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

Intelligent electric meter fault prediction method and system based on BiLSTM-CNN model

A smart meter and fault prediction technology, which is applied in prediction, neural learning methods, biological neural network models, etc., can solve the problem that the neural network model cannot extract valid information of smart meter fault data well, so as to ensure high-quality operation, The effect of reducing operation and maintenance costs

Pending Publication Date: 2022-07-05
GUANGXI POWER GRID CORP
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The present invention proposes a smart meter fault prediction method and system based on the BiLSTM-CNN model, which solves the problem in the related art that a single neural network model cannot extract valid information from smart meter fault data well

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
  • Intelligent electric meter fault prediction method and system based on BiLSTM-CNN model
  • Intelligent electric meter fault prediction method and system based on BiLSTM-CNN model
  • Intelligent electric meter fault prediction method and system based on BiLSTM-CNN model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0040] like Figure 1 to Figure 2 As shown, this embodiment proposes a smart meter fault prediction method based on the BiLSTM-CNN model, which includes the following three main steps:

[0041] 1. Data preprocessing

[0042] The data preprocessing steps include feature selection and data cleaning.

[0043] (1) Feature selection: In addition to selecting smart meter device data, operating data, and fault data, it is also necessary to select environmental factors in the operating environment where the smart meter is located, such as temperature and humidity.

[0044] (2) Data cleaning: Clean the data contained in the data, such as repeated values, missing values, and outliers, which affect the results of the model analysis.

[0045] 2. Unbalanced data sampling

[0046] There is a problem of category data imbalance after data cleaning through fault data features, so it is necessary to sample imbalanced data before modeling.

[0047] For the data imbalance problem, the present...

Embodiment 2

[0082] Based on the same inventive concept of Embodiment 1 above, the present invention also proposes a smart meter fault prediction system based on the BiLSTM-CNN model, including:

[0083] The acquisition module is used to acquire the historical fault data of the smart meter as characteristic data, including equipment data, operation data, fault type, fault time, temperature, and humidity;

[0084] a sample module, configured to use the equipment data, operation data, failure time, temperature, and humidity as the model input X, and the failure type as the model output Y, to construct a training sample set (X, Y);

[0085] The training module is used to initialize the model training parameters, and repeat the model training steps until the preset number of iterations of the model is reached;

[0086] A normalization module, for inputting to a fully connected neural network of (N*N) dimension through the training sample set, and outputting a normalized vector of each feature ...

Embodiment 3

[0096] Based on the same inventive concept of Embodiment 1 above, the present invention also provides a computer-readable storage medium, wherein,

[0097] The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, implements the steps of the method for predicting faults of a smart meter based on the BiLSTM-CNN model.

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 the technical field of intelligent electric meter fault prediction, and provides an intelligent electric meter fault prediction method and system based on a BiLSTM-CNN model, and the method comprises the steps: obtaining historical fault data of an intelligent electric meter as feature data, including equipment data, operation data, a fault type, fault time, temperature and humidity; the feature data is subjected to data cleaning, the data cleaning comprises cleaning of duplicate values, missing values and abnormal values contained in historical fault data features, and then a BiLSTM-CNN model is constructed; according to the technical scheme, the feature set influencing the fault of the intelligent electric meter is encoded by using the advantage of extracting sequence time sequence information by the bidirectional long and short time memory network and the advantage of extracting local feature information by the convolutional neural network, and finally, the encoded vector is used for classifying and predicting the fault of the intelligent electric meter. The problem that effective information in the fault data of the intelligent electric meter cannot be well extracted by a single neural network model in the prior art is solved.

Description

technical field [0001] The invention relates to the technical field of smart meter fault prediction, in particular to a smart meter fault prediction method and system based on a BiLSTM-CNN model. Background technique [0002] As an important link in the power production process, smart energy meters are one of the important technical and economic assessment indicators for power companies. They play an important role in my country's economic construction and are directly related to the economic interests of power generation, power supply and electricity consumption. With the gradual increase of smart meter devices, the problem of electric energy meter faults has become more and more prominent. Most of the traditional smart meter fault prediction methods use machine learning or a single neural network model to analyze the device information and operation information of the smart meter itself. Information modeling and analysis, while ignoring the impact of external factors 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
IPC IPC(8): G06Q10/04G06N3/04G06N3/08G06K9/62G06Q50/06
CPCG06Q10/04G06N3/084G06Q50/06G06N3/047G06N3/045G06F18/23213G06F18/2415Y04S10/50
Inventor 杨舟陈珏羽周政雷梁炜皓蒋雯倩唐志涛林秀清
Owner GUANGXI POWER GRID CORP
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