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

Power load prediction method based on window mobile machine learning

A technology for electric loads and moving machines, which is applied in the field of electric load forecasting based on window mobile machine learning, and can solve problems such as poor adaptability of unary regression, large amount of data, and low prediction accuracy

Active Publication Date: 2021-02-09
STATE GRID HEILONGJIANG ELECTRIC POWER CO LTD ELECTRIC POWER RES INST +1
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the problem of low prediction accuracy caused by the large amount of data, strong nonlinearity and poor adaptability of machine learning to unary regression in the existing methods for predicting electric load, and proposes a machine learning based on window movement Electric Load Forecasting Method

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 load prediction method based on window mobile machine learning
  • Power load prediction method based on window mobile machine learning
  • Power load prediction method based on window mobile machine learning

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0033] Specific implementation mode 1: In this implementation mode, the specific process of the power load forecasting method based on window mobile machine learning is as follows:

[0034] Step 1, collecting raw power load data;

[0035] The original power load data is the power load data set at the current moment and the time data set at which the power load is collected;

[0036] The current moment electric load Y={y 1 ,y 2 ,...,y n};

[0037] The moment X=[x 1 , x 2 ,...,x n};

[0038] Step 2. Upscaling the original power load data, the specific process is:

[0039] Step 21. Group the electric loads at the current moment:

[0040] Using a moving window of size l, moving groupings are performed in chronological order of electric loads:

[0041] Y={a 1 , a 2 ,...,a m}

[0042] a i ={y p(i-1)+1 ,y p(i-1)+2 ,...,y p(i-1)+l}

[0043] Among them, Y is the power load at the current moment, a i is the electric load of group i at the current moment, m is the tot...

Embodiment

[0067] Collect raw power load data ( figure 2 ): 35 days of power load collection time in 2019, the time interval is one hour, a total of 840 times and the power load value of a substation corresponding to each time.

[0068] For this section of load data, carry out mobile grouping, where l=20*24=480, p=24, k=15*24=360, that is, take the previous 15 days as the characteristic power load Next 5 days as target power load Adjacent two groups a i and a i+1 The interval is one day. Taking the load of the first 30 days for model training, a total of 11 sets of load data after dimension upgrading can be obtained. Here, since the target power load data is 5 days, and the moving step is 1 day, in order to prevent the false accuracy caused by repeated forecasting, this verification process uses the load value of the 16th-30th day as the characteristic power load of the test data, using The 31-35 day load value is used as the target power load of the test data.

[0069] Use KNN,...

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 discloses a power load prediction method based on window mobile machine learning, and relates to the field of power load prediction. The objective of the invention is to solve the problem of low prediction accuracy of an existing power load prediction method for a data set with large data volume and strong nonlinearity. The method comprises the following steps: step 1, collecting original power load data; step 2, raising the dimension of the original power load data; step 3, obtaining a power load prediction model by using the power load training after dimension raising; step 4,verifying the accuracy of the power load prediction model; and step 5, during actual charge prediction, inputting the power load data into the power load prediction model to obtain a prediction result. The invention is mainly used for predicting the power load.

Description

technical field [0001] The invention relates to the field of power load forecasting, in particular to a power load forecasting method based on window moving machine learning. Background technique [0002] In the process of power production, the load is closely related to the safe and stable operation of the power grid, the consumption of new energy, the economic operation of the power grid, and optimal dispatching. Power load data is affected by many aspects such as industrial production, people's daily life, climate, etc., and is a data type with high nonlinearity. Therefore, it is necessary to propose a more reasonable and accurate power load forecasting method. [0003] At present, in power load forecasting, when the relationship between independent variables and dependent variables is relatively clear, polynomial fitting, least squares method, etc. are mainly used, but the above methods are aimed at the accuracy of data with strong nonlinearity and large amount of data. ...

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): G06Q50/06G06Q10/04G06N20/00
CPCG06Q50/06G06Q10/04G06N20/00Y04S10/50
Inventor 郑君徐明宇陈晓光武国良刘洋祖光鑫刘智洋刘进张美伦郝文波张睿
Owner STATE GRID HEILONGJIANG ELECTRIC POWER CO LTD ELECTRIC POWER RES INST
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