摘要
热误差和切削力误差是影响数控机床精度的最重要的两个误差源,误差补偿技术是一种消除机床误差经济有效的方法,而有效的误差补偿依赖于准确的误差模型。在对切削加工过程中的热变形和切削力分析的基础上,选取合理的参量,采用BP神经网络和PSO算法相结合的优化方法建立了热误差和切削力综合模型。BP-PSO建模方法改善了网络模型的收敛速度和预测精度。基于所建误差模型,对一台精密车削中心加工实时补偿后使得径向加工误差从27μm提高到8μm,大大提高了车削加工中心的加工精度,验证了模型精度。
Thermal deformation and cutting force induced deformation of the machine tool structure are two most significant causes of machining errors, and error compensation technique is an effective way to improve the manufacturing accuracy of the NC machine tools, Effective compensation relies on an accurate error model that can predict the relevant errors during machining. Since a PSO-BP neural network modeling technique was developed to build the error model of thermal error and cutting force induced error. The PSO-BP neural network model not only enhances the prediction accuracy of the thermal errors but also reduce the training time of the neural networks. Experimental results showed that the machining error of a turning center has been reduced from 27μm to 8μm.
出处
《四川大学学报(工程科学版)》
EI
CAS
CSCD
北大核心
2008年第2期165-169,共5页
Journal of Sichuan University (Engineering Science Edition)
基金
高等学校全国优秀博士学位论文作者专项资金资助项目(200131)
关键词
热误差
切削力误差
粒子群算法
精密车削中心
thermal error
cutting force induced error
PSO algorithm
NC machine tool