【作者】 张伟；

【导师】 叶文华；

【作者基本信息】 南京航空航天大学 ， 机械电子工程， 2013， 硕士

【摘要】 本文以QLM27100数控机床为研究对象，以灰色系统理论为基础，进行温度测点优化和热误差建模，并对温度测点优化和热误差建模方法进行了改进。主要研究内容如下：（1）对QLM27100数控机床的结构进行了介绍，对机床热源进行了初步分析，确定了机床上温度传感器和位移传感器的安装位置，并进行了数据采集试验。（2）利用灰色关联分析法与模糊聚类分析法相结合对机床关键温度测点进行优化选择。先利用灰色关联分析理论选取对机床热误差影响较大的温度测点，再采用模糊聚类分析理论对已选择的温度变量进行聚类分析，得出最佳温度变量分组。最后，用两种不同方法对QLM27100数控机床进行温度测点优化选择，建立多元回归热误差模型进行比较，结果表明：采用灰色关联模糊聚类分析法对温度测点进行优化选择较合理。（3）利用灰色系统模型和时序分析模型相结合的方法对热误差进行建模。先采用灰色系统模型对数控机床热误差进行建模，再利用时序分析法对机床剩余热误差进行建模，得到最终的综合模型。结果表明：灰色系统模型与时序分析模型相结合的建模方法，能有效地提高机床热误差模型精度。（4）对热误差补偿系统进行了介绍，以QLM27100数控机床为试验对象，对灰色系统时序分析模型进行了验证，基于SINUMERIK840D数控系统的温度补偿功能进行补偿试验，取得了较好的补偿效果。更多还原

【Abstract】 A CNC machine tool QLM27100is as the research object. The methods of the temperaturemeasuring point optimization and the thermal error modeling are based on the theory of Gray system,and the methods are improved. The main research contents are as follows:The structure features of CNC machine tool QLM27100are introduced and the heat resourcesare analyzed. The location of temperature sensors and displacement sensors fixing on the machine isdetermined, and the data of temperature and thermal error was collected.The method to optimize the temperature measuring points is combined gray correlation analysiswith fuzzy clustering analysis. The theory of gray correlation analysis is used to select sometemperature measuring points by the influence of thermal error. The selected temperature measuringpoints are clustered by fuzzy clustering analysis, and the group of optimum temperature measuringpoints is got. Finally, two different methods of temperature measuring point optimization are used onQLM27100CNC machine tool, and the two groups of optimum temperature measuring points areused to build multiple regression model. The result of comparing the precision of the two modelsshows that the method of gray correlation analysis and fuzzy clustering analysis is better.The method of thermal error modeling is used grey system model and time series analysis model.The grey system model is used to build thermal error modeling, and the residual thermal error isstudied by time series analysis model. The comprehensive model is composed of the two models. Theresult shows that the comprehensive model can effectively improves the precision of thermal errormodeling.The thermal error compensation system is introduced. The QLM27100CNC machine tool is asthe research object. The grey system and time series analysis model is verified. Based on thetemperature compensation function of SINUMERIK840D CNC system, the temperaturecompensation experiment obtains better compensation effect.更多还原