Besides the low static stiffness of the machine structure and tool wear, thermal deformations are a major reason for dimensional and geometric work piece errors. As an alternative to cooling, insulation and other design measures, an indirect control compensation offers an inexpensive possibility to reduce thermally caused displacements which occur at the cutting point. The thermo-clastic displacements are calculated from the temperatures of different structure points and are corrected via the NC. In order to reduce the implementation expenditure and to achieve a good transformation capability for other thermal loads, it is necessary to only consider the displacement-relevant temperature measurement points. Finding out the relevant temperatures used to be a difficult and time consuming process. This paper presents a procedure based on a dynamic neural net that automatically determines relevant temperature probes. The quality of neural models and their ambility to generalise depends strongly on the net size, but this parameter is difficult to define in advance. Apart from the calculation of the correction parameters, the developed neural net also modifies its size during the learning phase. After the introduction of this model, compensation results for a milling machine are shown and compared to classical linear equation models.
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.