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

Power equipment state estimation method and device based on improved LSTM

A technology of power equipment and state, which is applied in the direction of neural learning methods, prediction, biological neural network models, etc., can solve problems such as poor numerical quality, low value density of power equipment state data, and difficult to accurately predict the trend of the state of the equipment, so as to improve data quality. The effect of improving authenticity and prediction accuracy

Pending Publication Date: 2020-12-25
STATE GRID ZHEJIANG NINGBO FENGHUA POWER SUPPLY CO LTD +2
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In order to solve the shortcomings and deficiencies existing in the prior art, the present invention proposes a power equipment state estimation method and device based on improved LSTM, which can overcome the problems of low value density and poor value quality of power equipment state data, which make the state trend difficult Accurately Predicted Defects

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
  • Power equipment state estimation method and device based on improved LSTM
  • Power equipment state estimation method and device based on improved LSTM
  • Power equipment state estimation method and device based on improved LSTM

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0058] Specifically, the embodiment of the present application proposes an improved LSTM-based power equipment state estimation method, such as figure 1 shown, including:

[0059] 11. Import the power equipment status detection data into the gradient descent tree GBDT for data processing to obtain high-quality data sequences;

[0060] 12. Calculate the relative error between the power equipment detection data and the high-quality data sequence;

[0061] 13. Use the obtained relative error to improve the forgetting gate of LSTM;

[0062] 14. Import the high-quality data series into the improved LSTM to predict the state trend of power equipment.

[0063] In implementation, the technical solution proposed in the embodiment of this application includes two key points, data quality improvement and LSTM model improvement. The improvement of data quality includes using GBDT to fit the real trend distribution of data; the improvement of LSTM model is to improve the forgetting gate...

Embodiment 2

[0097] On the other hand, the embodiment of the present application proposes an improved LSTM-based power equipment state estimation device 3, such as image 3 shown, including:

[0098] The data processing unit 31 is used to import the power equipment state detection data into the gradient descent tree GBDT for data processing to obtain high-quality data sequences;

[0099] An error calculation unit 32, configured to calculate the relative error between the power equipment detection data and the high-quality data sequence;

[0100] The forget gate improvement unit 33 is used to improve the forget gate of LSTM by using the obtained relative error;

[0101] The trend prediction unit 34 is used to import high-quality data series into the improved LSTM to predict the state trend of electric equipment.

[0102] In implementation, the technical solution proposed in the embodiment of this application includes two key points, data quality improvement and LSTM model improvement. Th...

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 relates to a power equipment state estimation method and device based on improved LSTM, and the method comprises the steps: improving data quality, improving an LSTM (Long Short Term Memory) model, wherein the data quality improvement comprises the fitting of the real trend distribution of data through employing a GBDT (Gradient Boosting Decision Tree), and the improved LSTM model improves a forget door of a conventional LSTM model. Compared with the prior art, the invention has the advantages that the data value is improved, and the prediction accuracy is high.

Description

technical field [0001] The invention relates to the technical field of power equipment reliability prediction, in particular to a method and device for power equipment state prediction based on improved LSTM. Background technique [0002] At present, from the perspective of data inherent law analysis, valuable knowledge for power equipment status evaluation, diagnosis and prediction has been discovered, and a multi-source data-driven power equipment status evaluation model has been established to realize personalized status evaluation and abnormal status of power equipment. Detection, accurate prediction of state changes, and intelligent diagnosis of faults can comprehensively, timely and accurately grasp the health status of power equipment, and provide auxiliary decision-making basis for intelligent operation inspection of equipment and optimal operation of power grids. However, the existing power equipment status detection data has problems such as low data value density ...

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): G06Q10/04G06Q50/06G06K9/62G06N3/04G06N3/08
CPCG06Q10/04G06Q50/06G06N3/049G06N3/08G06F18/2148G06F18/24323
Inventor 夏家峰秦立明张明达孙益辉王思谨崔昊杨夏晟邹轩
Owner STATE GRID ZHEJIANG NINGBO FENGHUA POWER SUPPLY CO LTD
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