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

Non-intrusive load decomposition and monitoring method based on difference

A differential and differential information technology, applied in the field of non-intrusive load decomposition and monitoring, can solve problems such as increasing the learning difficulty of neural networks, limiting decomposition accuracy, etc., achieving the effect of easy learning and training, reducing requirements, and improving usability

Pending Publication Date: 2019-09-20
UNIVERSITY OF MACAU
View PDF4 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing neural network models directly input the original total load data to the neural network for training, and in this way, key features such as electrical power changes are implicitly included in the original data, which increases the neural network. learning difficulty, which limits the decomposition accuracy

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
  • Non-intrusive load decomposition and monitoring method based on difference
  • Non-intrusive load decomposition and monitoring method based on difference
  • Non-intrusive load decomposition and monitoring method based on difference

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0066] This embodiment provides a non-intrusive load decomposition method based on differential input strategy and neural network, the flow chart is shown in the attached figure 1 , the specific processing flow includes the following steps:

[0067] S1. First obtain the total real-time power X of each electrical appliance on the master meter, and at the same time acquire the real-time power Y of the target electrical appliance (target load) at the corresponding time.

[0068] S2. Calculate the differential information of the total real-time power X on the master meter, that is, calculate the real-time variation of the total power, and obtain the real-time power variation sequence Xd.

[0069] S3, using the power change sequence Xd as the input of the neural network model, and using the real-time power Y at the time corresponding to the target load as the output of the neural network model to obtain a training sample set;

[0070] S4. Using the training sample set to train the...

Embodiment 2

[0107] Through the above description, we understand the implementation process of the non-intrusive load splitting method based on the differential input strategy and neural network of the present invention. This process can be realized by the non-intrusive load splitting device. Next, we will describe the interior of the non-intrusive load splitting device of the present invention Structure and function are introduced.

[0108] see Figure 5 , in the embodiment of the present invention, the non-intrusive load decomposition device 10 includes: an acquisition module 101 , a difference module 102 , a sliding module 103 and an analysis module 104 .

[0109] The acquiring module 101 is used to acquire the total real-time power X of each electrical appliance on the master meter, and at the same time acquire the real-time power Y of the target electrical appliance at a corresponding time.

[0110] The difference module 102 is used to calculate the difference information of the tota...

Embodiment 3

[0129] Figure 7 It is a schematic structural diagram of a non-intrusive load decomposition and inference device 100 based on a differential input strategy and a neural network according to Embodiment 3 of the present invention. The device 100 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (central processing units, CPU) 1001 (for example, one or more processors) and memory 1002 . Wherein, the program for executing the non-intrusive load decomposition and inference method of the present invention can be stored in the memory 1002, and the central processing unit 1001 can be configured to communicate with the memory 1002 to execute the non-intrusive load decomposition and inference method on the device 100 A series of command operations.

[0130] Those skilled in the art should understand that the embodiments of the present invention may be provided as methods, systems, or computer prog...

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 provides a non-intrusive load decomposition and monitoring method based on a differential information input strategy and a deep neural network method. The load decomposition method comprises the following steps: an obtaining step: obtaining the total real-time power X of each electric appliance on a total table, and obtaining the real-time power Y of a target electric appliance at the corresponding time; a difference step: calculating difference information of the total real-time power X on the total table to obtain a real-time power change sequence Xd; a sample set forming step: taking the power change sequence Xd as input, taking the real-time power Y of the target electric appliance at the corresponding moment as output, obtaining a plurality of sample pairs, and forming a training sample set; and an analysis step: training by adopting the training sample set to complete the analysis model. The method has the characteristics of high accuracy, small neural network scale and strong generalization capability, and can overcome the problem of high sampling frequency requirements of other methods.

Description

technical field [0001] The invention relates to the field of electric power technology, in particular to a non-invasive load decomposition and monitoring method based on a differential information input strategy and a neural network method. Background technique [0002] With the increasingly prominent energy problems and the development needs of smart homes / factories, we not only hope to monitor the total load power consumption in a room or building, but also hope to monitor the power consumption of each electrical appliance in this building, so that Help users understand the usage of electrical appliances, formulate more reasonable power consumption plans, and improve the utilization rate of electric energy. At the same time, these data can also be used as raw data to help the realization of smart homes and smart factories. In addition, this technology plays a vital role in power supply management on the power supply side, smart grid, real-time electricity pricing, etc. ...

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): G06Q10/06G06Q50/06H02J3/00G06N3/04
CPCG06Q10/06393G06Q50/06H02J3/00H02J2203/20G06N3/045
Inventor 张渊猛马少丹杨光华施政
Owner UNIVERSITY OF MACAU
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