基于实时反馈的机床热误差在线补偿模型

发布时间：2018-04-29 22:34

本文选题：数控机床 + 热误差　；参考：《中国机械工程》2015年03期

【摘要】：为建立一种能够适应机床不同工况且具有准确预测能力的热误差补偿模型,提出一种基于限定记忆递推最小二乘法辨识热误差模型参数的机床热误差预测建模方法。该方法随着机床工作状况的改变,根据实时反馈的温度和热误差数据,采用递推方法对模型参数进行即时修正,使热误差模型能够及时跟踪机床系统的热特性变化,实现以较高的预测精度对机床热误差进行补偿。通过数控车床主轴轴向热误差辨识建模及补偿实验可以看出,限定记忆递推最小二乘法比一步最小二乘法辨识精度有较大提高,最大残差值减小了52.3%,标准差减小了67%。实验结果表明,利用该方法进行机床热误差模型参数辨识具有较高的预测精度和鲁棒性,有效可行。
[Abstract]:In order to establish a thermal error compensation model which can adapt to different working conditions of machine tools and have accurate prediction ability, a method for predicting thermal error model parameters of machine tools based on finite memory recursive least square method is proposed. With the change of the working condition of the machine tool, according to the temperature and thermal error data of real-time feedback, the recursive method is used to modify the parameters of the model in real time, so that the thermal error model can track the change of the thermal characteristics of the machine tool system in time. The thermal error of machine tool can be compensated with higher prediction precision. Through the identification modeling and compensation experiment of the spindle axial thermal error of NC lathe, it can be seen that the precision of identification by the finite memory recursive least square method is much higher than that of the one-step least square method, the maximum residual value is reduced by 52.3 percent and the standard deviation is reduced by 67 percent. The experimental results show that the method is effective and feasible for the parameter identification of the thermal error model of machine tools with high prediction accuracy and robustness.
【作者单位】：西安理工大学;
【基金】：国家自然科学基金资助项目(51375382) 国家科技重大专项(2012ZX04012032) 陕西省自然科学基础研究计划重点项目(S2009JC1400)
【分类号】：TG502

【参考文献】