Quality treatment method and device for power grid data

A data and power grid technology, applied in the field of power grid data quality management, can solve problems such as large errors, large error application effects, and variation, and achieve the effect of improving the quality of data assets

Pending Publication Date: 2022-08-02
国家电网有限公司大数据中心
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] However, in the existing approximate calculation methods, only a small part of the data will produce a large error in the approximate calculation, and the large error produced by these small amounts of data is the main reason for the deterioration of the entire application effect

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
  • Quality treatment method and device for power grid data
  • Quality treatment method and device for power grid data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] See attached figure 1 , figure 1 It is a schematic flowchart of the main steps of the quality control method for power grid data according to an embodiment of the present invention. like figure 1 As shown, the quality control method for power grid data in the embodiment of the present invention mainly includes the following steps:

[0050] Step S101: Test the pre-built deep learning network that has been approximately calculated to obtain the error generated by the pre-constructed deep learning network that has been approximately calculated;

[0051] Step S102: determining a data identification network based on an error generated by the approximated pre-built deep learning network;

[0052] Step S103: Use the data identification network to classify power grid data.

[0053] In this example, the data identification network is determined based on the error generated by the approximately calculated pre-built deep learning network, including:

[0054] If the error gene...

Embodiment approach

[0079] Based on the above scheme, the present invention provides a best embodiment, which specifically includes:

[0080] 1) First collect K pieces of real data, including K1 positive samples and K2 negative samples, and K1+K2=K. For K1 positive samples, expand the data to N1 approximate data according to the normal distribution rule; for K2 negative samples, expand the data to N2 approximate data according to the random distribution rule, and obtain N1+N2=N approximate data in total.

[0081] 2) Convert the N pieces of approximate data into training data that the neural network can use, in order to make it fit the input and output characteristics of the selected neural network model.

[0082] 3) After the collected data is transformed, it will be used as the training data of the neural network for training. This patent proposes a network training method and structure for approximate calculation of parallel convolutional neural networks, which is characterized in that the dee...

Embodiment 2

[0085] Based on the same inventive concept, the present invention also provides a quality management device for power grid data, such as figure 2 As shown, the quality control device for power grid data includes:

[0086] The test module is used to test the approximated pre-built deep learning network to obtain the error generated by the approximated pre-built deep learning network;

[0087] a determining module for determining a data identification network based on errors generated by the approximately calculated pre-built deep learning network;

[0088] The classification module is used for classifying the grid data by using the data identification network.

[0089] Preferably, the data identification network is determined based on the error generated by the approximated pre-built deep learning network, including:

[0090] If the error generated by the approximated pre-built deep learning network is not greater than 2 times the preset error limit, the approximated pre-bui...

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 power grid data security, and particularly provides a power grid data quality treatment method and device, and the method comprises the steps: testing a pre-constructed deep learning network which is subjected to approximate calculation, and obtaining an error generated by the pre-constructed deep learning network which is subjected to approximate calculation; determining a data identification network based on an error generated by the approximately calculated pre-constructed deep learning network; and classifying the power grid data by using the data identification network. The technical scheme provided by the invention not only can protect core power grid data, but also can maintain data characteristics, and can be used for test verification of various algorithms.

Description

technical field [0001] The invention relates to the technical field of power grid data security, in particular to a quality control method and device for power grid data. Background technique [0002] Data sorting is the preliminary data preprocessing work in the process of mining and refining data value. Studies have shown that more than 80% of the work of many big data analysis tasks is spent on data sorting, which brings huge labor costs to data analysis. More importantly, due to the lack of systematic and theoretical support, the quality of data collation varies widely, which brings great uncertainty to the results of data analysis and greatly affects the mining and refining of the value of big data. Therefore, it is necessary to pay attention to data sorting, which is an important basic work in the data quality governance link. [0003] Compared with traditional data quality governance, data quality governance is endowed with significant big data technical characteris...

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): G06K9/62G06N3/04G06N3/08G06F16/215
CPCG06N3/08G06F16/215G06N3/045G06F18/217G06F18/214G06F18/24Y04S10/50
Inventor 王路涛陈振宇武丽莎杨畅秦明王家凯吕宏伟张乐
Owner 国家电网有限公司大数据中心
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