摘要
数控铣削加工变形问题一直是自动化制造领域的瓶颈问题。铣削过程的复杂性及引起变形的多因素性使加工变形问题很难得到精确的解析解。本文在相关课题研究的基础上 ,将自动控制领域的前沿科学———人工神经网络引入该问题的研究进程之中 ,采用三层反向传播BP网络模拟铣削参数与变形间的非线性关系 ,为加工变形预测及进一步实现变形控制提供科学依据。
Processing deformation in the course of numericalcontrolmilling is a key problem in the roboticized manufacturefield. The complicacy of milling processing and other factors of causing deformation make it difficulty to obtain anexact analytic solution from the processing deformation. In this paper, artificial neural networks were used to study this problem, and a three-layer BP neural network was adopted to find the nonlinear connection between milling parameters and parts deformation. Simulation results show that our method is feasible and effective.
出处
《机械科学与技术》
CSCD
北大核心
2004年第2期209-211,共3页
Mechanical Science and Technology for Aerospace Engineering
基金
军工项目"型号工程数控零件变形控制与校形技术"
国家自然科学基金"大型整体结构零件变形控制的关键技术研究"(5 0 175 10 2 )资助
关键词
加工变形
铣削参数
神经网络
BP算法
Cutting deformation
Milling parameter
Neural network
BP algorithm