基于神经网络的固体氧化物燃料电池电堆建模

针对现有的固体氧化物燃料电池(SOFC)模型过于复杂,难以满足工程上对SOFC系统实时控制设计的需要,提出了利用遗传算法(GA)优化径向基函数(RBF)神经网络实现对SOFC电堆建模。在建模过程中,利用遗传算法优化RBF神经网络的输出权值及高斯基函数的中心向量和基宽向量,采用优化后的参数作为网络初始值,然后利用梯度下降法对各参数进行调整。通过仿真对该建模的有效性和建模精度进行了检验。

Modeling SOFC Stack via Neural Networks

The solid oxide fuel cell(SOFC)models are too complicated to be used for on-line controller design,therefore a SOFC stack model was set up using a radial basis function(RBF)neural network based on a genetic algorithm(GA).During the process of modeling,the GA aiming to optimize the parameters of RBF neural networks and the optimum values were regarded as the initial values of the RBF neural network parameters.Furthermore the gradient descent learning algorithm was utilized to adjust the parameters.The validity and accuracy of modeling were tested by simulation.