Modeling And Simulation Study Of Proton Exchange Membrane Fuel Cell Power Generation System

In the 21 st century,people is facing great challenges as well as unprecedented opportunities.At the same time,with rapiding development of economy and technology,environmental pollution and ecological damage are also intensifying.The new energy is an alternative to our challenge.Fuel cell technology has become one of the most attractive schemes in many new energy schemes because of its advantages of clean,low noise and high productivity.The reaction mechanism of fuel cell is convert chemical energy into a power source for human using.The chemical reaction of the power generation material and the oxide gives out the electric energy.Meanwhile the product is water without pollution.The fuel cell was first applied to the field of Aeronautics and Astronautics,then gradually into the related industries such as vehicle mobile power supply,distributed generation.Because fuel cell has such a wide application prospect,fuel cell technology has been regarded as a national technology project in the United States,Japan,Germany,France,Britain and Italy.Our country also clearly put forward the development of fuel cell technology during the period of "Nine-Five" and "Ten-Five".At present,our research on Proton Exchange Membrane Fuel Cell(PEMFC)is mainly in two aspects: 1,PEMFC internal material and structure 2,PEMFC PEMFC stack operation mechanism mathematical model.There are few scientific achievements in the research of PEMFC system control model and algorithm.This paper is based on the two research topics of "proton exchange membrane fuel cell power system simulation" and "20kW vehicle proton exchange membrane fuel cell system".Establishing empirical model and mathematical model for PEMFC system.Through the simulation data shows that the influence of PEMFC system about internal temperature,current density,potential difference and the oxidant gas pressure on the output of numerical model.According to the control method of PEMFC system model,4 neural network algorithms are used to establish PEMFC single cell model.The error and step number of voltage characteristic curve of PEMFC single cell model under four kinds of neural network algorithm are compared.The main creative aspects of this paper are as follows:In this paper,a new feasible scheme is proposed for the modeling of PEMFC system:Using of 4 neural network algorithms to simplify the PEMFC system.The voltage characteristic curve of the PEMFC system is established by the training and learning of the experimental data.By comparing the running time,the simulation step and the error index of the four algorithms,a new way is provided for the optimization of the PEMFC system.