A Model-based Study On Gas Leakage Fault Of SOFC

Since the new period,with the rapid development of scientific and technological civilization and the renewal of computer technology,human living standards began to improve day by day,the use of automobiles increased exponentially,the heavy use of automobiles brought serious air pollution.At the same time,the use of energy has also made rapid progress,human beings continue to increase the use of traditional energy,but neglect the protection of the environment.This requires the study of a new clean energy source to alleviate the plight of traditional energy sources,so research and development of mature fuel cell technology has become the demand of the times.Solid oxide fuel cell(SOFC)has attracted the attention of the scientific community because of its high efficiency and pollution-free properties,but life and reliability are the two key issues restricting its successful commercialization.Fault diagnosis is an important means to ensure the stable operation of SOFC system,extend its life and improve its reliability.Compared with SOFC’s progress in materials,reaction technology and so on,its fault diagnosis research is relatively small.Gas supply system is an important part of fuel cell,the leakage of gas supply system affects the normal operation of fuel cell,and there is a lack of research on SOFC leakage failure at home and abroad.Therefore,this paper will be based on the model of SOFC gas leakage failure research,the specific content of the study is as follows:1.According to the research results of domestic and foreign scholars on SOFC leakage failure,the mechanical model of SOFC hydrogen and oxygen leakage failure is explored.The output characteristics(voltage,power)of SOFC in the event of gas leakage fault are studied in view of the undersupply of hydrogen and oxygen.The temperature of the battery stack is the key factor affecting SOFC output voltage,power,gas channel pressure,gas reaction rate,etc.,and most of the troubleshooting models in the existing literature assume constant temperature,so this paper focuses on the effects of leak faults on reactor temperature.so that the so FC gas leakage fault diagnosis model is more reasonable and more practical value.2,Based on the SOFC gas leakage fault model,this paper respectively on the hydrogen leakage failure,oxygen leakage failure and hydrogen oxygen leakage failure when the performance of SOFC is simulated and analyzed.The results show that:(1)the above three gas leakage faults have a certain impact on the output voltage,output power and temperature of SOFC,(2)when the leakage rate of the above three gas leakage faults reaches 15% of the normal value,the operating temperature of SOFC will reach the upper limit of its safe operating temperature,and effective measures need to be taken to ensure the safe operation of SOFC.In addition,the correlation between hydrogen and oxygen to SOFC operation is explored by gray correlation analysis.Taking the output characteristics of SOFC when hydrogen and oxygen leak at the same time as the reference sequence,and evaluating the output characteristics of hydrogen leakage and oxygen leakage respectively,the simulation results show that the correlation of hydrogen leakage is close to 1,which shows that compared with the air supply system,hydrogen supply system has a more significant impact on the normal operation of SOFC.3,In order to ensure the safe operation of SOFC,it is necessary to take effective measures to avoid obstacles for different gas leakage faults.To this end,this paper classifys the leakage fault of SOFC.Unlike other classification functions with polynthics,Bayesian discrimination takes the probability of characteristic vectors into a certain category as the main basis for classification,it has no strict requirements for the classification object,for this reason,the first use of Bayesian discrimination method to classify the three gas leakage faults of SOFC,the study shows that the fault classification error rate based on Bayesian discrimination is 83.76%.In order to further improve the classification effect,this paper classifies three gas leakage faults based on Fisher’s discrimination method.The method can classify a variety of data,the principle of which is to use the projection method to reduce the dimensional processing of fault data,and then realize fault classification according to the criteria with the smallest data deviation in the group and the largest deviation of data between groups,and it is suitable for cases where there is less data and there is no obvious law of data.Compared with the so FC gas leakage fault classification results of Bayes’ s classification method,it is found that Fisher’s classification correctness rate is higher,94.62%,which indicates that Fisher’s classification method is more suitable for SOFC’s gas leakage fault classification study.4.In order to extend the life of SOFC and improve its reliability,it is necessary to predict its fault and take effective diagnostic measures.Kalman filtering can predict the value of the latest moment based on the state of the previous moment,and constantly adjust the deviation between the forecast value and the expected value,through a series of iterations to achieve the prediction purpose.Based on the Kalman filter prediction model,this paper predicts the fault voltage,fault power and fault temperature in the event of hydrogen leakage failure in SOFC,and the results show that the average variance between the prediction data and the actual fault data is large,and the prediction effect is not ideal.Grey prediction model can predict the system with unclear relationship between various factors,which can be applied to situations where there is little data and the data changes easily.This method achieves the desired effect by establishing the corresponding model of the different equations at the same time as the correlation analysis of each influence factor.In order to improve the prediction effect of SOFC gas leakage failure,the fault voltage,fault power and fault temperature in hydrogen leakage failure are further predicted by grey prediction model.The results show that compared with the Kalman filter prediction model,the gray prediction model has higher prediction accuracy and smaller average variance.In addition,the method can also be extended to the so FC fault prediction when oxygen leak,hydrogen and oxygen leak at the same time,which shows that the gray prediction model is an effective method to realize SOFC gas leakage fault prediction.