提高冷轧高强度钢板屈服强度控制水平的研究

在高强度钢板中,超低碳烘烤硬化钢板(简称BH钢板)是新型优质汽车用薄板.在生产BH钢板时,需要有效地控制屈服强度,以保证较好的深冲性能.采用四方图识别了影响屈服强度的关键因素,包括化学成分、热轧参数、冷轧参数等;基于大量的生产过程历史数据,分别用回归分析、神经元网络、决策树三种不同方法进行分析,建立各关键输入变量(KIV)对屈服强度(KOV)影响的多变量模型.以神经元网络模型为例,进行了模型评价.以该模型为基础,在已知某些KIV取值的情况下,能够较为准确地预测BH钢成品的屈服强度,合格率提高7个百分点,取得了较好的经济效益.

Research on Improving High Strength Steel's Yield Strength

Among high strength sheets,extra low-carbon bake-hardening steel sheet(BH) is a new kind of automobile sheet with sound quality.To guarantee good deep-drawing property of BH steel sheet,it is required that the yield strength(YS) be effectively controlled in production.A method called Four Square Puzzle has been used to(identify) key factors that affect YS.They include chemical composition,hot rolling and cold rolling parameters, etc.A multi-variable analytical model has been built to study the effect of key input variables(KIV) on YS(KOV) through analysis of a great number of historical data on production with three different methods including regression analysis,neural network and decision tree.The model has been evaluated by taking neural network as an example.Based on this model,it is possible to predict finished BH steel product's YS if certain KIV values are available.As a result,qualification rate has climbed by 7% and better economic benefits have been reaped.