%.The best production process parameters for gear shaft hot forging forming are die preheating temperature of 300℃,blank heating temperature of 1150℃,friction coefficient of 0.3,and hot forging speed of 40 mm/s.

关 键 词：RBF神经网络 齿轮轴 热锻成形 工艺优化 

分 类 号：TG306[金属学及工艺—金属压力加工]