High-voltage power cable partial discharge phenomenon detection method based on deep learning

A technology of high-voltage electric power and partial discharge, applied in neural learning methods, testing dielectric strength, computer components, etc., can solve problems such as difficult to quantify, unbalanced positive and negative samples, and difficult to distinguish, so as to reduce the frequency of failures and simplify Effects of failure detection and loss avoidance

Inactive Publication Date: 2020-08-25
TIANJIN UNIV
View PDF8 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But it is difficult to quantify, and it is not easy to distinguish the interference signal of the equipment in operation
[0007] 3) Optical measurement method: It mainly uses sound wave compression to change the properties of the optical fiber and cause the characteristics of the output information of the optical fiber to change, and then measure the partial discharge signal, but this technology is not yet mature
[0011] However, for the task of detecting partial discharge phenomena, no deep learning method has yet been applied to this field.
In this field, there are problems such as fewer samples, extremely unbalanced positive and negative samples, and data noise, etc. These problems have a great impact on the classification task of voltage time series
How to obtain a higher accuracy for the final classification under the influence of these problems is still a challenging problem

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • High-voltage power cable partial discharge phenomenon detection method based on deep learning
  • High-voltage power cable partial discharge phenomenon detection method based on deep learning
  • High-voltage power cable partial discharge phenomenon detection method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] A method for detecting partial discharge phenomena in high-voltage power lines using a deep learning method, the method comprising the following steps:

[0038] 101: Determine whether manual processing of the original high-voltage power cable voltage time-series signal data is required;

[0039] For example: normalization, data compression, data enhancement, etc.

[0040] Wherein, the preprocessing method in the step 101 is specifically:

[0041] For the three-phase AC, firstly process the signal of each phase: divide the time-series voltage signal data of each phase into several segments, and then calculate the mean mean, standard deviation std, mean plus and minus standard deviation mean±std( Indicates that the single measurement standard deviation and the random error normal distribution curve are used as the standard to describe its degree of dispersion), the amplitude max_range of this interval, and this interval is respectively located at 0%, 1%, 25%, 50%, and 75%....

Embodiment 2

[0049] The following combined with specific examples, Figure 1-Figure 4 The scheme in embodiment 1 is further introduced, it mainly has four parts to form:

[0050] 1) Preprocessing the original high-voltage power cable voltage time-series signal data; 2) The overall structure of the model used; 3) Fine-tuning of parameters; 4) Analysis of model prediction results.

[0051] The data set used in this example is some voltage timing signals related to partial discharge in high-voltage power lines collected by the new electric meter designed by the ENET Center of Ostrava Technical University (VSB-STUDIO). Its data set contains 2904 samples of high-voltage wires, such as figure 1 As shown, it shows the distribution of positive and negative samples in the data set. Since the measured line is a high-voltage AC line, each sample contains data of three phases, and the data is collected every 0.02s. The single phase of each sample contains data of 800,000 time points, of which figur...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a method for detecting the partial discharge phenomenon of a high-voltage power cable based on deep learning. The method comprises steps of carrying out preprocessing of the voltage time sequence signal data of an original high-voltage power cable, obtaining features of a high-voltage power line signal, and skipping the step if data preprocessing is not needed; converting the preprocessed high-voltage power cable voltage time sequence signal or the unprocessed signal into time sequence data inputted by a partial discharge phenomenon detection model; and obtaining a final partial discharge phenomenon detection model through training of the training set, judging performance of the detection model according to the evaluation indexes, and then improving the detection model till a model with a satisfactory effect is obtained. The method is advantaged in that whether a partial discharge phenomenon occurs in a monitored voltage time sequence signal of a certain sectionof high-voltage power line can be accurately judged.

Description

technical field [0001] The invention relates to the field of high-voltage power cables, in particular to a method for detecting partial discharge phenomena of high-voltage power cables based on deep learning. Background technique [0002] As an important medium for urban power transmission, high-voltage transmission lines play an important role in various fields such as national development and people's lives. Under the new situation of China's rapid economic development, the total electricity consumption of people, enterprises, factories, etc. is also increasing, and the scale of the power system is also becoming more complex and expanding. In this huge circuit network, high-voltage power cable distribution is like the entire skeleton of the human body, bearing the needs of people's daily life and production work. [0003] In the early stage of power failure in high-voltage power lines, a phenomenon called partial discharge (PD) often occurs, which is equivalent to the dis...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G01R31/12G06K9/00G06N3/04G06N3/08
CPCG01R31/12G06N3/08G06N3/044G06N3/045G06F2218/08
Inventor 孙美君许广远王征
Owner TIANJIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap