非表示:
キーワード:
-
要旨:
Fuel cell systems are electrochemical energy converters, which can be applied for both, stationary as well as mobile power supply. Due to the high efficiency, low emissions and noise-free operation, fuel cells are considered to play an important role in the future electricity market. The complexity of these systems results from the fact, that they consist of three-dimensional stacks of membrane-electrode-assemblies interacting with a number of peripheral components. Moreover fuel cells can exhibit a strongly nonlinear behavior, as the electrochemical reactions do not only depend on reactant concentration and temperature, but also on electrostatic potential along the electrodes. The application of a dynamic process model can considerably contribute to the optimal design and effective operation of a fuel cell system. In the area of case studies or development of control strategies, often rough and simplified models are sufficient. In contrast, models with a very high level of detail are necessary for process analysis and the understanding of the physicochemical phenomena, taking place within the fuel cell.
The intention of this contribution is to propose a platform-independent, hierarchical structuring concept for the modeling of fuel cell systems, which is based on the network theory for chemical engineering processes (Gilles 1997, Mangold et al. 2002). This concept is extended to the structuring of electrochemical systems, such as fuel cells (Hanke et al. 2005) and allows a uniform and flexible treatment of the above-mentioned model classes. The proposed concept structures the model on the one hand in a vertical direction into different hierarchical levels (process unit level, phase level, storage level). On the other hand, the different levels are structured in a horizontal direction into components (macroscopic thermodynamic storages) and coupling elements (generalized fluxes). In this way the model features first a high flexibility on different hierarchical levels, secondly a good transparency and reusability of the models and thirdly a guaranteed consistency and compatibility of the models. Following this concept, users can focus, through the hierarchical levels, on the detail they are most interested in and they can modify this detail by replacing one elementary building block by another. They are relieved from algebraic manipulation of their model equations and from mechanical coding work. The risk of time-consuming implementation and programming errors is reduced.
