Arc Recognition Method For Series Fault Based On Morphological Filtering And OTSU

In this paper,series arc fault experimental device is used to conduct a series of arc fault experiments,and the performance of the device is verified.By analyzing the experimental data,morphological filtering and OTSU are combined to identify the series fault arc.The series fault arc experimental device designed in this paper can be used to study the characteristics of fault arc under AC and DC conditions.By switching different resistors and inductors to form different power factor loads,the device also has an external load interface,which can connect the external load to the device for fault arc experiment.At the same time,the device can change the external environment of the arc to carry out fault arc experiments under different conditions,and control the arc gap to study the influence of the arc gap on the characteristics of the fault arc;adjust the humidity of the fault arc environment to study the influence of humidity on the fault arc;simulate the vibration of the contact separation state to study the characteristics of the fault arc.The initial temperature of contact is controlled to study the effect of contact temperature on arcing.The performance of the series fault arc device is verified by experiments under different loads and different environments.The characteristics of series fault arc are different due to different types of power supply and load.In order to obtain the general conclusion of fault arc characteristics,this paper presents a method of identifying series fault arc based on morphological filtering and OTSU.Fault arc experiments are carried out for different types of pure resistive load,inductive load,frequency converter-motor loads,industrial computer loads,etc.The current signals collected in normal working state and arc fault state are decomposed into five layers by using DB4 wavelet basis.The decomposed waveforms of current signals in 32 frequency bands are obtained as the identification characteristics of fault arc.By calculating the variance of each frequency band at the same time,the decomposed frequency band signal is reconstructed into a new signal.Morphological algorithm is used to filter the reconstructed signal,which highlights the current characteristics in the case of fault.The waveform threshold th is extracted by OTSU algorithm,and the number of waveform peaks above th after filtering is counted.The results show that there are obvious differences in the number of wave peaks above the waveform threshold th between the normal state and the series fault arc state,which can be used as the fault arc identification feature.