Precision and productivity are very important criteria for the evaluation of modular tool systems and require a thermally stable process with tolerances in the micrometer range. During the past decades there has been an increasing interest in compensating thermally induced errors. In this paper we investigate wheather a prediction of thermal displacement based on a nonlinear regression analysis is possible, namely using the alternating conditional expectation algorithm (ACE) introduced by Breiman and Friedman, 1985. The data we are analyzing were generated by two different finite element spindle models of modular tool systems. As the main result we find that the ACE-algorithm is a powerful tool to model the relation between temperatures and displacements. It could also be a promising approach to handle well-known hysteresis effects. Limitations of this study are the model restricted results, next our findings have to be validated on real data.

This content is only available via PDF.
You do not currently have access to this content.