Research And Application Of Aviation AC Arc Fault Detection Technology

With the development of aviation industry,aviation safety has always been a very important issue.Reducing the fire safety risk of aviation electrical power system can effectively improve the reliability of aircraft.And arc fault is one of the most important reasons that cause fire in aviation electrical power system.In recent years,with the increasing capacity of aircraft power distribution system,due to the increase of the probability of system fault caused by arc fault,the aviation safety problem has become particularly severe.Therefore,it is of great significance to study the characteristics of arc fault and find an effective method for detecting aviation AC arc fault for the aviation safety.Firstly,In this paper,according to the relevant standards,the series arc and parallel arc in the aviation electrical power system are simulated by the established arc fault simulation test platform,and the related current data are collected.The load types include linear loads and nonlinear loads.Each kind of test collects the data of different current values.This paper summarizes the arc characteristics with the change of the current values and compares the similarities and differences of the arc current waveform of the different load types.Secondly,two kinds of arc fault characteristic features,characteristic feature based on fractal dimension and wavelet analysis and characteristic feature based on singularity of the wavelet reconstruction signal,are proposed and two aviation arc fault detection methods are proposed according to each characteristic feature.It is found that the two arc fault detection methods have certain arc fault identification capability,but they both can not effectively identify all arc faults.Finally,considering arc fault detection methods based on the single feature,the accuracy of arc identification is low,therefore the BP neural network is used to determine whether there is an arc fault in the line.The first step is to train part of feature data through the neural network,then using the untrained feature data tests the correctness of this neural network.the result shows that the detection accuracy is high and this method can be used as one of the aviation AC arc fault detection methods.This paper also propose the hardwareand software design scheme of aviation arc fault detector,which lays good foundations for the development and research of aviation arc fault detector.