论文部分内容阅读
通过对铸件缺陷与影响因素间因果关系的分析 ,利用MATLAB中的NeuralNetworkToolbox仿真环境和BP模型算法建立了用于铸件缺陷分析与控制的神经网络模型 ,详细论述了模型结构的设计、数据处理、网络初始化、训练与仿真的过程。实践表明 ,基于MATLAB的铸件缺陷分析与控制模型能有效地提高效率及直观地结果显示 ,对提高铸件质量及进一步研究具有积极作用。
Through the analysis of causality between casting defects and influencing factors, the neural network model for casting defect analysis and control is established by using MATLAB environment NeuralNetworkToolbox and BP model algorithm. The design, data processing, network Initialization, training and simulation process. Practice shows that, based on MATLAB casting defect analysis and control model can effectively improve the efficiency and intuitive results show that to improve the casting quality and further research has a positive effect.