Research project: Smart Grind

Project director: Prof. Dr.-Ing. Peter Zeiler

Project worker: Marc Hönig

Brief description of the research project:

The objective of this research project is to develop a universal retrofit system for the process optimisation of surface grinders. This will allow the path planning of the grinding process to be optimised by taking account of the actual material removal. Hence, the only surfaces which are subjected to grinding are those on which grinding work is actually necessary. This approach, which is based on the intelligent evaluation of high-frequency data of the NC control, has the potential to considerably shorten the processing time and can thus significantly increase the economic efficiency of the surface grinding process. The project will also use machine learning methods to produce a condition diagnosis and a condition prognosis for the grinding disc. Information on the condition of the grinding disc helps companies to meet the high quality demands placed on the surface finish and the form and positional tolerances of the grinding process. Furthermore, this knowledge allows maintenance intervals to be implemented as required for optimum exploitation of the tool life (resource efficiency). The machine-oriented evaluation of the data to calculate the optimised path planning, and also of the condition diagnosis and prognosis, takes place in an edge. The advantages lie not only in the fact that high security standards are maintained when dealing with sensitive production data but also in the low latency of the data transmission.

Research work to be undertaken in the project

• Condition monitoring and diagnosis of machine tools
• Methods of machine learning
• Data transmission and storage in the cloud
• Data analysis in a machine-oriented edge cloud
• Process optimisation and path planning
• App programming