摘要
针对质子交换膜燃料电池(PEMFC)这一非线性复杂被控对象,采用基于C-均值模糊聚类的Takagi-Suge-no(T-S)模糊神经网络辨识方法,建立了系统的电特性模型;在此基础上应用广义预测控制策略,实现了PEMFC的输出功率控制。仿真实验比较了该方法与基于时间绝对偏差乘积积分(ITAE)指标的PID控制器和LQG控制器方法,结果表明所提出的方案在负荷跟踪、克服扰动及鲁棒性方面具有较理想的控制性能。
In this paper,a fuzzy neural identification method based on C-means is proposed for a proton exchange membrane fuel cell(PEMFC),and then a generalized predictive controller is designed to achieve the PEMFC load response.Simulation results show that the proposed method is characterized by better performance in disturbance rejection and set-point tracking when compared with an integrated time and absolute error(ITAE)-based proportional-integral-derivative(PID) and linear-quadratic-Gaussian(LQG) controller.
出处
《北京化工大学学报(自然科学版)》
CAS
CSCD
北大核心
2013年第3期104-108,共5页
Journal of Beijing University of Chemical Technology(Natural Science Edition)
基金
国家自然科学基金(61273132)
北京市优秀人才培养资助项目(2009D013000000003)
关键词
质子交换膜燃料电池
模糊神经网络
T-S模型
广义预测控制
proton exchange membrane fuel cell(PEMFC)
fuzzy neural network
Takagi-Sugeno model
generalized predictive control