摘要
基于神经网络原理 ,对微合金钢热轧控制参数的选取进行了研究 .制订了一套获取样本数据的实验方案 .该方案利用Gleeble - 15 0 0热力模拟机提取了轧制温度、应变量、应变速率和相应的应力应变曲线 ,并通过显微观察获取了实验后样品断面的奥氏体晶粒尺寸 .通过归一化把实验所得数据进行必要的处理 .采用改进BP算法训练网络 ,对热轧控制参数 (轧制温度、应变量、应变速率 )和描述微合金钢组织性能的参数 (奥氏体晶粒尺寸 )之间的映射关系进行了函数逼近 ,建立了奥氏体晶粒尺寸及流变应力神经网络模型 .实践证明 ,将该神经网络模型运用于热轧控制预报 。
Based on the method of artificial neural networks, a new approach is provided to estimate hot rolling quality of the micro alloyed steel. The new means is developed, which obtained data of hot rolling. In the research rolling temperature,strain,strain rate and the curve of stress and strain are gained by using Gleeble-1500 thermal mechanical machine and then grain sizes and flow stress are measured. The data are deal with by ANN, and the artificial neural networks originated BP arithmetic is built through training. Based on the result of networks' estimating, the prediction of austenite grain size and flow stress of micro alloyed steel in hot rolling can realized and the mathematics model generated by ANN has higher accuracy.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2002年第1期40-43,74,共5页
Journal of Harbin Institute of Technology
基金
国家自然科学基金资助项目 (5 97810 0 4)
东北大学轧制技术及自动化国家重点实验室资助项目 .
关键词
微合金钢
轧制控制
奥氏体晶粒尺寸
神经网络
BP算法
预报模型
热轧
micro alloyed steel
controlled rolling
austenite grain size
neural network
BP method
update BP method
prediction model
