Arcfault Research And Application Based On HHT And Neural Network

With the development of national economy and modern science and technology, civil and industrial demand for electricity is increasing, according to the"2012 China Fire Yearbook" statistics, up to 37,960 electrical fire caused in 2011, accounting for 30.3% of the total number of fire. It reported over 25,000 in Europe each year, accounting for over 25% of the total number of fires. Electrical fire has become one of the fire safety hazards. For the series arc fault caused the majority of electrical fires in all the electrical fire, the circuit breaker and residual current protection device used in the line are unable to effectively isolate and protect. It is necessary to R&D of the arc fault detection technology to improve the electrical safety. And the detection technique for domestic 220V/50Hz electrical environment has important theoretical and practical value.In this thesis, the 220V/50Hz electrical environment arc fault detection is taken as research object to find the arc fault test method. And Hilbert-Huang Transform Theory is applied to the arc fault feature extraction. The main research covers:(1) According to UL1699-2008 FCI standard, based on the arc fault test platform, different load arc fault test has been performed, and arc fault current signal has been obtained for further studing the arc fault feature.(2) Based on the relevant theory of the arc fault detection, the application characteristic of the Hilbert-Huang Transform (HHT) method and the wavelet transform method has been analyzed.(3) Due to the unique characteristics of the Hilbert-Huang Transform (HHT) method, it is widely used in various fields of signal processing. In this thesis, Hilbert-Huang transform and algorithms have been studied and used to the application of the arc fault feature extraction. And based on the above analysis, the neural network is applied for training the feature extraction Hilbert spectrum. The results show that this proposed method has fine recognition results for the arc fault detection.