基于一种改进自适应模糊神经技术的PEMFC系统建模和控制

从质子交换膜燃料电池(PEMFC)实际应用的角度出发，应用自适应模糊神经网络技术对PEMFC系统进行建模与控制．在建模过程中，同时应用实验数据和专家经验对模型进行辨识，使模糊节点具有明确的物理意义和初始参数的选择更加容易．在控制过程中，将训练好的网络模型作为PEMFC控制系统的参考模型，采用自适应神经网络学习算法(ANA)在线对控制器参数进行自适应调整，采用最近邻聚类算法(NCA)对控制器的模糊规则库进行更新．在仿真实验中，将自适应模糊控制算法与PID和传统模糊算法进行比较，结果表明本算法控制性能优良．

Modeling and Novel Adaptive Fuzzy Neural Network Control of Proton Exchange Membrane Fuel Cell (PEMFC)

From practical application, adaptive fuzzy identification and control models of proton exchange membrane fuel cell (PEMFC) were developed based on input-output sampled data and experts' experience. In the modeling process, experimental data and experts' experiences are used to identify the operating temperature of PEMFC. It makes the nodes of network possess distinct physical meanings, and chose initial value easily. In the control process, the trained network model is used as the reference model of PEMFC control system. ANA is applied to regulate parameters on-line, and NCA is applied to update the rule database of controller. At the end, the simulation and experimental results of PEMFC control system were presented, with the show of the effectiveness.