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

AI-based intelligent diagnosis method, system and device

A technology of electricity consumption data and models, applied in the field of AI-based intelligent diagnosis, can solve problems such as large time span, misjudgment, and data ineffectiveness, achieve low-latency business response, fast network service response, and improve energy The effect of the indicator

Active Publication Date: 2020-12-18
马欣 +1
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] 1. "Energy index management is not predictable" - the current energy index management system is mainly based on monitoring and data storage, a large amount of data does not play a role, and it is impossible to predict energy indexes in real time, so that it is impossible to accurately judge the abnormal situation of energy indexes;
[0009] 2. "Difficulty in diagnosing the cause of abnormal energy indicators" - most of the energy indicator management systems currently in use are based on data calculation, with weak analysis capabilities, and cannot accurately locate the real-time abnormalities of each station, equipment, line, station area, and user It is very difficult to manage energy indicators, and the diagnosis of abnormal energy indicators depends heavily on the professional level of personnel, and there are risks of misjudgment and missed judgment;
[0010] 3. "Governance decisions rely on manual work" - the current energy index management is at a semi-automated level, heavily dependent on manual experience and manual decision-making, and lacks effective technical and data support
[0011] 4. "Insufficient utilization of data value" - most of the current analysis work of energy index management relies on the linkage of major business systems. Insufficient value mining

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
  • AI-based intelligent diagnosis method, system and device
  • AI-based intelligent diagnosis method, system and device
  • AI-based intelligent diagnosis method, system and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0062] like figure 1 As shown, the present invention provides a kind of intelligent diagnosis method based on AI, comprising:

[0063] Collect electricity consumption data;

[0064] Input the collected electricity consumption data into the trained artificial intelligence AI model to obtain relevant data.

[0065] The artificial intelligence AI model includes a power stealing AI model to detect power stealing data, wherein the power stealing AI model is trained according to the following procedures: collecting labeled power consumption data; preprocessing the collected power consumption data; The preprocessed electricity consumption data is used to train the electricity stealing AI model.

[0066] Specifically, for the training of the model, the real power consumption data of the station area is collected through the IOT edge device for the training of the model. The data feature set includes time, single-phase / three-phase voltage, single-phase / three-phase current, neutral l...

Embodiment 2

[0092] A kind of intelligent diagnosis method based on AI of the present invention comprises:

[0093]We have collected a large amount of real power consumption data in the station area through IOT edge devices for model training. The total amount of training data collected by each IOT device is about 20w, including time, single-phase / three-phase voltage, single-phase / three-phase current, neutral line voltage / current, phase angle, whether to open the cover and other index data as features set, and also contains the actual user classification data (normal users, power-stealing users) as the actual results of user classification as labels to train the model.

[0094] The 20w data collected through IOT edge devices contains some missing and erroneous data, so we need to preprocess the data first so that the data can reach the standard of training data. Through the EDA (Exploratory Data Analysis) method, we found that the distribution of high-quality data is more obvious, showing...

Embodiment 3

[0102] An AI-based intelligent diagnosis method of the present invention also includes:

[0103] Collect electricity consumption data;

[0104] Input the collected electricity consumption data into the trained artificial intelligence AI model to obtain relevant data.

[0105] Preferably, the artificial intelligence AI model also includes the station area list AI model, the technical loss AI model, the household change relationship AI model, the collection AI model, the measurement AI model, and the archives AI model.

[0106] Among them, the relevant models are as follows:

[0107] Table area AI model

[0108] The station area meter is used to measure the difference between power supply and electricity sales (called line loss electricity). The station area meter counts the power supply and electricity sales of the 10kV / 400V low-voltage sub-area, and provides basic data on the line loss of the station area. The AI ​​module of the station area table mainly completes the follo...

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 present invention relates to the field of smart grids, in particular to an AI-based intelligent diagnosis method, system and device. The core is AI+IoT, which consists of standardized and industrialized IoT edge computing hardware devices plus customized AI modules, including: the collection unit of IoT devices collects power consumption data; the cloud platform trains AI models based on historical data, and uses new data to Constantly improve the AI ​​model dynamically, and the edge computing unit of the IoT device inputs the collected power consumption data into the trained artificial intelligence AI model to obtain relevant data. Realize the real-time prediction of governance, effectively improve the real-time monitoring level of abnormally exceeded energy indicators, and change the previous working mode of semi-manual system detection and manual verification to a fully automatic intelligent AI analysis mode. Finally, a new software and hardware form of cloud platform + edge computing equipment + smart field terminal + smart IoT chipset module will be realized, the user-side energy system and power distribution management system will be redefined, and the era of "software-defined energy" will be opened.

Description

technical field [0001] The invention relates to the field of smart grids, in particular to an AI-based intelligent diagnosis method, system and device. Background technique [0002] At present, the construction of the ubiquitous electric power Internet of Things is progressing steadily, and tasks in six major fields, including basic support, data sharing, and internal and external business, have been carried out successively. As one of the core economic and technical indicators of power grid enterprises, line loss reflects the operating costs and economic benefits of the enterprise. Strengthening line loss management is a long-term strategic task and systematic project for power grid enterprises. The smart grid is a multi-objective function constrained by balance. Its factors are numerous, the conditions are random, the line loss is changeable, and the causes are complex. Factors change dynamically, making physical modeling difficult. Manual experience and existing methods ...

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 Patents(China)
IPC IPC(8): G06Q10/04G06Q10/06G06Q50/06G06K9/62
CPCG06Q10/04G06Q10/06393G06Q50/06G06F18/24323G06F18/214Y02P90/82
Inventor 马欣孙钊
Owner 马欣
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