基于神经网络的汽车连杆锻压工艺优化  

Optimization of Automobile Connecting Rod Forging Process Based on Neural Network

作　　者：郝刚[1] 刘艳春[2] 

机构地区：[1]石家庄职业技术学院,河北石家庄050081 [2]河北科技大学理学院,河北石家庄050018

出　　处：《热加工工艺》2017年第13期176-179,共4页Hot Working Technology

基　　金：河北省人力资源与社会保障厅课题(JRS-2015-1127)

摘　　要：以模具加热温度、预热温度、始锻温度、终锻温度和锻压速度为输入层参数,以冲击性能、耐磨损性能为输出层参数,构建了汽车连杆锻压工艺优化的5×25×15×2四层神经网络模型。结果表明,神经网络模型的预测误差小于3%,具有较强的预测能力和较高的预测精度。与生产线原锻压工艺相比,采用优化工艺生产的汽车连杆冲击吸收功增大24%,磨损体积减小35%。使用优化工艺生产的汽车连杆冲击性能和耐磨损性能得到提高。Taking die heating temperature, preheating temperature, initial forging temperature, finish forging temperature and forging speed as input parameters, and taking the impact performance and wear resistance as the output parameters, the 5× 25×15×2 four layers neural network model of optimized automobile connecting rod forging process was established. The results show that the prediction error of neural network model is less than 3%, which has strong prediction ability and high prediction accuracy. Compared with the original forging process in production line, the impact absorption energy of the automobile rods produced by the optimized forging process increases by 24%, and the wear volume decreases by 35%. The impact property and wear resistance of the automobile connection rods produced by the optimized process are improved.

关 键 词：汽车连杆 神经网络 锻压工艺 冲击性能 耐磨损性能 

分 类 号：TG316[金属学及工艺—金属压力加工]