Research On Fault Diagnosis And Control System Of Fuel Cell Backup Power

With the continuous growth and development of the civilization,there were some solid reasons to expect that the demand for low-pollution and high-efficiency energy to become more important.PEMFC(Proton exchange membrane fuel cell,PEMFC)which use hydrogen as an energy source can provide reliable power in a steady state.PEMFC showed great potential in the field of backup power.First,according to the theory of air-cooled PEMFC,the mathematical model was built.The model could be divided into several parts: the anode gas diffusion model,the anode gas consumption model,the cathode gas diffusion model,the cathode gas consumption model,the output voltage model,the PEMFC temperature model and the anode gas pressure drop model.Then,these models were simulated in MATLAB.A 2.5kW air-cooled PEMFC stack was operated in testing platform.The experimental data was used to validate the steady and dynamic performance of this model.The results showed that this model can simulate the state of PEMFC stack well.Then,in this paper,the faults of the PEMFC system was summarized as follows: flooding in anode of PEMFC,the raise of the hydrogen penetration,the fault in cooling subsystem and the fault in hydrogen supply subsystem.By analyzing the model and experiment of fuel cell system,the mechanisms and symptoms of these four kinds of faults were studied.An method of neural network ensemble was presented to diagnose these faults.The test result showed that,compared with the previous method for fault diagnosis of PEMFC systems,the proposed fault diagnosis method has higher accuracy and generalization ability.Moreover,the partial structure of this method can be altered easily,along with the change of the PEMFC systems.In general,this method for diagnosis of PEMFC has value for certain applications.Thirdly,in this paper,the methods of fuzzy PID control and RBF neural network intelligent control were presented for the control of the PEMFC stack temperature and hydrogen flow.These two methods had the ability about self-adaptation.Then,Hydrogen flow and stack temperature were controlled using these two methods and traditional PID control scheme.The testing results drew the conclusions that were in hydrogen flow control loop,fuzzy PID control worked best,and in stack temperature control loop,RBF neural network intelligent control was optimal.At last,this paper studied the various state in the fuel cell backup power systems.A program logic control method was proposed to optimize and improve the overall performance of this system.Base on the research on system control and fault diagnosis,a control module of fuel cell backup power was designed using Siemens S7-1200,and a remote monitoring system aslo was designed with LABVIEW.This monitoring system could simultaneously monitor multiple backup power sites.An fuel cell backup power prototype was developed,and then a series of tests were made to testing its overall performance.The testing results showed that the fuel cell backup power system could provide safe and reliable power to the communication base stations.