灰色理论与神经网络在形状记忆合金热变形行为中的应用

• 作 者: 刘春茂 王俊京 王宏颖
• 机 构: 河南工业职业技术学院计算机学院,河南工业职业技术学院机械工程学院,
• 出 处: 铸造技术
• 发表日期: 2015年02期
• 关键词: 形状记忆合金 神经网络 灰色理论 热变形行为 预测模型 最小误差 仿真数据 变形温度 应变速率 预测值

在传统BP神经网络预测模型基础上,分析了各个因素的灰色关联度权值,建立了基于灰色理论和BP神经网络的Ni-Ti形状记忆合金热变形行为的预测模型。结果表明,仿真数据的相对误差都位于-3%~3%,最小误差仅为0.94%。实验值和预测值的相关性高达0.9971,与传统BP神经网络的预测精度相比大为提高。

Based on the traditional BP neural network prediction model, the gray relational weight of each factor was analyzed and the prediction model of thermal deformation behavior of Ni-Ti shape memory alloy was established based on gray theory and BP neural network. The results show that the relative error of the simulation data is between -3% ~ 3% and the minimum error is only 0.94%. The correlation between the experimental and predicted values ​​is as high as 0.9971, which is greatly improved compared with the prediction accuracy of the traditional BP neural network.