Based On The Five Time Harmonic Component Probability Distribution Of Arc Fault Identification Method Of Research

Since the 19 th century the invention of electricity,electricity has received a wide range of applications.In today's life,home appliances have been a lot of popularity,and there are many new types of household appliances continue to emerge,the frequency of occurrence of electrical fires are getting higher and higher,causing more and more harm.The fault arc is the main cause of the fire.The traditional protection of electrical equipment can be short-circuit fault and ground short-circuit fault when the effective protection is for high current overcurrent protection.The series fault arc is characterized by small current,high temperature,short duration,when the fault arc,the traditional over-current protection and can not effectively protect the line.Therefore,it is of great significance to effectively detect and identify the series fault arc.At present,the research on low voltage fault arc is mainly in the thesis stage,almost no product appears,and foreign research on fault arc is more mature than domestic,AFCI has been related to foreign products and has been applied.The main research methods of arc fault in China is the main characteristics of arc fault current and voltage waveform analysis,wavelet analysis,Fourier analysis,autoregressive model and independent component analysis method in the time domain and frequency domain analysis of arc fault.However,due to the diversity of building electrical characteristics of the load,some load normal operation of the current characteristics and fault arc current characteristics are very similar.Therefore,based on the research results at home and abroad,this thesis builds a fault arc experimental platform,and collects a variety of current waveforms when the load is in normal operation and fault arc.Through the analysis of the collected load current waveform,observe the difference between the load current and the current waveform when the fault current occurs.Then the Fourier analysis of the current waveform is carried out by using MATLAB software to extract the harmonic components of the current waveform during the normal operation and the fault arc.The harmonic components of the two cases are compared and analyzed,determine the quintuple harmonic component as the characteristic parameter of arc fault occurs.Then,50 sets of data were collected for each feature of the load,and the frequency distribution histogram was used to observe the probability distribution type of the feature quantity similar to the t-distribution,normal distribution and logarithmic normal distribution.Then,the t-distribution test of single parameter and the k-s test of single parameter are carried out by SPSS software respectively,and the distribution of the characteristic quantity was determined to be lognormal distribution.And then determine the threshold range of the feature quantity of the fault arc criterion.In this thesis,a fault arc identification method based on probability distribution of quintuple harmonic component is proposed,which provides a reliable reference for fault arc detection.