Project description

Industrial manufacturing of components involves the sequential combination of manufacturing processes to create products with a defined property profile. The challenges lie in optimizing the manufacturing process, addressing process-related uncertainties, and performing global optimization across the entire process chain. Multi-criteria optimization is challenging because there are conflicting target values with regard to processing and functional properties.

Process chain optimization in ProModFun is to be carried out using an experimental approach and data-based modeling. First, the process chain is set up consisting of (1) thermal spraying, (2) turning, and (3) diamond smoothing to produce a functional surface with a graded increase in hardness near the surface. The innovative property profile will be implemented using the example of the “guide roller” component. Thermal spraying (primary shaping) is used to apply manganese hard steel surfaces to rotationally symmetrical components, which are then adapted in terms of core and edge properties by turning and diamond smoothing. The target variables of the process chain to be optimized are to be influenced as follows: (1) maximize surface hardness, (2) minimize surface roughness, (3) maximize the oxide content of the coating system, and (4) maximize the energy efficiency of surface production.

Sensors are integrated into the manufacturing processes and the measurement data is used for modeling. Statistical modeling and multidimensional, pattern-based description using grey-box AI algorithms (fuzzy pattern classification) serve as procedural approaches. They enable the quantification of various types of uncertainty and the forward coupling of the process chain. This is followed by the development of a procedure for inverse multi-criteria optimization across the entire process chain for the purpose of global optimization.

Process chain