中碳结构钢淬透性的两种预报方法的比较研究  

Comparison of two methods for hardenability prediction of medium-carbon structural steels

作　　者：白宁[1] 刘国权[1] 宋月鹏[2,3] 吴晋彬[1] 

机构地区：[1]北京科技大学材料科学与工程学院,北京100083 [2]山东农业大学机械与电子工程学院,山东泰安271018 [3]中国科学院理化技术研究所,北京100080

出　　处：《金属热处理》2009年第2期59-64,共6页Heat Treatment of Metals

基　　金：国家自然科学基金(No.50271009)

摘　　要：采用三层反向传播(BP)人工神经网络(ANN)技术建立钢淬透性预测预报的数学模型,基于Visual Basic编程语言建立了中碳结构钢淬透性的BP-ANN预测预报系统。将其对45、35Mn2、48CrMnMo、30CrMnNiMo钢的预报结果与近年来发展的改进型非线性数学模型所得结果以及对应的试验数据进行了比较研究,发现BP-ANN预测预报系统可以根据钢的化学成分较准确地预测钢的端淬试验各个距离点处的硬度值,但是各点处预报结果之间相不关联,且难以与钢的端淬显微组织建立有机联系;而改进型非线性数学模型可以较好地预报整条端淬曲线,辅以梅尼尔(Maynier)模型还可对各点处对应的显微组织进行预测预报。将两种模型结合使用来对钢材的淬透性进行预测预报,则有望进一步提高预测预报结果的可靠性。Using three-layer back propagation（BP） artificial neural networks（ ANN ）, a numerical model for predicting hardenability of steel was proposed, and then a predicting system for medium-and-low-carbon steel was correspondingly developed using Visual Basic programming language. The predicted results of steels 45,35Mn2,48CrMnMo, and 30CrMnNiMo by the BP-ANN system were compared with experimental values and with simulated results of other models. It is shown that the BP-ANN system can predict the hardness values at particular points precisely,though these values are independent of each other, and the prediction is not based on microstructure-hardness relationship; on the other hand, the modified non-linear model can predict the shape of whole end-quench hardness curve quite well, and can predict the corresponding microstructure with the aid of Maynier model. Thus, the quality of prediction can be further improved by combining BP-ANN model and modified non-linear model.

关 键 词：淬透性 中碳结构钢 数值模拟 人工神经网络 改进型非线性模型 

分 类 号：TG115.61[金属学及工艺—物理冶金] TP183[金属学及工艺—金属学]