PEMFC阴极系统的混合智能PID解耦控制器设计

Hybrid intelligent PID decoupling control design for PEMFC cathode system

60 kW PEM FC发电系统要求安全性、可靠性和持久性,理想的电堆湿度和压力控制很重要。在电堆阴极系统的动态仿真和分析基础上,针对系统强非线性动态耦合的特点和控制要求,设计了阴极湿度和压力的混合智能PID解耦控制器。系统实时控制的仿真实验结果说明,利用带二次型性能指标的单神经元PI控制器,调节空气增湿器水泵的注入水流量来控制阴极湿度,控制结果快速而且稳定。以对角回归神经网络(diagonalrecurrentneuralnetw ork,DRNN)动态辨识为基础的增量式PID控制器,调节空压机送往电堆阴极的空气流量,使阴极压力的控制更加稳定和精确。两个控制器的良好配合不仅能够克服系统的非线性动态和耦合特性,而且能够克服系统负载的实时功率需求的影响。

For the excellent security,stability and durability of 60 kW proton exchange membrane fuel cell（PEMFC） power generator,it＇s important for cathode humidity and pressure of the fuel cell stack to be maintained in certain ideal levels.Based on the simulation and analysis of the cathode system,a hybrid intelligent PID decoupling controller was proposed for the close coupling and non-linear dynamic characteristics and requirements of the system.The simulation control results show that,the cathode humidity can be fast controlled stable by adjusting the water injection for the air supply,which is realized by a single neuron PI controller with the optimal quadratic index.Another incremental PID controller,based on the diagonal recurrent neural network（DRNN） dynamic identification,controls the cathode pressure more accurate by adjusting the air flow exit the compressor.Coordinating these two controllers can protect the stable operation of system from non-linear dynamic and coupling characteristics and the real-time power demand of the system load.