Abstract：The short circuit fault has become the most common and harmful fault in the power system, and the signal filtering is of great significance to improve the accuracy of the early detection of short circuit fault. In view of the characteristics of short circuit fault in low voltage distribution system and the rapid requirements of early detection technology, this paper researched mathematical morphology, wavelet denoising, mean filtering and median filter, which are real-time and easy to implement. The actual noise signal produced of the typical air conditioning load with emergency power supply UPS was taked as an example. Analysising and comparising of four filtering method, the conclusion is that the generalized mathematical morphology method couldn’t remove the white noise, but it can filter out some of the impulse noise except that the noise of the pulse width is more than the width of the structure element. The improved generalized morphological filter is better for removing impulse noise, and it can be used to filter out the impulse noise which is wider than structure element. The dbN wavelet base function has almost no filtering effect on impulse noise. Median filtering could remove most of the noise, and mean filter is good for suppressing noise.
产焰萍, 缪希仁. 低压配电系统短路故障早期检测的信号滤波方法研究[J]. 电气技术, 2016, 17(8): 9-13.
Chan Yanping, Miao Xiren. Signal Filtering Method for Early Detection of Short Circuit Fault in Low Voltage Power Distribution System. Electrical Engineering, 2016, 17(8): 9-13.
2016, Vol. 17 
