This project is part of project network 8 - "Integrated data management, workflow and visualization to enable an integrative systems science" in the Cluster of Excellence Simulation Technology. The goal of the project network is to create a simulation environment that supports the user in handling of complex simulations. Among other things, the system should take over essential, but inconvenient tasks like, e.g., logging the results and the used parameters and computation steps. The creation and execution of simulation runs should also be eased. The basis for this is workflow technology, like it is used in the field of business processes. It allows building up simulations from modular components, which can be flexibly combined and reused. To enable efficient work of the user with simulations, the integration of visualization components into the system is an important aspect of this project. Thereby, the user has uniform access through a single system to all steps necessary when working with simulations - design of the simulation, execution of the simulation, and (visual) analysis of the results.

With the increasing complexity of simulations, the requirements grow for the visualization of the generated data. The visualization has to be "tailored" to every single question and task. In many cases, new questions appear during the inspection and analysis of data. A visualization environment as flexible as possible can help to analyze simulation results faster and more thoroughly. Therefore, it should be possible to compose the required visualization from different subcomponents. In the case of simulations with long running times, monitoring of the simulations is of great importance. Visualization can support this task and allow an early recognition of problems and errors in the simulation, e.g., by displaying intermediate results.

Furthermore, new visualization techniques should be developed for the field of simulation technology. This includes the visualization of uncertain data and systems, and also methods for visual comparison of different simulation results. To get a comprehensive view into the data, the simultaneous application and combination of different visualization methods is also of interest.

Moreover, new methods for the visualization of vector and tensor fields were developed within the project. A method for visualizing coherent structures in vector fields was successfully transferred to tensor fields. An application for this is, e.g., the visualization of data resulting from diffusion tensor magnetic resonance imaging. With this, it is possible to visually separate areas of coherent nerve fiber tracks in the human brain. Coherent regions of stress distribution in a simulated thigh bone were also visualized with this method. Further research resulted in a method that reduces the computational effort of an established method for vector field visualization. The effort can be reduced from linear to logarithmic complexity. For example, a time-dependent visualization in terms of 3D animation can be computed with the accelerated method in only 6 minutes instead of 3.5 hours.