Research And Application Of Photovoltaic Array Fault Diagnosis Technology

With the ever-increasing energy crisis,countries around the world are paying more and more attention to clean renewable energy.Among the renewable energy,the photovoltaic industry that uses solar power to generate power has developed rapidly.As the core component of photovoltaic power station,photovoltaic array works in harsh outdoor environment.Failures will affect power generation efficiency and power generation efficiency.Therefore,the study of fault diagnosis technology for photovoltaic arrays is of great significance for ensuring the stability of photovoltaic power generation.The research work of this article mainly includes two aspects: the research of photovoltaic array fault diagnosis methods and the software and hardware design of improved communication nodes.For the research of photovoltaic array fault diagnosis methods,two fault diagnosis algorithms are proposed: genetic algorithm(GA)optimized back propagation(BP)neural network fault diagnosis algorithm(GA-BP),Improve butterfly optimization algorithm(IBOA)optimized the probabilistic neural network(PNN)fault diagnosis algorithm(IBOA-PNN).(1)Establish a simulation model of the photovoltaic array,analyze the output characteristic curves of the photovoltaic array under five conditions,and determine the input vector that reflects the fault characteristics under different conditions;(2)Introduce the genetic algorithm with good global search performance to optimize the weights and thresholds,improve the search speed and convergence accuracy of the BP neural network,and obtain the GA-BP fault diagnosis model for judging the state of the photovoltaic array;(3)Butterfly optimization algorithm(BOA)has problems such as slow search speed and easy to fall into local optimum.Chaos mapping is introduced to IBOA,and an IBOA-PNN fault diagnosis model is proposed.Aiming at the software and hardware design of the improved communication node,the following work is done:(1)The GA-BP algorithm is transplanted from the PC to the embedded platform to realize the fault diagnosis function,which reduces the consumption of communication traffic and reduces the burden on the server;(2)With the STM32F407VET6 microprocessor as the core,the core control module,communication module,environmental information collection and other hardware modules are developed;the software design of the improved communication node includes system initialization,data reception and collection,and data processing and uploading.By building an actual photovoltaic array experimental platform,obtaining actual operating data samples,verifying the classification performance of the GA-BP model and the IBOA-PNN model for photovoltaic array fault diagnosis,and verifying the effectiveness of the improved communication node.