Electric power medium-and-long-term load prediction method based on a long-and-short-term memory model

A long-short-term memory and load forecasting technology, which is applied in forecasting, neural learning methods, biological neural network models, etc., can solve problems such as weak adaptability to changes, limited forecasting accuracy, and limited algorithm complexity, so as to deepen cognition and accuracy Improve, reduce the effect of the build

Pending Publication Date: 2019-04-19
安徽数升数据科技有限公司
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The advantage of the above algorithm is that it is relatively intuitive and easy to use; the disadvantage is that the complexity of the algorithm is limited, the adaptability to changes is weak, and the prediction accuracy is limited

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
  • Electric power medium-and-long-term load prediction method based on a long-and-short-term memory model
  • Electric power medium-and-long-term load prediction method based on a long-and-short-term memory model
  • Electric power medium-and-long-term load prediction method based on a long-and-short-term memory model

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment 1

[0026] Such as figure 1 As shown, the present invention is a method for medium and long-term load forecasting of electric power based on a long-short-term memory model, comprising the following steps:

[0027] Step 1. Collect the historical data of regional factors by year as sample characteristics; divide the regional load in summer into base load and cooling load, and calculate the base load in different years, that is, the average value of regional load without cooling load, to obtain the load ratio : Load ratio=(area load-base load) / base load; sample features include daily maximum temperature value, daily minimum temperature value, daily average temperature, high temperature duration days, sunshine level, wind speed, rainfall situation and maximum load value; daily average The calculation of temperature is: daily average temperature = (daily maximum temperature + daily minimum temperature) / 2, and the number of high temperature continuous days of the day is the number of da...

specific Embodiment 2

[0048] Such as figure 2 As shown, the present invention is a method for medium and long-term load forecasting of electric power based on a long-short-term memory model, comprising the following steps:

[0049] Step 1. Selectively extract the data sources of the historical data of regional factors by year, and regularly extract the historical data after regular update;

[0050] Step 2: periodically analyze the selectively extracted historical data, and perform data cleaning and feature construction on it; at the same time, perform data cleaning and feature construction on the regularly updated historical data extracted regularly;

[0051] Step 3: Integrating and summarizing the data and characteristics of the two, using the Markov model to select the high-temperature sequence segment with the greatest possibility of predicting the year, and constructing a long-short-term memory model;

[0052] Step 4: Evaluate and apply the long-short-term memory model, and obtain application...

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 an electric power medium and long term load prediction method based on a long-and-short-term memory model, and relates to the technical field of electric power prediction. Themethod comprises the following steps of dividing regional loads in summer are into basic loads and cooling loads, and acquiring the load ratio by calculating the basic loads of different years; Selecting a high-temperature sequence fragment with the maximum probability of prediction years by utilizing a Markov model; Constructing a long-short-term memory model: summarizing and integrating the loadratio data and regional characteristic factors of the high-temperature sequence fragments so as to train and generate the long-short-term memory model for power load prediction; Performing model training and optimization; And predicting the power load of the time sequence segment of the future year. According to the method, the medium-and-long-term load of the electric power is predicted based onthe long-and-short-term memory model, and the relevance between the load of the region in the summer of the attack peak and the external condition characteristics is found, so that the regional loadunder different external conditions in the future is predicted, and the operation stability of a regional power grid is improved.

Description

technical field [0001] The invention belongs to the technical field of electric power forecasting, in particular to a medium and long-term electric power load forecasting method based on a long-short-term memory model. Background technique [0002] Power grid planning is the basis and prerequisite for power grid construction. The quality of planning will seriously affect the construction investment cost of the power grid and the safe operation of the power grid. The medium and long-term load forecasting is the theoretical basis of power grid planning, and the accuracy of the forecasting algorithm directly affects the planning scheme. credibility. Most of the current load forecasting algorithms are based on the relationship between power and regional economy, such as elastic coefficient method, output value unit consumption method, etc., or simple mathematical models, such as linear regression model, gray model, etc. The advantage of the above algorithm is that it is relativ...

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): G06Q10/04G06Q50/06G06N3/04G06N3/08
CPCG06N3/08G06Q10/04G06Q50/06G06N3/044G06N3/045
Inventor 刘峰
Owner 安徽数升数据科技有限公司
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