CN-122017387-A - Dynamic power supply fault monitoring system, method, equipment and medium for multi-load distribution network

CN122017387ACN 122017387 ACN122017387 ACN 122017387ACN-122017387-A

Abstract

The invention discloses a dynamic power supply fault monitoring system, method, equipment and medium for a multi-load distribution network, wherein the method comprises a first power supply signal acquisition module, a second power supply signal acquisition module, a time domain mutation feature extraction module, a frequency domain mode feature extraction module and a fault monitoring result output module, wherein the first power supply signal acquisition module is used for arranging a multi-branch sensor on the multi-load distribution network and acquiring a first power supply signal of a target power distribution branch, the second power supply signal acquisition module is used for carrying out disturbance desensitization processing on the first power supply signal through a signal disturbance desensitization mechanism to obtain a second power supply signal, the time domain mutation feature extraction module is used for extracting time domain mutation features of the second power supply signal, the frequency domain mode feature extraction module is used for extracting frequency domain mode features of the second power supply signal. The invention can effectively inhibit the interference signals generated by load switching, frequency conversion operation and the like by constructing a signal disturbance desensitization mechanism and adopting a processing mode of combining frequency band weighting, sub-band filtering and time domain enhancement, and simultaneously retains and enhances the characteristics of real fault signals.

Inventors

LI XU
XIAO XINFU
HU ZUGUO
ZHANG LU
XIAO LI
ZHU LEI
ZHOU JINGUO
SHEN YOUQIANG
LI SHENGFU
ZOU YIN
LI KANG
QIAN ZHENGCHAO
TIAN DI
GAN YI
ZHAO YUHAN
WANG ZIHUA
CHEN ZELIN
LUO YAN
ZHANG XINYANG
Yuan Xingqian
GE JUNSHAN
TANG XINGXIN
YANG YI
LUO XUAN
PENG YILIN
LI WEIXING
Jia Bensheng
XU XIAOYUN
JIAN BEI
TIAN FEI
Nie Yuanjie
LI MINGHU
ZHOU ZHONGBO

Assignees

贵州电网有限责任公司

Dates

Publication Date: 20260512
Application Date: 20251225

Claims (10)

1. A dynamic power failure monitoring system for a multi-load distribution network, comprising: The first power supply signal acquisition module is used for arranging a multi-branch sensor on a multi-load distribution network and acquiring a first power supply signal of a target power distribution branch; The second power supply signal obtaining module is used for carrying out disturbance desensitization processing on the first power supply signal through a signal disturbance desensitization mechanism to obtain a second power supply signal; the time domain mutation feature extraction module is used for extracting time domain mutation features of the second power supply signal; the frequency domain modal feature extraction module is used for extracting frequency domain modal features of the second power supply signal; The fault monitoring result output module is used for inputting the time domain mutation characteristic and the frequency domain modal characteristic as a time-frequency double-domain characteristic set into a double-domain discrimination model for analysis, analyzing the time domain mutation characteristic according to the discrimination result and outputting a fault monitoring result.
2. The multiple load division dominant network oriented dynamic power failure monitoring system of claim 1, wherein the second power signal acquisition module performs the steps of: constructing a sliding window disturbance sample library; Performing waveform similarity matching on the first power supply signal through a signal disturbance desensitization mechanism and the sliding window disturbance sample library to obtain waveform similarity; and when the waveform similarity is higher than a set threshold value, activating the signal disturbance desensitization mechanism to perform disturbance desensitization processing on the first power supply signal to obtain a second power supply signal.
3. The multiple load division dominant network oriented dynamic power failure monitoring system of claim 2, wherein the signal disturbance desensitization mechanism performs the steps of: the signal disturbance desensitization mechanism performs disturbance desensitization processing including frequency band weighting, subband filtering and time domain enhancement; The frequency band weighting is used for identifying a disturbance frequency band of the first power supply signal, and a weight factor is set to carry out weight suppression on the disturbance frequency band; The sub-band filtering is used for identifying a disturbance sub-band of the first power supply signal and carrying out filtering processing on the disturbance sub-band; And the time domain enhancement is to identify time domain mutation characteristics of the first power supply signal, and slope enhancement processing is carried out on the time domain mutation characteristics.
4. The multi-load division dominant network oriented dynamic power failure monitoring system of claim 3, wherein the frequency domain modal feature extraction module performs the steps of: Decomposing the second power supply signal through a preset layer wavelet packet to obtain a plurality of decomposed sub-bands, wherein the preset layer is X; Calculating the energy of the plurality of sub-bands to form a frequency band energy feature vector, and outputting the frequency band energy feature vector as wavelet packet decomposition frequency domain features; Performing Fourier transform on the second power supply signal to obtain spectrum distribution data; and outputting the frequency peak value frequency, the spectrum center of gravity and the spectrum amplitude range of the spectrum distribution data as Fourier transform frequency domain characteristics.
5. The multiple load division dominant network oriented dynamic power failure monitoring system of claim 4, wherein the failure monitoring result output module performs the steps comprising: Initializing a model architecture; collecting disturbance sample power supply signals marked with disturbance events and fault sample power supply signals marked with fault events in a multi-load branch power distribution system according to a preset sampling period; Combining the disturbance sample power supply signal and the fault sample power supply signal to obtain a time-frequency dual-domain feature training sample; Training the model framework according to the time-frequency double-domain feature training sample to obtain a trained double-domain discrimination model; and deploying the trained two-domain discriminant model to a monitoring master station of the multi-load branch power distribution system.
6. The multiple load division dominant network oriented dynamic power failure monitoring system of claim 5, wherein the step performed by the failure monitoring result output module further comprises: The dual-domain discriminant model calculates a modal distance index according to the time domain mutation characteristic and the frequency domain modal characteristic ; Setting a disturbance mode threshold; When the modal distance index is When the fault detection value is larger than the disturbance mode threshold value, outputting a fault event discrimination result, performing fault index calculation on the time domain mutation feature, and generating a fault alarm signal according to the fault index; When the modal distance index is And outputting a disturbance event discrimination result when the disturbance event discrimination result is not larger than the disturbance modal threshold value.
7. The dynamic power failure monitoring system for multiple load distribution network of claim 6, wherein a modal distance index is calculated The method comprises the following steps: ; Wherein, the The frequency domain features are decomposed for the wavelet packet of the current supply signal, Decomposing frequency domain feature samples for wavelet packets stored in the time-frequency dual-domain feature training samples, Is the fourier transform frequency domain characteristic of the current supply signal, For fourier transform frequency domain feature samples stored in the time-frequency dual domain feature training samples, Is a norm used to characterize the distance measure over the feature space.
8. A method for monitoring dynamic power supply faults oriented to a multi-load distribution network, applying the system as claimed in any of claims 1 to 7, comprising the steps of: arranging a multi-branch sensor on a multi-load distribution network, and collecting a first power supply signal of a target power distribution branch; Performing disturbance desensitization processing on the first power supply signal through a signal disturbance desensitization mechanism to obtain a second power supply signal; extracting time domain mutation characteristics of the second power supply signal; extracting frequency domain modal characteristics of the second power supply signal; and taking the time domain mutation feature and the frequency domain modal feature as a time-frequency double-domain feature set to be input into a double-domain discrimination model for analysis, analyzing the time domain mutation feature according to discrimination results, and outputting fault monitoring results.
9. An electronic device, comprising: A memory and a processor; the memory is configured to store computer executable instructions that, when executed by the processor, implement the steps of the multi-load division multiple access network oriented dynamic power failure monitoring system of any one of claims 1 to 7.
10. A computer readable storage medium storing computer executable instructions which when executed by a processor implement the steps of the multi-load division dominating network oriented dynamic power failure monitoring system of any of claims 1 to 7.

Description

Dynamic power supply fault monitoring system, method, equipment and medium for multi-load distribution network Technical Field The invention relates to the technical field of distribution network fault monitoring, in particular to a dynamic power supply fault monitoring system, method, equipment and medium for a multi-load distribution network. Background With the continuous expansion of the scale of the distribution network, the increasingly diversified types of loads and the wide access of distributed power supplies, the operation environment of the multi-load branch distribution network becomes increasingly complex, and the occurrence frequency of power supply faults and the range of the influence of the faults on the outside are also continuously increased. However, in actual operation, because a certain disturbance of current or voltage is inevitably introduced in frequent load switching, variable frequency operation, start-stop action and the like, the generated signal time sequence waveform, abrupt slope, energy change and the like are very similar to the waveform of the actual fault to a great extent. However, the conventional fault detection means is difficult to deeply dig complex features of signals, the disturbance and the real faults are difficult to effectively distinguish, interference signals are brought to parallel operation of loads of multiple branches, and the disturbance of normal loads is easily misjudged as faults, so that misoperation of a protection device is caused, or when the real faults occur, the situation of refusal occurs because the corresponding signal features are covered by the interference signals, the reliability and the stability of a power supply system are seriously influenced, waste of power resources and equipment damage are caused, and even the safety of electricity utilization is threatened. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, the invention provides a dynamic power supply fault monitoring system, method, equipment and medium for a multi-load distribution network, which solve the problems of insufficient fault monitoring accuracy and high misjudgment rate caused by poor modal identification capability and dynamic disturbance adaptability to complex signals in the prior art, and the power supply is not stable and reliable enough. In order to solve the technical problems, the invention provides the following technical scheme: In a first aspect, the present invention provides a dynamic power failure monitoring system for a multi-load distribution network, including: The first power supply signal acquisition module is used for arranging a multi-branch sensor on a multi-load distribution network and acquiring a first power supply signal of a target power distribution branch; The second power supply signal obtaining module is used for carrying out disturbance desensitization processing on the first power supply signal through a signal disturbance desensitization mechanism to obtain a second power supply signal; the time domain mutation feature extraction module is used for extracting time domain mutation features of the second power supply signal; the frequency domain modal feature extraction module is used for extracting frequency domain modal features of the second power supply signal; The fault monitoring result output module is used for inputting the time domain mutation characteristic and the frequency domain modal characteristic as a time-frequency double-domain characteristic set into a double-domain discrimination model for analysis, analyzing the time domain mutation characteristic according to the discrimination result and outputting a fault monitoring result. The method comprises the steps of establishing a sliding window disturbance sample library, performing waveform similarity matching on a first power supply signal through a signal disturbance desensitization mechanism and the sliding window disturbance sample library to obtain waveform similarity, and activating the signal disturbance desensitization mechanism to perform disturbance desensitization processing on the first power supply signal when the waveform similarity is higher than a set threshold value to obtain a second power supply signal. The dynamic power supply fault monitoring system for the multi-load distribution network comprises a signal disturbance desensitization mechanism, wherein the signal disturbance desensitization mechanism performs disturbance desensitization processing comprising frequency band weighting, sub-band filtering and time domain enhancement, the frequency band weighting is used for identifying a disturbance frequency band of a first power supply signal, a weight factor is set for carrying out weight suppression on the disturbance frequency band, the sub-band filtering is used for identifying a disturbance sub-band of the first power supply signal, the disturbance sub-b