Research On AFCI Technology In Power Cable Network

With the rapid development of China's economy after the Nineteenth National Congress and the construction of power cable network,more and more cities and countryside begin to update distribution boxes.More and more power cables are used in distribution network,and the use of power cables in distribution network tends to be complex,resulting in the occurrence of defects in the bolt parts of power cable network rising year by year.The full name of AFCI is Arc-Fault Circuit-Interrupter,which is translated into Arc-Fault Circuit-Interrupter in Chinese.AFCI is a circuit breaker for protecting electrical equipment.Its principle is to cut off the power supply when the fault develops into a greater disaster according to the fault signal in the conductor.Distribution cable bolts are in high temperature and vibration state for a long time,and will loosen.The loosening fault of distribution cable bolt joint may lead to intermittent and continuous arc generation,which is very easy to cause overheating of distribution box.Once there are combustibles around,it will easily cause fire and other faults,and even cause explosion when the situation is serious.Life.Therefore,it is of great practical significance to study AFCI technology in power cable network and put forward methods of diagnosis and judgment.In this paper,the research of AFCI arc fault diagnosis and identification technology in power cable network is carried out,and the arc characteristics of power cable caused by the loosening of bolt position in power cable network are studied.In the course of the research,an arc generator is set up to identify and diagnose arc faults.Under the experimental conditions of adjusting the degree of looseness,loop power factor and loop current,many scientific experiments are carried out under different experimental conditions at the bolted part,and the temperature and electrical characteristics are verified by the data acquisition card.The experimental data are collected to serve as the basis for arc fault diagnosis and identification of the bolted parts in the power cable network.In order to distinguish the characteristics of current between different loosening States and normal states,and the differences between different states,this paper uses the method of combining wavelet energy entropy and probabilistic neural network to analyze the time-frequency domain of contact current at fault points.When arc fault occurs in the circuit,it can accurately diagnose the situation of different loosening degrees at the connecting parts of power cables.The experimental data show that the characteristic data can be quickly calculated from the current waveform by using the wavelet energy entropy when different loosening states occur in the connecting parts of power cables.The probabilistic neural network takes this characteristic as the input of the neural network,and can effectively identify the real situation of arc fault in power cable network.The theoretical accuracy is above 95%.In order to simulate the occurrence of arc fault,the probabilistic neural network takes this characteristic as the input of the neural network.The test results show that the recognition accuracy of the above methods is higher than 90% under different conditions.The comparison with the accuracy of AFCI products in the market proves that this method can be used to identify and judge the loosening faults of power cable bolts in distribution network.This paper has 38 figures,17 tables and 50 references.