保持架超塑性挤压工艺参数的BP神经网络优化

Parameters Optimization of Superplastic Extrusion Process for Solid Cage Using BP Neural Network

建立了铅黄铜超塑性拉伸温度、初始应变速率与延伸率、流变应力之间的BP神经网络预测模型,分析了变形条件与超塑性能之间的关系,根据得到的铅黄铜最佳超塑条件进行了轴承保持架超塑挤压试验。结果表明利用BP网络对轴承保持架超塑挤压工艺参数进行优化是切实可行的,所预测的铅黄铜最佳超塑变形条件能够满足成形工艺的实际需要。

The BP neural network predicting model is established among superplastie tension temperature, initial strain rate and elongation, flow stress of lead brass. Then the relationship between superplastie condition and superplastie performance are analyzed. According to the optimal superplastic forming parameters, the superplastic extrusion of the solid cage was performed. The results show that the process parameters optimization of the solid cage using BP neural network is feasible. The obtained optimization could meet the practice demand of lead brass deformation well.