基于神经网络辨识的质子交换膜燃料电池建模

针对质子交换膜燃料电池(PEMFC)系统过于复杂，难以建模，而已建立的模型难以满足PEMFC控制系统设计和应用的要求。本文利用神经网络具有逼近任意复杂非线性函数的能力，将神经网络辨识方法应用到PEMFC强非线性系统的建模中，避开了PEMFC系统内部的复杂性。模型以电池工作温度为神经网络辨识模型的输入量，电池电压、电流密度为输出量，利用500组实验数据作为训练样本，采用改进型BP算法，建立了不同温度下电池电压—电流密度动态响应模型。仿真结果表明，方法可行，建立的模型精度较高，从而为设计PEMFC实时控制系统奠定了基础。

Modeling Proton Exchange Membrane Fuel Cell (PEMFC) Based on Neural Networks Identification

For the serious complexity of Proton Exchange Membrane Fuel cells (PEMFC), modeling of it is very difficult and the existing models are too complicated to be applied in designing and controlling of the system, especially in on-line controlling. In this paper, we try to establish the voltage and current model of PEMFC by using neural networks identification technique. The operating temperature of the fuel cells is taken as the input and the voltage and current response as the output of the neural networks model. In this way, we can avoid the internal complexity of PEMFC. The 500 groups experimental data are used, and the structure and the novel BP algorithm of neural networks identification system are given. The validity and accuracy of the model are proved by the simulation results. The neural networks modeling makes it possible to design on-line controller of PEMFC.