Voiceprint image feature-based transformer state recognition method, voiceprint image feature-based transformer state recognition systems and medium

An image feature and state recognition technology, applied in image enhancement, image analysis, image data processing and other directions, can solve the problems of easy omission of features, insufficient judgment experience, and low accuracy in acoustic diagnostic systems, and achieves improvement in human judgment. The effect of high diagnostic accuracy and fast diagnostic speed

Inactive Publication Date: 2019-12-31
STATE GRID HUNAN ELECTRIC POWER +2
View PDF7 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The technical problem to be solved by the present invention: Aiming at the above-mentioned problems in the prior art, a transformer state recognition method, system and medium based on voiceprint image features are provided. The present invention can use voiceprint image features to identify abnormal operation of transformers related to noise It can significantly improve the problems of insufficient experience and large errors in human judgment, and the problems of easy omission of features and low accuracy in existing acoustic diagnosis systems. It can diagnose and classify abnormal operating conditions of transformers related to noise without relying on human experience. Significantly improve the diagnostic accuracy of noise-related transformer operating status, with the advantages of high diagnostic accuracy and fast diagnostic speed

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
  • Voiceprint image feature-based transformer state recognition method, voiceprint image feature-based transformer state recognition systems and medium
  • Voiceprint image feature-based transformer state recognition method, voiceprint image feature-based transformer state recognition systems and medium
  • Voiceprint image feature-based transformer state recognition method, voiceprint image feature-based transformer state recognition systems and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0040] Such as figure 1 As shown, the implementation steps of the transformer state recognition method based on voiceprint image features in this embodiment include:

[0041] 1) Acquire the sound signal of the target transformer in the running state;

[0042] 2) Generate a voiceprint feature image based on the sound signal;

[0043] 3) Extracting the feature information of the voiceprint image in the voiceprint feature image;

[0044] 4) Using the voiceprint image feature information and the preset transformer state feature recognition library to identify the state of the target transformer related to noise. mapping relationship between them.

[0045] In this embodiment, the voiceprint feature image generated in step 2) is a two-dimensional color image, and the horizontal axis and the vertical axis are time and frequency respectively, and the signal amplitude is represented by the color depth of the color image.

[0046] Such as figure 2 As shown, the lower side of the fig...

Embodiment 2

[0066] This embodiment is basically the same as Embodiment 1, and its main differences are:

[0067] 1. The method of extracting the feature information of the voiceprint image in the voiceprint feature image in step 3) is different. Extracting the feature information of the voiceprint image in step 3) of this embodiment specifically refers to using a convolutional neural network to extract the features in the feature image of the voiceprint to obtain the feature information of the voiceprint image, which is compared to the method in Embodiment 1. In terms of generality, it is better.

[0068] 2. In step 4), the method of identifying the noise-related state of the target transformer based on the feature information of the voiceprint image is different. Such as Figure 4 As shown, in step 4) of this embodiment, the method of identifying the noise-related state of the target transformer based on the feature information of the voiceprint image is as follows: input the feature i...

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 voiceprint image feature-based transformer state recognition method, voiceprint image feature-based transformer state recognition systems and a medium. The recognition methodcomprises the following steps of: generating voiceprint feature images according to sound signals, extracting voiceprint image feature information from the voiceprint feature images, and comparing the voiceprint image feature information with a preset transformer state feature recognition library or inputting the voiceprint image feature information into a pre-trained artificial intelligence machine learning model to recognize the noise-related state of a target transformer. According to the invention, the voiceprint image features can be used to recognize the noise-related abnormal operationstate of the transformer, the problems of insufficient human judgment experience and large errors and the problems of easy omission of the features and low accuracy caused by an existing acoustic diagnostic system can be significantly improved, diagnostic accuracy of the noise-related transformer operation state can be greatly improved, diagnostic accuracy is high and diagnostic speed is fast.

Description

technical field [0001] The invention relates to the technical field of power transformer operation state evaluation, in particular to a transformer state recognition method, system and medium based on voiceprint image features, which can identify noise-related abnormal operation states of transformers by using voiceprint image features. Background technique [0002] Power transformer failure is the key cause of large-scale power outages in power systems. More than 30% of the typical fault defects of the transformer (such as: overexcitation, three-phase unbalanced load, harmonic load, cooling system defect, DC bias, loose winding and core, loose accessories, overload, etc.) can be detected by noise. Existing fault detection methods related to transformer noise mainly rely on operating personnel to use human ears to judge the type of fault based on experience, or use the waveform and frequency spectrum of the sound signal to extract a certain number of sound features as the ba...

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): G10L17/26G10L21/10G06T7/00
CPCG06T7/001G06T2207/30108G10L17/26G10L21/10
Inventor 吴晓文卢铃曹浩胡胜陈炜彭继文吕建红
Owner STATE GRID HUNAN ELECTRIC POWER
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