Simplified evaluation of PEM-fuel cells by reduction of measurement parameters and using optimised measurement algorithms

PEM-fuel cells operated with hydrogen and air offer promising possibilities for the decentralised energy supply in stationary and mobile applications. But, there is still a remarkable need for more research for the optimisation of the single components even though the research, especially the development of membrane-electrode units and bipolar plates, has made considerable progress within recent years. This also applies to the definition of suitable test algorithms and parameters for recording the characteristics as well as long time tests. The investigations for single cells or stacks can thereby be subdivided into investigations in the stationary and the dynamic state. This paper shows a simplified approach for the evaluation and modelling of PEM-fuel cell stacks for the stationary state. Based on the definition of regression approaches for the dependence of several parameters the number of stack parameters is initially reduced to a sufficient number of values. The remaining parameters are used to form an energy model which can be combined with the energy models of auxiliary components like air compressors and sinus inverters [1]. Algorithms for recording the parameters still have to be defined. Therefore, the tests for the preparation of a set of characteristic curves can be very time consuming because of the multitude of variable parameters. This paper presents and discusses optimised measurement algorithms for the evaluation of PEM-fuel cell stacks to reduce this time exposure