Research On Fault Diagnosis Method Of Photovoltaic Array

In the future,the energy and power structure of the world will inevitably become a high proportion of renewable energy structure.In the high proportion of renewable energy structure,the proportion of solar energy will reach a higher level.The continuous development of photovoltaic power generation technology has made the widespread use of solar energy an inevitable trend.However,a large number of photovoltaic power plant arrays are built outdoors in harsh environments,various faults will inevitably occur during operation.Therefore,appropriate fault diagnosis methods can be used to accurately and quickly identify the faults of photovoltaic arrays and timely Alarms have become an important means to ensure the safe and stable operation of photovoltaic power generation systems.This paper focuses on the four common faults of photovoltaic arrays: short-circuit,open-circuit,local shadowing and aging faults,and proposes a fault diagnosis method for photovoltaic arrays based on genetic algorithm to optimize BP neural network.First,this paper introduces the main classification of photovoltaic power generation systems and the working principles of photovoltaic cells.A simulation model of photovoltaic cells and photovoltaic arrays is established in Matlab/Simulink,and the output characteristics under standard conditions are derived.After that,the four fault states of the photovoltaic array are analyzed in detail,and the photovoltaic array models corresponding to these four faults are established respectively,and then the array output characteristic curves under various fault states are obtained.The output parameter corresponding to the obvious change of the fault is used as the input variable of the fault diagnosis model.Secondly,the structure and working principle of BP neural network and genetic algorithm are studied in detail.Aiming at the problem that the BP neural network algorithm is liable to fall into a local minimum during network training,this paper uses genetic algorithms to optimize the weight and threshold of the BP neural network.Finally,based on the characteristics of BP neural network and genetic algorithm,this paper designs the BP neural network part and genetic algorithm part of the fault diagnosis model respectively,and obtains a complete photovoltaic array fault diagnosis model based on genetic algorithm to optimize the BP neural network.Select multiple sets of data to train the model,get the network training error curve,and test the trained fault diagnosis model to get the corresponding fault diagnosis results.By comparing with the network training error curve and test results of the traditional BP neural network,it can be found that the convergence speed of the BP neural network optimized by the genetic algorithm is improved,and the four types of faults studied in this paper are more accurate.Therefore,it is verified that the photovoltaic array fault diagnosis method based on genetic algorithm optimized BP neural network adopted in this paper can improve the diagnosis effect of photovoltaic array fault and improve the reliability of photovoltaic array fault diagnosis.