论文部分内容阅读
以热压缩实验所得数据为基础,建立了BP人工神经网络模型,并对9Cr-1Mo钢的流变应力进行了预测。结果表明,神经网络模型预测值与实验值的相关系数达到0.999 8,平均误差为0.15%。该模型具有较高的预测精度和较好的泛化能力。
Based on the data obtained from the thermal compression test, a BP artificial neural network model was established and the flow stress of 9Cr-1Mo steel was predicted. The results show that the correlation coefficient between the predicted value and the experimental value of the neural network model reaches 0.999 8, with an average error of 0.15%. The model has high prediction accuracy and better generalization ability.