Parameter based characterization of topographies and the adjustment of grinding processes by self-learning models (T12#)

点击翻译按钮获取中文摘要

Focus of the sub project E1 of the CRC 653 was the interaction of grinding tool topography and the ground work piece topography. In sub project K2 process models for the prediction of tool wear and form errors of components as well as control strategies of the cutting process were developed. The present funding proposal focuses to transfer these results into applications in a cooperation between Walter Maschinenbau GmbH and the Institute of Production Engineering and Machine Tools (IFW Hannover).The aim of the planned project is the extension of a method to describe grinding tool topographies with roughness parameters. At first the used measurement method will be optimized with regard to measurement time, robustness and the application in an industrial environment. After that compiled technological knowledge about the relationship between tool wear and dressing process at the one hand site and roughness parameter of work piece and grinding tool at the other hand forms the basis for the transfer of self-learning models from K2 into application. The process model predicts tool wear and reacts with the adjustment of the process parameter or recommends a dressing process. Considering the roughness parameter the dressing process in the machine tool will be more efficient and economic. After adjusting the process models the tool grinding machine Walter Helitronic Vision 400L is able to measure and characterize grinding tools and adjust the process parameter for the next work piece. Furthermore the grinding machine knows the necessary infeed to sharpen and dress a grinding tool in a sustainable way.