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

A microgrid load prediction method based on an LSSVM neural network

A load forecasting and neural network technology, applied in neural learning methods, biological neural network models, forecasting, etc., can solve problems such as increased forecasting errors, low forecasting accuracy, and uneven load characteristic curves, so as to improve accuracy, learn High efficiency and global optimal effect

Inactive Publication Date: 2019-04-23
KUNMING UNIV OF SCI & TECH
View PDF0 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The microgrid load has obvious fluctuations and sudden changes, resulting in an unsmooth load characteristic curve. For load forecasting, the greater the change in load characteristics, the lower the prediction accuracy.
Scholars at home and abroad have covered a wide range of research on microgrids, but there are still too many uncertain factors; some methods have improved the convergence speed and solved local optimization problems, but with the increase of the prediction period, the prediction error will also increase. increase

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
  • A microgrid load prediction method based on an LSSVM neural network
  • A microgrid load prediction method based on an LSSVM neural network
  • A microgrid load prediction method based on an LSSVM neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034] Embodiment 1: a kind of microgrid load forecasting method based on the neural network of LSSVM, comprises the following steps:

[0035] Step 1 normalizes and preprocesses the historical data of the microgrid to form a training sample set; since the dimension and magnitude of the characteristic indexes are not the same, the effect of a characteristic index with a particularly large magnitude on the classification may be highlighted during the calculation process. In order to eliminate the influence of the difference in characteristic index units and the different magnitudes of characteristic indexes, it must be normalized so that each index value is unified in a certain common numerical characteristic range.

[0036] (1) Normalization of load data

[0037] Apply logarithmic treatment to the loadings:

[0038] x' ab = lg(x ab ) (2.1)

[0039] where x ab is the original load, x′ ab is the normalized load.

[0040] (2) Division and standardization of day types

[00...

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 relates to a microgrid load prediction method based on an LSSVM neural network, and belongs to the field of microgrid loads. The invention provides a load prediction method based on a least square support vector machine technology in combination with a neural network, and the method is characterized in that firstly, model parameters of a least square support vector regression machineare optimized by using an ant colony algorithm, the regression machine is trained by using an optimized model, the regression machine provides better structure and parameters for an RBF neural network, and finally, a prediction result is obtained through matlab simulation training. According to the invention, the micro-grid coincidence prediction accuracy is improved.

Description

technical field [0001] The goal of the microgrid load forecasting proposed in this paper is to take into account the algorithm efficiency and network performance, improve the accuracy of power forecasting, and make the forecasting reach the global optimum. Background technique [0002] With the rapid development of modern information technology and the continuous improvement of the automation level of the power system, the power load forecasting method is also continuously improved and perfected. The existing power load forecasting methods can be roughly divided into two categories: traditional forecasting methods and artificial intelligence forecasting methods. Traditional forecasting methods include time series method, trend extrapolation method, gray model method and regression model method, etc. Artificial intelligence forecasting methods include fuzzy reasoning, neural network, wavelet analysis and expert system methods. The microgrid load has obvious fluctuations and...

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): G06Q10/04G06Q50/06G06N3/00G06N3/04G06N3/08G06N20/10
CPCG06N3/006G06N3/08G06Q10/04G06Q50/06G06N3/045Y04S10/50
Inventor 唐菁敏马含
Owner KUNMING UNIV OF SCI & TECH
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