摘要
研究硬旋铣加工时工件的热变形对提高工件加工精度、掌握硬旋铣加工技术是至关重要的。文章基于高效环保的滚珠丝杠螺纹硬旋铣工艺,围绕提高硬旋铣加工螺距精度问题,针对加工过程中的工件热伸长及误差补偿方法,通过基于BP神经网络算法的热伸长研究及补偿实验研究,探索了工件热伸长变化的特征值提取、BP预测模型的建立及验证、热伸长误差的补偿方法。结果表明:根据特征值法建立的BP神经网络热伸长预测模型精度较高,根据模型预测结果进行螺距误差插补补偿加工能够提高滚珠丝杠硬旋铣加工的螺距精度。
The study of whirlwind hard milling thermal deformation of the workpiece is essential to improve the workpiece precision and to master whirlwind hard milling technology. Hard whirling is an efficient and green processing of Ball Screw thread. This paper focuses on the improving of hard whirling pitch accuracy,for workpiece thermal elongation and error compensation method,through research of thermal elongation and based on BP neural network algorithm and experimental study of error compensation,explores the feature extraction of workpiece thermal elongation curve,and the establishment and verification of BP forecasting model,and the method of error compensation. The results show that BP neural network prediction model has high accuracy,and pitch error compensation based on prediction results can improve the hard whirling ballscrew pitch accuracy.
出处
《山东建筑大学学报》
2014年第6期530-534,563,共6页
Journal of Shandong Jianzhu University
基金
国家自然科学基金项目(51375279)
国家重大科技专项项目(2012ZX04002013)
国家青年基金项目(51105232)
关键词
滚珠丝杠
硬旋铣
BP神经网络
热伸长
螺距误差
ballscrew
whirlwind hard milling
BP neural network
thermal elongation
pitch error
