Optimisation of the active surface design of high-speed progressive tools using machine and deep learning algorithms

• Applicants: Professor Dr.-Ing. Peter Groche; Professor Dr. Kristian Kersting
• Subject Area: Primary Shaping and Reshaping Technology, Additive Manufacturing ; Image and Language Processing, Computer Graphics and Visualisation, Human Computer Interaction, Ubiquitous and Wearable Computing
• Term: since 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 520459405

The main goal of the cooperative project "Optimisation of the active surface design of high-speed progressive tools using machine and deep learning algorithms" is the identification, based on machine or deep learning methods, of non-linear correlations between sensorily and numerically recordable process and quality data on the one hand and product and process parameters on the other hand in a high-speed progressive tool. Based on these data-driven models, the aim is to describe correlations between changes in the active surface in the process and product and process parameters and to use these correlations for optimisation of the active surface design. To achieve this goal, both the further development of an existing, modular progressive tool, the integration and qualification of corresponding sensor technology, the comprehensive experimental set up, the simulative mapping of the process and the development of corresponding AI models must be addressed. In the development of AI models, multimodal approaches are to be tested for the first time, in which heterogeneous experimental and simulative data are merged and used as inputs for machine and deep learning models. Identified correlations or sensitivities between product properties will then be used to optimise active surface parameters. Both Explainable Artificial Intelligence and Human-in-the-loop approaches will be tested and used in order to incorporate domain-specific knowledge into the modelling and to check the plausibility of identified correlations and exclude fake correlations. A six-stage tool consisting of deep-drawing, ironing and blanking operations for the manufacturing of a sensor housing serves as a demonstrator process. The active surfaces are to be optimised through the geometric design, dimensions and surface structuring of the tools as well as through the adaptation of temperature and lubrication conditions.

• DFG Programme: Priority Programmes
• Subproject of: SPP 2422: Data-driven process modelling in metal forming technology
• Co-Investigator: Professor Devendra Singh Dhami