固体氧化物燃料电池的建模与仿真

摘要

能源短缺和环境问题已成为本世纪全球面临的最重要课题，作为一种新的能源形式．固体氧化物燃料电池（SOFC）技术日益受到重视。由于现有的SOFC模型过于复杂，难以满足工程上对SOFC系统实时控制的需求，提出利用粒子群算法（Ps0）优化径向基函数（RBF）神经网络，从而实现对SOFC的建模。PSO对RBF神经网络的中心值和连接权值进行优化，提高了网络的泛化性能，使其非线性逼近能力更强，从而达到精确模型的目的。仿真实验验证了粒子群算法在SOFC建模的有效性。%With energy shortages and environmental issues become the most important issue in the world in this century, as a new form of energy, the solid oxide fuel cell (SOFC) technology has received increasing attention. SOFC models are too complicated to be used for on-line controller design, therefore, a SOFC model was set up using a radial basis function (RBF) neural network based on a particle swarm optimization (PS0). The PSO optimizes the centers and widths of RBF, so that the network's generalization performance is improved and has stronger nonlinear approximation ability, at the same time, the model becomes more accurate. Simulations show the validity of the PSO in SOFC modeling