基于神经网络的铣削复杂薄壁件受力变形分析和建模研究

铣削过程的复杂性使加工变形问题很难得到精确的解析解。为研究铣削过程中复杂薄壁件受力变形模型，将人工神经网络引入到摆线轮加工变形模型研究过程中，以有限元仿真结果为依据，通过改进的BP神经网络算法，建立了高速铣削轴承钢摆线轮铣削力与变形之间的非线性映射模型。结果显示所建立的网络模型具有较高的精度和良好的泛化能力，为进一步实现变形控制提供科学依据。

Forced Deformation Analysis and Modeling Research of Milling of Complex Thin-walled Parts Based on Neural Network

It is difficult to obtain an accurate analytic solution to the machining deformation because of the complicacy of milling process. To study the deformation model of the complex thin-walled parts,the neural networks are adopted in the research process of machining deformation model of cycloid wheel. On the basis of finite element simulation result and with the help of the improved BP neural network algorithm,a nonlinear mapping model between the milling force and deformation is established. The results show ...