Research On Intelligent Control Algorithm Of Solid Oxide Fuel Cell

Solid oxide fuel cell(SOFC)is a new type of clean energy,and its modeling and control research will have a great impact on the actual grid-connected power generation technology.In this paper,based on the SOFC single battery power generation principle and actual running status,under the condition of reasonable assumptions,combined with the ideal gas state equation and the conservation of mass,an adjustable parameter model of the SOFC single cell is established with two adjustable parameters,the utilization of hydrogen Q_f and the ratio of hydrogen to oxygen M.In addition,the PID control strategy and intelligent control method on SOFC single cell are studied.Firstly,a PID controller is designed to control the output voltage of SOFC single cell and trial and error method is used to realize the PID tuning.However,the trial and error method needs to change the values of K_P,K_I and K_D for many times based on the user's inherent experience and continuously analyze the variation of the output voltage of SOFC single cell.Besides,when the change of external conditions has an impact on the battery,it is necessary to reset the PID.This process wastes a lot of time and it is difficult to achieve a good control effect when the parameters are not properly matched.Therefore,the intelligent control algorithm on SOFC is studied in this paper.On the basis of determining the reasonable value range of parameters K_P,K_I and K_D,particle swarm optimization algorithm and genetic algorithm are designed to achieve the optimal tuning of PID parameters K_P,K_I and K_D,which improved the speed of parameter optimization and the output performance of SOFC.The simulation results show that the adjustable parameter model of SOFC single battery can accurately reflect the relationship among the hydrogen input molar flow,the water vapor input molar flow and the output voltage of SOFC single battery and has some flexibility.And the control scheme designed according to the input-output characteristics of SOFC single battery adjustable parameter model is reasonable.Under the condition of the ideal state and the interference state,both particle swarm PID and genetic algorithm PID can effectively control the hydrogen input molar flow in order to ensure that the output voltage of battery reaches the expected value.These two controllers can quickly adjust PID parameters K_P,K_I and K_D according to the actual running condition of SOFC single battery,and have better anti-interference performance.Moreover,by analyzing the optimization and control effects of particle swarm algorithm and genetic algorithm,it is found that particle swarm PID is better than genetic algorithm PID when controlling SOFC single battery.