论文部分内容阅读
以钢材牌号、钢材长度、表面硬度和热疲劳级别为输入参数,以预热温度、预热时间、淬火温度、淬火时间、回火温度和回火时间为输出参数,构建4×12×3×6四层结构的热挤压模具钢热处理工艺神经网络模型,并进行了试验验证和现场应用确认。结果表明,神经网络模型的的热处理温度预测误差小于4℃,热处理时间预测误差小于2 min,预测精度较高。模型对生产现场的4Cr5MoSiV1热挤压模具钢预测出的热处理工艺,完全能满足企业设计要求,实用性强。
The input parameters of steel grade, steel length, surface hardness and thermal fatigue grade were taken as input parameters. The parameters of preheating temperature, preheating time, quenching temperature, quenching time, tempering temperature and tempering time were used as output parameters to construct 4 × 12 × 3 × 6 four-layer structure of hot extrusion die steel heat treatment process neural network model, and the experimental verification and field applications confirmed. The results show that the prediction error of heat treatment temperature is less than 4 ℃, the prediction error of heat treatment time is less than 2 min, and the prediction accuracy is high. Model on the production site 4Cr5MoSiV1 hot extrusion die steel predicted heat treatment process, fully able to meet the design requirements of enterprises, practicality.