Visual Analysis of Multi-run Multi-field Simulation Data

People: Prof. Dr.-Ing. Lars Linsen, Alexey Fofonov

Funding: Deutsche Forschungsgemeinschaft (DFG) project grant 50557 DFG-LI-21

Description: Research in science and engineering typically relies on generating or using a theoretical model for an investigated phenomenon and running simulations with that model to validate the model or to produce new insights about the phenomenon. Spatio-temporal simulations are a major subgroup of such simulations. They investigate data fields with multiple variables and their change over time. Often, the fields are volumetric and contain variables of different type such as scalar, vector-, or tensor-valued. The fields are sampled within the (3D) spatial domain at discrete locations, which may or may not exhibit a regular structure. Moreover, the models are, typically, influenced by a number of simulation parameters or numerical parameters. The simulation results depend on the choice of the parameters. Commonly, the intervals from which to choose the parameter settings are known, but the exact values are determined empirically by running the simulation multiple times using different parameter settings. To investigate the dependence of the process on initial conditions, many simulation runs have to be performed with varying combinations of parameters values. Hence, in total, one looks into multi-run multi-variate time-varying volumetric simulation data. Aim of the project is to develop a comprehensive approach for the interactive visual analysis of multi-run multi-field simulation data addressing all their facets. Consequently, the objective is to develop visualization methods that integrate and link the aspects of volume visualization, exploration of multiple (output) simulation variables in form of multi-fields, investigation of spatio-temporal patterns including detection of features and their changes over time, and comparative analysis of multiple simulation runs including influence and correlations of (input) multi-run simulation parameters.