论文部分内容阅读
采用3×12×2三层拓扑结构,以退火温度、淬火温度、回火温度作为输入层参数,以抗弯强度和磨损体积为输出层参数,构建了M2高速钢(HSS)热处理工艺的神经网络优化模型。并对模型进行了训练、预测验证与现场应用。结果表明:该神经网络优化模型有高的预测能力和预测精度,输出的抗拉强度和磨损体积的平均预测相对误差分别为1.93%、1.89%。与原结果相比,使用神经网络优化工艺参数热处理的M2高速钢的抗弯强度增加35%、磨损体积减小41%。M2高速钢的最佳退火、淬火和回火温度分别为:(810±10)、(1230±10)、(520±10)℃。
The 3 × 12 × 2 three-layer topology was used as the input layer parameters based on the annealing temperature, quenching temperature and tempering temperature, and the flexural strength and wear volume were taken as output layer parameters to construct the nerve of M2 HSS heat treatment process Network optimization model. And the model has been trained, forecast verification and field application. The results show that the neural network optimization model has high predictive ability and prediction accuracy, and the average relative errors of the predicted tensile strength and wear volume are 1.93% and 1.89% respectively. Compared with the original results, the bending strength of M2 high speed steel heat-treated by using neural network is increased by 35% and the wear volume is reduced by 41%. The optimum annealing, quenching and tempering temperatures of M2 high speed steel are (810 ± 10), (1230 ± 10) and (520 ± 10) ℃, respectively.