论文部分内容阅读
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 ,提高了预测精度并取得较好的效果
Based on the theory of neural network, the selection of control parameters for hot rolling of microalloyed steel was studied, and a set of experimental scheme for obtaining sample data was developed. The program used the Gleeble - 1500 thermal simulator to extract the rolling temperature and strain, Strain rate and the corresponding stress-strain curve were obtained, and the austenite grain size of the cross-section of the sample was obtained by microscopic observation.The data obtained by the experiment were processed by normalization.An improved BP algorithm was used to train the network, The relationship between the rolling control parameters (rolling temperature, strain, strain rate) and the parameters describing the microstructure and properties of austenitic steel (austenite grain size) was approximated by function, and the grain size of austenite and Flow stress neural network model.It has been proved by practice that the neural network model is applied to the prediction of hot rolling control to improve the prediction accuracy and achieve better results