神经网络在梯度功能材料制备中的应用

基金项目:

“863”青年科学基金资助项目

详细信息

Application of Neural Network to the Process of Functionally Gradient Materials Fabrication

摘要: 针对梯度功能材料(FGM)制备过程的复杂性,提出了利用神经网络信息处理机制进行制备材料的特性预估;实例分析表明,这一方法是有效的.同时,针对BP学习算法速度较慢,易陷入局部极小的缺点,改用函数型连接网络来提高学习速度.试验表明学习速度提高显著.
- 神经网络/梯度功能材料 /
- BP算法 /
- 函数型连接网络
Abstract: Taking account of complecities of the Functionally Gradient Materials (FGM) fabrication process, a neural network based expert system which estimates the properities of the material is presented. The experimental results show that this method is effective. To improve the learning speed of the system and reduce the possibility of local minimum, a functional-linked net is introduced. The application indicates that both learning speed and accuracy of the estimation are satisfactory.
- neural networks/functionally gradient materials /
- BP algorithm /
- functional linked net