Grey-Box-Processing – data processing and analysis at the interface between experiment and simulation

The optimal utilization of results from real-world crash tests for the validation and optimization of numerical crash simulations requires the integration of experimental and virtual data sets. © FE model: NHTSA/DOT. © Photo: Adobe Stock

The continuously increasing requirements on future vehicle systems regarding safety, costs, development times and the environmental impact lead to growing challenges within the automotive development process. To overcome these challenges, the continuation of the digitalization and virtualization of the development process is of vital importance. A reliable virtual vehicle development calls for simulation methods and models, which are optimized and validated based on the results of few real-world experiments.

To allow for a reduction of the amount of time- and costintensive crash tests, the detection of the greatest possible amount of information about the experimental behavior as well as the optimal utilization of this information is necessary. For this reason, along with novel measurement systems, methods for data processing and analysis at the interface be- tween experiment and simulation are developed at Fraunhofer EMI. These new methods allow for the transfer of complexly structured experimental data sets into the simulation environment in the shape of a virtual solution corridor. The resulting integration of virtual and experimental data sets allows for the identification and assessment of deviations between experiment and simulation as well as for the determination of improvement measures derived from physically based characteristic values.