Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Transformer operation state prediction method based on long short term memory network and transformer operation state prediction system based on long short term memory network

A long- and short-term memory and operating state technology, applied in neural learning methods, biological neural network models, instruments, etc., can solve problems such as threats to normal production and life, damage to transformers, etc.

Active Publication Date: 2018-05-15
SHANGHAI JIAO TONG UNIV +1
View PDF8 Cites 25 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Once a fault occurs, it will not only seriously damage the transformer, but also greatly threaten the normal production and life of the people.

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
  • Transformer operation state prediction method based on long short term memory network and transformer operation state prediction system based on long short term memory network
  • Transformer operation state prediction method based on long short term memory network and transformer operation state prediction system based on long short term memory network
  • Transformer operation state prediction method based on long short term memory network and transformer operation state prediction system based on long short term memory network

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0100] In order to increase the learning speed and effect, and reduce the risk of the network falling into a local minimum during learning, the weight matrix in LSTM is first randomly initialized by obeying a Gaussian distribution with a mean of 0 and a variance of 1, and then using singular value decomposition to obtain an orthogonal base matrix as an initialization value [22] . The LSTM bias item and the output layer bias item are initialized to 0, and the output layer weight matrix is ​​multiplied by 0.01 by a random number that obeys a Gaussian distribution with a mean of 0 and a variance of 1. The prediction model consists of one layer of LSTM network and Softmax network layer. The size of the input layer is 100, and the number of neurons in the hidden layer of LSTM is 128. To prevent overfitting, the signal loss rate is set to 0.2, and the size of the output layer is 4.

[0101] Taking a 500kV #2 main transformer as an example, the basic situation of the transformer is...

example 2

[0107] Taking a 220kV #1 main transformer as an example, the transformer was shipped in April 2000 and put into operation on June 23, 2000. After it was put into operation, the operation of the main transformer was basically good, and the overall load rate was relatively high. Periodic chromatographic testing found that the total hydrocarbon value in the main transformer oil increased significantly after the main transformer reached its peak in 2010 and passed the summer. The value of total hydrocarbons increased slowly year by year, but did not exceed the value for attention, and the data of other characteristic gases except total hydrocarbons were normal. Table 2 shows some oil chromatographic online monitoring data from June to July 2013.

[0108] Table 2. Online monitoring data of a 220kV #1 main transformer oil chromatography (unit: uL / L)

[0109] time

H 2

CH 4

C 2 h 4

C 2 h 6

C 2 h 2

total hydrocarbon

CO

CO 2

...

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 transformer operation state prediction method based on a long short term memory network. The transformer operation state prediction method comprises the steps that (1) historical information related to the transformer operation state is acquired; (2) the historical operation state of the transformer is assessed based on the historical information; (3) a transformer operation state prediction model based on the long short term memory network is constructed; (4) the transformer operation state prediction model is trained based on the historical information and the historical operation state; and (5) the operation state of the transformer in the future is predicted based on the historical information through the transformer operation state prediction model. Besides,the invention also discloses a corresponding system. The operation state of the transformer in the future can be effectively predicted; the potential threat of the transformer can be perceived in timeand the fault development trend of the transformer can be mastered by prediction of the transformer operation state; and the method and the system have important meaning for enhancing the equipment operation safety and reliability.

Description

technical field [0001] The invention relates to the field of power equipment monitoring, in particular to a transformer operation state prediction method and system. Background technique [0002] During the service process, power transformers suffer from long-term thermal, electrical and mechanical stresses, and gradually deteriorate from a completely good state until they fail. Once a fault occurs, it will not only seriously damage the transformer, but also greatly threaten the normal production and life of the people. Predicting the operating status of transformers is helpful to timely perceive potential threats to transformers and grasp the development trend of transformer faults. [0003] The continuous accumulation of transformer status panorama information provides a prerequisite for the evaluation and prediction of transformer operating status. Transformer status panorama information usually includes various information related to transformer operation status, such ...

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): G01R31/00G06N3/08G06N3/04
CPCG01R31/00G06N3/049G06N3/084
Inventor 代杰杰盛戈皞徐玲玲宋辉侯慧娟江秀臣
Owner SHANGHAI JIAO TONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com