This cooperation project will create the basis for new methods and tools for the development of fuel cell systems. Although fuel cells are already used successfully in vehicles, the service life of the cells in vehicles, among other things, is the subject of ongoing investigations. Dynamic loads and transient processes that typically occur in-vehicle applications can cause the fuel cell to age faster than constant operation. This usually means a permanent reduction of the maximum power due to local damage to the cells. In order to be able to consider performance losses due to damage at an early stage during development, methods for highly dynamic control of fuel cells are to be developed in this project. These methods can be applied in fuel cell systems and on modern stack test rigs as well. In order to identify transient operating conditions that deliberately cause or prevent damage, a comprehensive model-based analysis method will be developed. The use of 3D-CFD models with high resolutions allows a detailed investigation of local damage mechanisms based on physical and chemical principles. Simplified models are to be derived from these high-resolution models with the aim of operating them in real-time. This is the prerequisite to enable the implementation of innovative online monitoring and diagnostic methods and to gain insight into the fuel cell processes.
A second focus of the project is the development of suitable innovative test methods that enable the identification and subsequent avoidance of damage-relevant state trajectories of the fuel cell. With the help of the developed models, dynamic test cycles are to be generated, which allow specific conclusions to be drawn about the performance and damage or ageing processes of the fuel cell. To this end, new methods of model-based test design are being developed, which provide for appropriate excitation of the stack on the test rig.
A third objective of the project is the defined control of a fuel cell system during transient phases. The thermodynamic states that occur, such as temperatures and pressures, must be realized quickly and reliably. For this purpose, suitable methods of non-linear multivariable control are used, which should enable a decoupling of these variables with maximum achievable dynamics. In addition to controlling the dynamic test cycles, such control methods also enable the precise thermodynamic emulation of balance-of-plant components on the stack test rig. This allows the behaviour of the stack in an overall system to be tested virtually, which enables additional quality improvement and time savings in the development of fuel cell systems. This project creates the basis and methods to improve the service life and efficiency of fuel cells specifically. By means of the developed test procedures, the development time of fuel cell systems should also be significantly reduced.