摘要
针对NURBS曲线插补中求导误差大、计算复杂等问题,提出了采用RBF神经网络对自由曲线进行插补的方法;并以弓高误差为约束条件,建立一个进给步长可变的神经网络数控插补模型。另外根据曲线曲率的变化规律找出插补过程中的速度敏感点,在到达该速度敏感点之前将速度降低到指定值,从而进行有效的速度规划。最后在MATLAB平台上进行仿真实验,结果表明,该模型结构简单、计算量小、插补精度高,能够在保证加工效率和加工质量的同时降低柔性冲击。
In order to solve the problem of large error derivation and complex calculation in NURBS curve interpolation,the method of interpolating free curves by using RBF neural network is proposed.And a neural network numerical control interpolation model with variable feed step is established by using the bow height error as the constraint condition.In addition,the velocity sensitive point in the interpolation process is found out according to the change law of curvilinear curvature,and the speed is reduced to the specified value before reaching the speed sensitive point,thus the effective speed planning is carried out.Finally,the simulation experiment on the MATLAB platform shows that the model has the advantages of simple structure,small calculation and high interpolation precision accuracy,which can reduce the flexible impact while ensuring the processing efficiency and quality.
作者
邬再新
李华兵
WU Zai-xin;LI Hua-bing(Institute of Mechanical and Electrical Engineering,Lanzhou University of Technology,Lanzhou 730000,China)
出处
《组合机床与自动化加工技术》
北大核心
2019年第2期49-52,共4页
Modular Machine Tool & Automatic Manufacturing Technique
关键词
自由曲线
神经网络
数控插补
S型加减速
free curve
neural network
numerical control interpolation
S type acceleration and deceleration