Thermal deformation in machine tools is one of the most significant causes of machining errors. A new approach to predict the thermal error of machine tool is proposed. The temperature variables and the thermal errors are measured using the Pt-100 thermal resistances and eddy current sensors respectively. Fuzzy c-means clustering method is conducted to identify the temperatures, and the representative as an independent variable are selected meanwhile it eliminates the coupling among the variables. The learning and prediction of the thermal errors is achieved using minimal-resource allocating networks by treating the issue as functional mapping between the thermal shifts and the temperature variables. The network is made to predict the error map of a machining center. A traditional radial basis function model is introduced for comparison. The experiment result shows that the fuzzy cmeans clustering method and minimal-resource allocating networks combination is a fast and accurate method for thermal error compensation in machine tools.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.