摘要
为寻求计算应力状态系数的新方法,以4200轧机轧制的大量实测数据为基础,利用M atlab神经网络工具箱,以轧制前、后钢板厚度为输入神经元,以实测轧制压力并依靠压力公式进行逆运算获得的Qp为输出神经元,建立了轧制变形区的应力状态系数与轧件轧制前后钢板厚度对应关系的BP神经网络模型和GRNN神经网络模型.结果表明,用人工神经网络算法预测应力状态系数是可行的;且通过GRNN神经网络模型和BP模型的对比,说明GRNN网络具有更高的精度和更强的泛化能力.
To find a new method of calculating the stress state modulus,according to a large number of experimental data in 4200 rolling mill,BP and GRNN prediction models are established for the relationship between stress state modulus which is the output element and thickness which is the input element before and after rolling by Matlab neural network toolbox.The results indicate that using the algorithm of artificial neural network is feasible,and compared with the model of BP neural network GRNN model has better ...
出处
《郑州大学学报(工学版)》
CAS
北大核心
2009年第2期103-106,共4页
Journal of Zhengzhou University(Engineering Science)
基金
国家自然科学基金资助项目(10176010)
关键词
应力状态系数
人工神经网络
中厚板轧机
stress state modulus
artificial neural network
medium plate mill