An Arc Fault Detection Method Based on Wavelet Feature Extraction and The Design & Realization by LabWindows/CVI

Arc fault is one of the primary reasons that cause electrical fire. When arc fault occurs in power supply line, the current can not make protective equipments act and the arc fault can not be found and cut off easily, so electrical fire comes into being. In the comparisons, the current signal is used for the detection physical parameter and the wavelet transform is used to study the arc fault detection. First according to the select principle of wavelet singularity detection, select and construct orthogonal quadratic spline wavelet as wavelet function, and use the porous algorithm dyadic wavelet transform to realize wavelet transform fast algorithm. Then carried out the arc fault’s wavelet singularity detection through modulus maxima detection method, finally analyzed the current signal by the method of wavelet approximation. LabWindows/CVI is used as development platform to design and implement the above analysis. Based on LabWindows/CVI, the upper computer program uses multithreading technology to detect and analyze the current signal. By arc fault detection algorithm, the system judges whether there has arc fault or not. The algorithm judges whether there has arc fault or not through detect if there has periodicity singularity points or not. The experiments show that this detection method of arc fault can detect arc fault in power supply circuits exactly and efficaciously.