Fault detection method of DC distribution system based on iewt and imdmf

Abstract

The present invention relates to a fault detection method for a DC power distribution system based on IEWT and IMDMF, comprising the following steps: acquiring data on DC power distribution line faults and AC system faults, and constructing a combination of improved multi-view depth matrix decomposition and soft distribution layer Fault classification model; protection devices are installed at both ends of the line of the DC power distribution system to monitor and collect the current and voltage waveforms of the DC line in real time, and obtain the inter-pole voltage u dc ;Use the improved empirical wavelet transform to adaptively decompose the current waveform to obtain the current component c 1~3 ; Calculate the maximum value of the detail coefficient A max , and set the threshold A th , when satisfied A max > A th When , it is judged that a fault occurs; the current component that will be obtained c 1~3 , inter-pole voltage u dc Input the fault classification model to realize the classification of fault types of the DC power distribution system. The invention can effectively shorten the fault classification time and improve the operation efficiency of the classification model.

Description

technical field [0001] The invention relates to the field of power distribution system fault detection, in particular to a direct current power distribution system fault detection method based on IEWT and IMDMF. Background technique [0002] With the rapid development of new energy and power electronics technology, the demand for grid-connected distributed power sources increases, and the proportion of DC loads continues to increase. The traditional AC distribution network needs to realize the consumption of distributed power and the power supply of DC loads through power electronic converters. The use of DC distribution network can not only directly supply power to various DC loads, but also save distributed power and energy storage devices. The power electronic conversion device when connected to the grid has the advantages of high transmission efficiency, good power quality, and high power supply reliability. However, the immaturity of protection technology limits the wi...

AI Technical Summary

Problems solved by technology

Signal processing algorithms include short-time Fourier transform, discrete Fourier transform, wavelet transform, and improved complementary set empirical mode decomposition algorithm. It is better to directly use electrical quantity detection, but each signal processing algorithm has some defects in the application of DC power distribution system fault detection field.

Images