论文部分内容阅读
采用9×27×9×1四层拓扑结构,以C、Cr、V、Ti、Mo、W、Ni、Cu含量和Cr/C比值为输入参数,以磨损体积为输出参数,构建了钒钛高铬铸铁耐磨损性能的神经网络优化模型,并进行了模型预测验证和铸铁试样的显微组织、物相组成和耐磨损性能的分析。结果表明,该神经网络模型预测精度较高,输出的磨损体积相对预测误差在1.1%~2.9%;优化出的高铬铸铁成分(wt%)由3C、18Cr、1.2V、0.4Ti、1.5Mo、1W、0.1Ni和0.2Cu组成。
In this paper, a 9 × 27 × 9 × 1 four-layer topology was constructed. The input parameters of C, Cr, V, Ti, Mo, W, Ni, Cu and Cr / C were taken as input parameters and the wear volume was used as output parameter. Neural network optimization model of wear resistance of high chromium cast iron, and the prediction of model and the analysis of microstructure, phase composition and wear resistance of cast iron samples were carried out. The results show that the prediction accuracy of the neural network model is high, and the relative prediction error of the output wear volume is between 1.1% and 2.9%. The optimized composition of high chromium cast iron (wt%) consists of 3C, 18Cr, 1.2V, 0.4Ti, 1.5Mo , 1W, 0.1Ni and 0.2Cu.