부산대학교 도서관

Proton-Exchange Membrane (PEM) plays a potential role in offering effective and sophisticated solutions to a wide range of real-world applications. It is seen as a suitable choice to fit the emission reduction schedules and to challenge other technologies in terms of efficiency and greenhouse gases production. In automotive systems, this technology offers the advantage of being able to operate at low temperature, consuming the oxygen from the air and having short automotive startup time. This makes it a challenging alternative to the traditional technologies used in automotive systems. The fuel cell consists of a membrane that separates two electrodes (cathode and anode). This work considers only the cathode which is divided into two chambers: the channel and the Gas Diffusion Layer (GDL). The hydrogen generated from the fuel processing system, is fed into the anode of the cell stack, while air is pumped into the cathode through an air compressor. Fuel cells produce water and heat by converting the chemical energy to electrical energy. As all chemical reactions, the fuel cell's optimal efficiency depends on the operating conditions: air flow, humidity, pressure, and temperature. For a good transportation of the reactant gases (hydrogen and air), the hydration of the membrane needs to be regulated. For this reason, this paper deals with the internal parameter identification of a PEM fuel cell system, especially the flooding phenomenon. This is performed by proposing a model-based observer, which is the core of an on-line monitoring method. The proposed procedure is simple: it consists of rebuilding the chosen internal parameters, which are the vapor and the oxygen partial pressures at the GDL, aiming to monitor the hydration parameter. This strategy is cost-effective and its approach can be extended to cover the whole stack.