Multi-source discharge signal separation method and analysis and discrimination method

Abstract

The invention discloses a multi-source discharge signal separation method and an analysis and discrimination method, and the multi-source discharge signal separation method comprises the following steps: positioning a corresponding discharge position L of a multi-source discharge signal in a detected cable, and calculating the equivalent duration T and the equivalent bandwidth W of the multi-source discharge signal; separating the multi-source discharge signal according to the discharge position L, the equivalent duration T and the equivalent bandwidth W. Through the LTW three-dimensional separation space, different discharge signals with time-frequency distribution characteristics can be distinguished, and discharge sources with similar time-frequency characteristics but different discharge positions can be further distinguished.

Description

technical field [0001] The invention relates to the technical field of communication, in particular to a multi-source discharge signal separation method and an analysis and discrimination method. Background technique [0002] High-voltage cables are the main way of power transmission, and the normal operation of the cables is the basis of the power supply system. During the working process of the cable, different partial discharge phenomena will cause different degrees of damage to the cable, affect the transmission and use of electric energy, and cause economic losses; therefore, partial discharge detection is an important indicator to measure the insulation performance of power cables. [0003] However, in the actual partial discharge detection process, due to the complexity of the site environment, a variety of discharge sources are mixed or the partial discharge detection equipment will inevitably be affected by various interference discharges, so the analysis and iden...

AI Technical Summary

Problems solved by technology

During the working process of the cable, different partial discharge phenomena will cause different degrees of damage to the cable, affect the transmission and use of electric energy, and cause economic losses; therefore, partial discharge detection is an important indicator to measure the insulation performance of power cables

[0003] However, in the actual partial discharge detection process, due to the complexity of the site environment, a variety of discharge sources are mixed or the partial discharge detection equipment will inevitably be affected by various interference discharges, so the analysis and identification of the collected signals is the key to partial discharge detection. Where the key points and difficulties are, it is necessary to seek an effective multi-source separation method

[0004] At present, the separation of multi-source discharge signals is mainly carried out through parameters or methods such as pulse waveform characteristics and time-frequency transformation; the former constructs the characteristic attributes of pulse waveforms at the far end and near end according to the pulse propagation attenuation characteristics, including pulse width, rise Multiple parameters such as / fall time, oscillation polarity, and main oscillation frequency are compared during the pulse signal acquisition. However, this method does not consider the variability of the cause of the discharge, and it is impossible to further identify the type of discharge. It can only be based on According to several kinds of discharge pulse waveforms in experience, the judgment is rough and arbitrary; the latter uses joint time-frequency analysis to map the pulse data into the standard density distribution characteristics of time domain and frequency domain. From a statistical point of view, it is indeed possible to distinguish between different type of pulse wave, but in the actual cable partial discharge detection, there will still be situations where the pulses are very similar but the phase distribution characteristics are still mixed, making it difficult to further effectively separate, so it is necessary to find a method that can solve such problems

Images