Design Of Power Control System For Multi-stack Direct Methanol Fuel Cell

Direct methanol fuel cell (DMFC) is a device translating chemical energy into electricitywith methanol as fuel. It has good application prospects in the fields of portable powersupplications and vehicle power supplications due to its advantages of simple structure, highenergy conversion efficiency and environment-friendly. The research institutions at home andabroad focus on the studies of DMFC chemical materials, such as: catalyst materials, protonexchange membrane performance and processing techniques. But the research on control ofits subsidiary bodies is still deficiency.The characteristics of DMFC are studied, then a comprehensive analysis of the keyparameters affecting the output performance is proposed. The stack operating temperature isimportant to the stack performance, thus an artificial intelligence modeling approach forDMFC temperature response is proposed. It overcomes the limitation of the traditionalmodeling methods. The complexity of an accurate mathematical model is avoided by usingneural network and adaptive neuro-fuzzy inference system(ANFIS). The optimal structureparameters of the three network types (BP, Elman, ANFIS) are given through the simulationcomparisons between different models, which makes it possible to predict and control thetemperature response of DMFC online.On the basis of the above research work, a joint power supply structure is proposed formulti-stack DMFC. The control system is designed with the AT89S52microcontroller.Real-time monitoring of the stacks’ working status parameters can be achieved. The controlsystem performs effective management on the entire multi-stack DMFC through themulti-stack coordination algorithm. The impact of stack attenuation could be avoided, whichimproves the stability of power supply.