SOFC System Fault Diagnosis Method Based on Maximum Mutual Information Coefficient and Particle Swarm Optimization Support Vector Machine

Solid oxide fuel cell (SOFC) will degrade the performance of the battery stack due to fuel deficit failure, which will affect the performance of the whole system. As a result, there has been a growing focus on fault diagnosis technology to increase the stability and dependability of SOFC systems. This work proposes and introduces an efficient fault diagnostic model (FDM) for SOFC system failure diagnosis. In SOFC system, a large number of sample data are collected, the maximum information coefficient (MIC) method is used to reduce the dimension of the feature quantity and extract the feature quantity. Particle swarm optimization (PSO) is used to optimize the parameters of support vector machine (SVM), and the FDM of SOFC system based on MIC algorithm and PSO-SVM is established. To improve the accuracy of SOFC system fault diagnosis. The findings demonstrate the excellent accuracy and dependability of the FDM of the SOFC system, which is based on MIC and PSO-SVM. The accuracy of the training and test sets can both exceed 98%.