基于BP神经网络的铝型材挤压模具优化设计

采用三层BP神经网络建立挤压模具的数学模型,利用三维刚塑性有限元模拟获得神经网络的样本信号,对神经网络模型进行训练。利用神经网络函数逼近功能,以U形铝型材在挤压工作带出口处具有最均衡的轴向挤压速度为目标,最终对U形铝型材挤压模具合理模孔位置进行了优化设计,并采用计算机模拟仿真对优化结果进行验证,表明优化结果是有效的。

Optimizing of aluminum sections extrusion die based on BP neural network

In this paper, A three-layer neural network is used and a back propagation algorithm is employed to train the network, the sampling data were got by using finite element simulation. It is determined by applying the ability of function approximation of the neural network to decrease the imbalance of metal flow at the die land exit. We achieved the optimization design of the aluminum sections extrusion die. The finite element simulation is used to verify the optimal result developed in the present work. It is demonstrated that the discussed approach is effective.