| 摘要: |
| 针对7050铝合金锻造后力学性能难以预测的特点，提出一种基于径向基函数（RBF）的神经网络预测模型。该模型以7050铝合金锻造工艺参数为输入层，力学性能为输出层，利用样本数据训练模型后进行7050铝合金锻件力学性能预测。实验结果表明，RBF神经网络在锻造7050铝合金力学性能的预测过程中收敛性良好、预测精度高，相对误差不超过6%；相较于BP神经网络，RBF神经网络具有更高的预测精度。该模型可以对7050铝合金的锻造生产过程起到良好的指导作用。 |
| 关键词: RBF神经网络 锻造工艺参数 7050铝合金 力学性能 |
| 基金项目: |
|
| Prediction of Mechanical Properties of Forged 7050 Aluminum Alloy Based on RBF Neural Network |
|
Li Xiaoqiang Du Changlin Wang Jun Song Jian Shan Xiufeng Yan Jun
|
|
Shanghai Space Propulsion Technology Research Institute, Shanghai 200000
|
| Abstract: |
| In view of the fact that’s difficult to predict the mechanical properties of 7050 aluminum alloy after forging, a network prediction model based on radial basis function (RBF) is proposed. The model takes 7050 aluminum alloy forging process parameters as input layer and mechanical properties as output layer, and the mechanical properties of aluminum alloy forging were predicted after the model was trained with sample date. The experimental results show that RBF neural network has good convergence and high accuracy in the prediction process of mechanical properties of forged 7050 aluminum alloy, and the relative error is less than 6%; compared with BP neural network has higher prediction accuracy. The model can play a good role in guiding the forging process of 7050 aluminum alloy. |
| Key words: neural network (RBF) forging process parameters 7050 aluminum alloy mechanical properties |
