Optimization of Support Vector Machine for Photovoltaic Array Fault Diagnosis Based on Sparrow Search Algorithm

Photovoltaic power generation system because of the use of energy green environmental protection has been widely concerned by the world, to ensure that the system operation with high efficiency is the premise that the system can operate normally, photovoltaic power generation system is installed in a bad environment, photovoltaic arrays often have a variety of failures, and the number of photovoltaic arrays in the system is large, it is difficult to determine the type of failure and maintenance in time. It seriously affects the efficiency of photovoltaic power generation system, so it is very important to study a method that can detect the failure of photovoltaic power generation system in real time. In this study, the output characteristic curves of various fault conditions were analyzed according to the output characteristic curves of photovoltaic cells, the characteristics that could be distinguished from other fault types were selected, and the fault characteristic values were collected. Then SSA-SVM is used to establish the fault diagnosis model of photovoltaic array, and combined with the simulation experimental data and the data of photovoltaic simulator, the validity of SSA-SVM method is verified. Through model design and algorithm improvement, SSA-SVM is applied to fault diagnosis of photovoltaic array. Compared with IGS-SVM and PSO-SVM, SSA-SVM algorithm has higher stability and shorter time.