Research On Detection Technology Of Electric Fire Arc In Low Voltage Distribution System

With the continuous development of China’s electric power technology,the increasing variety of electrification products,residential electricity and industrial demand for electricity increased,the electric arc fault caused by electrical fire bring the huge loss to the people.Arc fault occurs in the line will be accompanied with light,heat and other phenomena occur,a very small arc current will lead to local high temperature in the circuit,it is easy to cause burning around the fuel,resulting in electrical fires.Due to research technology,signal processing methods and the limitations of the study,traditional fire alarm or protection device can not play a comprehensive circuit protection,often false positives,omission and other phenomena.So it is very important to detect the fault of low voltage arc timely and accurately.The hazards of electrical fire and the relationship between electrical fire and arc fault are analyzed.The mechanism of AC arc generation,the classification of arc and the characteristics of AC arc are expounded from microscopic angle.According to the theory of three kinds of arc mathematical models,three kinds of arc model deduction process are analyzed.Based on the assumptions of the mathematical model and the scope of applicability,the simulation arc model suitable for this subject is selected.A simulation platform is built in MATLAB,the load is simulated and analyzed,a typical load simulation database is established,and the basic characteristics of arc current are analyzed.An experiment platform to simulate arc fault was set up under standard UL1699.Several typical load experiments were carried out,and the characteristics of the current waveform before and after the arc were compared and analyzed in time domain.According to experimental data collected from before and after of the arcing,the experimental database was established.Simulation and experimental results are compared,to ensure the accuracy of the experiment and simulation.Finally,due to the limitation of time domain,arc fault occurring in the circuit prone to false positives and omission phenomenon.Therefore,in order to further study the characteristics of arc fault and achieve the correct diagnosis of arc fault identification,presents two methods of arc fault detection.one is based on the power spectrum relative area difference,by analyzing the characteristics of the power spectrum before and after of the arcing,the power spectrum curve fitting and calculating the relative area difference,to achieve the arc fault diagnosis;The other one is based on the wavelet coefficient modulus maximum method,which can detect the arc fault accurately and timely by the periodic realization of the singularity of the wavelet signal.According to the results of simulation and experiment,it is known that the above two methods can detect the arc fault accurately and timely.reduce the loss of electrical fire bring arc fault and provide a new research method for the development of new type arc fault detection device and arc fault.