精密坐标镗床进给系统热误差分析与预测

发布时间：2018-04-04 11:22

本文选题：进给系统　切入点：热误差　出处：《西安交通大学学报》2015年10期

【摘要】：为了预测数控机床运行时热误差对进给系统定位精度的影响,以精密坐标镗床为研究对象,采用红外热像仪和激光干涉仪分别测量进给系统在每个测点的丝杠温度和定位精度,提出进给系统热误差的最小二乘支持向量机(LS-SVM)预测方法,建立了关于温度与位置的预测模型。模型引入最小二乘支持向量机方法对机床进给系统热误差进行预测分析,较好地描述了进给轴热误差与温度、位置之间的非线性关系,且对样本的依赖度小,有很好的泛化能力,解决了目前线性拟合模型用特征平均温度替代当前测点温度进行计算而存在较大误差的问题。实验结果表明,与目前已经在数控机床上实际应用的线性预测模型相比,LS-SVM模型对进给系统热误差的预测精度可达90%,预测精度提高30%以上,取得了非常好的预测效果,具有较高的现实应用价值。
[Abstract]:In order to predict the influence of thermal error on the positioning accuracy of the feed system, the infrared thermal image and laser interferometer were used to measure the lead screw temperature and positioning accuracy of the feed system at each measuring point, taking the precise coordinate boring machine as the research object.A least square support vector machine (LS-SVM) method for predicting the thermal error of feed system is presented. The prediction model of temperature and position is established.The model introduces the least square support vector machine (LS-SVM) method to predict and analyze the thermal error of the machine tool feed system, and describes the nonlinear relationship between the thermal error of the feed shaft and temperature and position, and the dependence on the sample is small.It has a good generalization ability and solves the problem that the linear fitting model uses the characteristic average temperature instead of the current measuring point temperature to calculate and there is a big error.The experimental results show that compared with the linear prediction model which has been applied in numerical control machine tools, the prediction accuracy of the LS-SVM model for the heat error of feed system can reach 90%, and the prediction accuracy is improved by more than 30%, and the prediction effect is very good.It has high practical application value.
【作者单位】：西安交通大学机械制造系统工程国家重点实验室;
【基金】：国家重点基础研究发展计划资助项目(2011CB706805) 国家重大科技成果转化资助项目(2012HZ01)
【分类号】：TG659

【参考文献】