基于BP神经网络的CrMnBH类钢淬透性预报  

Prediction of CrMnBH Types Steel Hardenability Based on BP Neural Network

作　　者：李长宏[1] 王萍[1] 沈千成 黄贞益[1] 吴勇[1] 

机构地区：[1]安徽工业大学冶金工程学院,安徽马鞍山243002

出　　处：《热加工工艺》2016年第20期248-251,共4页Hot Working Technology

摘　　要：实际生产过程中汽车用钢的淬透性很难控制,各因素之间的关系呈非线性映射。通过对影响淬透性化学元素的分析和改进的BP人工神经网络,构建了优化的汽车用钢淬透性预测模型。结果表明:实验值与预测值之间的误差在6%以内,其预测的准确性高,成功应用到某钢厂的现场生产预报。In actual production process, it is difficult to control the hardenability of steel for automobile, and the relationship between the factors is nonlinear mapping. Through the analysis of the chemical elements of influencing the hardenability, the optimized forecast model of the hardenability of the steel for automobile was built by improved BP artificial neural network. The results show that the error between the experimental value and the predicted value is within 6%. Its prediction accuracy is high, and it is successfully applied to the field production forecast of a steel mill.

关 键 词：改进 BP 神经网络 淬透性 汽车用钢 

分 类 号：TG156.3[金属学及工艺—热处理]