基于BPNN、SVR和RF模型的7050合金高温流动应力预测  被引量:3

Prediction of high temperature flow stress of 7050 aluminum alloy based on BPNN,SVR and RF models

在线阅读下载全文

作  者:马斌[1,2] 梁强[1,2] 贾艳艳 徐永航 MA Bin;LIANG Qiang;JIA Yan-yan;XU Yong-hang(College of Mechanical Engineering,Chongqing Technology and Business University,Chongqing 400067,China;Chongqing Key Laboratory of Manufacturing Equipment Mechanism Design and Control,Chongqing Technology and Business University,Chongqing 400067,China)

机构地区:[1]重庆工商大学机械工程学院,重庆400067 [2]重庆工商大学制造装备机构设计与控制重庆市重点实验室,重庆400067

出  处:《材料热处理学报》2023年第3期196-204,共9页Transactions of Materials and Heat Treatment

基  金:重庆市高校创新研究群体(CXQT21024);2021年度校内资助项目计划(2152026);重庆市自然科学基金项目(CSTB2022NSCQ-MSX0473)。

摘  要:采用Gleeble-3500热物理模拟机对7050铝合金进行等温热压缩实验,获得了合金在变形温度为300~450℃以及应变速率为0.001~1 s^(-1)条件下的应力应变数据。在此基础上,建立了经灰狼优化算法(Grey wolf optimization,GWO)优化的反向传播神经网络(BPNN)、支持向量机(SVR)和随机森林(RF)模型并验证其预测精度。结果表明:经过GWO优化的BPNN、SVR和RF模型预测精度高于原始模型;GWO-BPNN与GWO-RF模型的预测精度比较接近,且均高于GWO-SVR;在外推数据预测上,GWO-BPNN模型的预测精度更高,在内插数据预测上,GWO-RF模型的预测精度更高。不同机器学习模型对流动应力数据的拟合效果不同,其预测精度也存在差异。The isothermal hot compression test of 7050 aluminum alloy was carried out with Gleeble-3500 thermophysical simulator,and the stress and strain data of the alloy at the deformation temperature of 300-450℃and the strain rate of 0.001-1 s^(-1)were obtained.On this basis,back propagation neural network(BPNN),support vector regression(SVR)and random forest(RF)models optimized by Grey wolf optimization(GWO)were established and their prediction accuracy was verified.The results show that the prediction accuracy of the BPNN,SVR and RF models optimized by GWO is higher than that of the original model.The prediction accuracy of the GWO-BPNN and GWO-RF models is close,and both are higher than GWO-SVR model.In terms of extrapolation data prediction,the GWO-BPNN model has higher prediction accuracy,and in terms of interpolation data prediction,the GWO-RF model has higher prediction accuracy.Different machine learning models have different fitting effects on flow stress data,and their prediction accuracy is also different.

关 键 词:7050铝合金 流动应力 反向传播神经网络(BPNN) 支持向量机(SVR) 随机森林(RF) 

分 类 号:TG146.2[一般工业技术—材料科学与工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象