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

Building short-term load prediction method based on ARIMA-LSTM combination model

A technology of short-term load forecasting and combined models, which is applied in forecasting, neural learning methods, biological neural network models, etc., can solve the problems of inaccuracy and single precision of building short-term load forecasting, and achieve the effect of improving forecasting accuracy and convergence speed

Inactive Publication Date: 2020-06-09
SHANGHAI UNIVERSITY OF ELECTRIC POWER
View PDF4 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In view of the above-mentioned existing building short-term load forecasting method based on the ARIMA-LSTM combined model, there is a single and inaccurate problem of building short-term load forecasting accuracy, the present invention is proposed

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
  • Building short-term load prediction method based on ARIMA-LSTM combination model
  • Building short-term load prediction method based on ARIMA-LSTM combination model
  • Building short-term load prediction method based on ARIMA-LSTM combination model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0065] Reference figure 1 with figure 2 , Provides a schematic diagram of the overall structure of a short-term building load forecasting method based on the ARIMA-LSTM combined model, such as figure 1 , A short-term building load forecasting method based on the ARIMA-LSTM combined model includes collecting influencing factor data through a data collector, and performing maximum and minimum normalization processing on the load data and each influencing factor data to obtain a dimensionless data set;

[0066] Select key influencing factors;

[0067] Calculate the cosine similarity and obtain the sample data of similar days as the training set;

[0068] Input the similar daily load training set into the ARIMA-LSTM combined model to obtain the load forecast result;

[0069] Wherein, the influencing factor data includes load data, weather data and date type data.

[0070] Specifically, the main body of the present invention includes S1: Collect influencing factor data through a data coll...

Embodiment 2

[0110] This embodiment is to verify and explain the technical effects used in the method. The different methods selected in this embodiment and the method are used for comparative testing, and the test results are compared by means of scientific demonstration to verify the true effects of the method.

[0111] First, collect the data of building load influencing factors through the data collector, including load data, meteorological data and date type data. The data sampling interval is 1 hour, and 24 data points are recorded every day; then the analysis and screening of building load forecasting When training sample data, analyze the load influencing factors, then preprocess the original sequence data, and finally select similar daily sequence data according to the gray correlation degree and cosine distance.

[0112] Among them, such as figure 1 As shown, the analysis and operation steps of the collected data are described as follows:

[0113] 1) Since meteorological factors have a...

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 building short-term load prediction method based on an ARIMA-LSTM combination model, and the method comprises the steps: collecting influence factor data through a data collector, and carrying out the maximum and minimum normalization of the load data and all influence factor data, and obtaining a dimensionless data set; selecting key influence factors; calculating cosinesimilarity, and obtaining similar day sample data as a training set; inputting the similar day load training set into an ARIMA-LSTM combination model to obtain a load prediction result, wherein the influence factor data comprises load data, meteorological data and date type data; according to the method, when the training sample data for building load prediction is analyzed and screened, the similar day data sequence is selected by considering the meteorological factors and the grey correlation degree of the date type sequence, so that the prediction precision is effectively improved.

Description

Technical field [0001] The invention relates to the technical field of electric power load forecasting, in particular to a short-term building load forecasting method based on an ARIMA-LSTM combined model. Background technique [0002] With the rapid development of urbanization in China, the proportion of building energy consumption will continue to rise; in order to alleviate the energy crisis and improve environmental degradation, reducing building energy consumption and improving energy efficiency management have received the focus of the industry, which in turn makes building load forecasting Become an important research content in the construction of the ubiquitous power Internet of Things; accurate load forecasting provides a decision-making basis for the building energy efficiency management system to formulate power demand response and load dispatch planning, which is conducive to optimizing the balance of supply and demand and improving the utilization of power equipment ...

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
CPCG06Q10/04G06Q50/06G06N3/08G06N3/044G06N3/045
Inventor 崔承刚李鹏辉官乐乐马波杨宁陈辉
Owner SHANGHAI UNIVERSITY OF ELECTRIC POWER
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