Ambient temperature-induced thermal error modelling for a special CMM at the workshop level based on the integrated temperature regression method

Guangjie Jia,Jianbin Cao,Xu Zhang,Nuodi Huang

DOI: https://doi.org/10.1007/s00170-022-09533-1

2022-07-24

Abstract:Coordinate measurement machines (CMMs) at the workshop level are gaining increasing significance for achieving high machining efficiency while ensuring machining accuracy. As thermal error significantly affects the measurement accuracy of CMMs at the workshop level, it is crucial to predict and compensate for this kind of error. Based on functional requirements, we investigated the thermal positioning error of the Z-axis of this special CMM. The main thermal error source is determined as the ambient temperature for the considered CMM running at a low measuring speed. Usually, the ambient temperature is used as a single variable. We believe that this is inaccurate, because when the temperature changes drastically, the temperature at different locations would be different owing to uneven heat transfer. Therefore, in the study, we split the ambient temperature into multiple temperature variables. This leads to a strong correlation among the variables and a reduction in the accuracy and robustness of the error model. Then, we developed an integrated temperature regression method for thermal error modelling. This method merged these temperature variables into one temperature variable for thermal error modelling based on error separation. Afterwards, the integrated temperature model is compared to the single temperature and ridge regression models. The results show that the proposed temperature regression method can eliminate the collinearity between input variables and simplify the overall thermal error modelling. In addition, the compensation accuracy of the proposed model can be controlled within 5 μm at a large ambient temperature range (10–35 °C) by using only two temperature sensors.

engineering, manufacturing,automation & control systems