Learning milling machine for the adaptive machining of carbon fibre reinforced plastics

• Applicant: Professor Dr.-Ing. Eckart Uhlmann
• Subject Area: Metal-Cutting and Abrasive Manufacturing Engineering
• Term: since 2019
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 420609123

Due to their advantageous mechanical properties, carbon fibre reinforced plastics (CFRP) are increasingly being used in the key industries of the automotive and aviation sectors. The investigation and optimization of milling of CFRP therefore remains relevant. The aim are processes that achieve high surface quality with high material removal rates and low tool wear rates. The challenge of processing CFRP results from the fibres which lead to faster cutting edge rounding. Machining with a worn tool can in turn result in post-process finishing operations due to unwanted workpiece damage such as delamination and fibre protrusions. To identify suitable process parameters many research projects have been carried out and, in some cases, regression models have also been set up, for example focussing on the delamination factor. The number of machining experiments carried out is usually in the lower two-digit range, so that non-linear influences of the process parameters and tool wear on the target values are inevitably left unconsidered. Due to the anisotropy of CFRP, FEM-supported simulation only provides very limited reliable statements about the process. The development of machining processes will therefore continue to be based on the implementation of test series and expert knowledge, at least in the medium term. Experimentally parameterised process monitoring based on high-resolution acoustic emission and modern evaluation methods can make a significant contribution to the development, optimisation and monitoring of machining processes. Due to the high complexity of machining CFRP as a material with material anisotropy and the changing fibre orientation angles, there is still potential for optimisation beyond previous research approaches. A lack of process understanding in CFRP machining can also be identified as a reason why the investigation of adaptive machining of CFRP materials has not yet been the focus of development work. The potential of intelligent process parameter adaptation to compensate for the influence of tool wear and changing fibre orientation angles on component damage has not yet been researched for milling. The development planned in this research project allows the fully automated parameterisation of tool-side, acoustic emission-based process monitoring for the detection of component dam-age and tool wear for all fibre orientation angles. Artificial intelligence methods for image-based qualification and quantification of component damage and feature-based extraction of this damage from the high-frequency acoustic emission signal are being researched for the first time. This enables the systematic investigation of the potential of adaptive CFRP machining in the further course of the project. The rotating measurement of acoustic emission hereby represents the key technology for adaptive CFRP machining.

• DFG Programme: Research Grants