摘要
热轧镁/铝复合板各层厚度直接决定了它能否满足不同环境条件下的使用要求,因此,研究决定各层厚度的轧制力和厚度预测模型对轧制过程控制和产品的性能提升具有重要的意义。本文建立了AZ31B镁合金和5052铝合金在轧制过程中变形抗力的PSO-BP预测模型,高精度地表征其在不同温度、应变和应变速率下的变形特性。结果表明:镁合金对应模型的的R2、RMSE和AARE分别为0.99165、3.4251和0.56%,铝合金对应模型的R2、RMSE和AARE分别为0.99496、2.8092和1.43%,验证了PSO-BP模型用于描述两种金属材料的准确性和可靠性。在镁/铝复合板轧制过程中,轧制力随着压下率和入口总厚度的增加而增大,随着入口厚比的减小,轧制力呈增大的趋势。镁/铝复合板出口厚比随着压下率和入口总厚度的增大而减小,随着入口厚比的减小,出口厚比也呈减小的趋势。
The thickness of each layer directly determines whether it can meet the requirements of use under different environmental conditions,so the study of rolling force and thickness prediction model that determines the thickness of each layer is of great significance to the control of the rolling process and the enhancement of product performance.The results show that the R2,RMSE and AARE of the corresponding model for AZ31B Mg alloy are 0.99165,3.4251 and 0.56%,respectively,and those for 5052 Al alloy are 0.99496,2.8092 and 1.43%,respectively,verifying the accuracy and reliability of the PSO-BP prediction model for describing the two metal materials.During the rolling process of Mg/Al clad plates,the rolling force increases with the increase of the reduction rate and the total inlet thickness,and the rolling force tends to increase with the decrease of the inlet thickness ratio.The outlet thickness ratio of Mg/Al clad plates decreases with the increase of reduction rate and the total inlet thickness,and with the decrease of the inlet thickness ratio,the outlet thickness ratio also shows a tendency of decreasing.
作者
苏军
郝平菊
刘元铭
王振华
王涛
SU Jun;HAO Pingju;LIU Yuanming;WANG Zhenhua;WANG Tao(College of Mechanical and Vehicle Engineering,Taiyuan University of Technology,Taiyuan 030024,China;Engineering Research Center of Advanced Metal Composites Forming Technology and Equipment,Ministry of Education,Taiyuan University of Technology,Taiyuan 030024,China;Hai’an Industry Institute of Advanced Manufacturing and Intelligent Equipment,Taiyuan University of Technology,Hai’an 226600,China)
出处
《中国有色金属学报》
EI
CAS
CSCD
北大核心
2024年第8期2627-2640,共14页
The Chinese Journal of Nonferrous Metals
基金
国家自然科学基金资助项目(52375367,52105390,52205404,52105391)
国家重点研发计划资助项目(2018YFA0707300)
海安太原理工大学先进制造与智能装备产业研究院开放研发项目(2023HA-TYUTKFYF028)。
关键词
镁/铝复合板
PSO-BP神经网络
变形抗力
轧制力
各层厚度
Mg/Al clad plates
PSO-BP neural network
deformation resistance
rolling force
thickness of each layer
