摘要
以奥贝球铁的一步等温、二步等温淬火温度和一步等温、二步等温淬火保温时间4个工艺参数作为神经网络的输入层参数,以拉伸性能为输出层参数,构建了4×4×1的三层结构的BP神经网络的奥贝球铁热处理工艺优化神经网络模型,并进行了模型的预测和验证。结果表明:该神经网络模型能较好地反映热处理工艺参数与拉伸性能之间的内在规律,BP神经网络预测平均相对误差不超过3.5%,采用BP神经网络对奥贝球铁热处理工艺进行优化,可明显提高奥贝球铁的拉伸性能。
Four parameters of one-step isothermal,two-step isothermal quenching temperatures,and one-step isothermal quenching and two-step isothermal quenching holding time of austempered ductile iron,were taken as the input layer parameters of the neural network,and the tensile properties as the output layer parameters,an optimized neural network model of austempered ductile iron heat treatment process based on 4×4×1 three-layer BP neural network was constructed.The prediction and verification of the model were also carried out.The results show that,the neural network model can better reflect the inherent law between heat treatment parameters and tensile properties.The average relative error of BP neural network prediction is not more than 3.5%.The BP neural network is used to optimize the heat treatment process of austempered ductile iron,and the tensile properties of austempered ductile iron are obviously improved.
作者
邹伟
王荣吉
张立强
俞杰
童希
ZOU Wei;WANG Rongji;ZHANG Liqiang;YU Jie;TONG Xi(College of Mechanical and Electrical Engineering,Central South University of Forestry and Technology,Changsha 410000,China)
出处
《热加工工艺》
北大核心
2020年第6期132-135,共4页
Hot Working Technology
基金
湖南省教育厅科学研究重点项目(14A157)
湖南省自然科学基金面上项目(2018JJ1262)。
关键词
BP神经网络
奥贝球铁
热处理工艺优化
拉伸性能
BP neural network
austempered ductile iron
heat treatment process optimization
tensile properties
