Research On Time-frequency Characteristic Analysis And Identification Method Of Series Fault Arc In Photovoltaic System

Due to the fact that solar energy is widely distributed across the globe,and solar power does not have problems such as fuel consumption and environmental pollution,it has rapidly developed under the background of global energy depletion and has increasingly replaced traditional energy generation.However,with the increase in the use of photovoltaic equipment,some serious problems are gradually exposed.Photovoltaic fire accidents have been reported in recent years,and most of the fire accidents have been caused by arc faults.The domestic and foreign direct current fault arc protection for photovoltaic system has successively issued a series of standard standards,and more and more scholars have passed into the study of photovoltaic fault arc.Taking into account that photovoltaic power is composed of a large number of battery plates in series and in parallel,and there are a large number of connection points inside,the probability of series arc faults is much greater than that of parallel arcs.Therefore,this paper mainly focuses on the research of DC series arc fault arc of photovoltaic systems.Based on previous researches on the fault arc of photovoltaic system,a series-connected arc fault experimental platform for photovoltaic system was set up,and the fault arc characteristics were studied based on the system current signal.First,the generation mechanism and mathematical model of the arc are introduced.Then the characteristics of the arc are analyzed in two aspects of the time-frequency domain.It is found that in the time domain,the average arc current and the peak-peak value are significantly different from the normal current,while the wavelet b5-band energy in the frequency domain can better reflect the characteristics of the arc and distinguish the environmental factors such as shading and shadowing.Based on the design of the fault arc detection algorithm.Finally,the arc fault is tested by using the time-frequency domain feature combined with the use of a threshold to determine the arc fault twice.The test results demonstrate the effectiveness and feasibility of the method.