An integration method of multiple prediction results of power load probability density

A technology of probability density and power load, which is applied in the field of integration of various prediction results of power load probability density, to achieve accurate description, improve prediction accuracy and reduce costs

Active Publication Date: 2019-04-02
TSINGHUA UNIV
View PDF6 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the field of electric load forecasting, there is no related research on the integration of probability density forecasting

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
  • An integration method of multiple prediction results of power load probability density
  • An integration method of multiple prediction results of power load probability density
  • An integration method of multiple prediction results of power load probability density

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0066] The integration method of various forecasting results of the electric load probability density proposed by the present invention, its flow chart is as follows figure 1 shown, including the following steps:

[0067] (1) The historical power load data D = [d 1 , d 2 ,... d t ,... d T ] is divided into three parts according to the set ratio, generally speaking, the ratio setting of the first part is larger than the sum of the ratios of the second part and the third part, and the ratio is 10:1:1 in one embodiment of the present invention The three parts of data are recorded as: training set D 1 , validation set D 2 and combination set D 3 ,

[0068]

[0069] Among them, the data set D 1 is of length T 1 , data set D 2 is of length T 2 , data set D 3 The length is T-T 1 -T 2 , in one embodiment of the present invention [ ] is the rounding down function;

[0070] (2) Using different hyperparameters, train D separately 1 The three probability prediction mo...

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 an integration method of multiple prediction results of power load probability density, and belongs to the technical field of power system analysis. According to the method, aplurality of probability density or quantile probability prediction models are obtained through training of a plurality of three-class regression models set by different hyper-parameters, and the output of the quantile prediction models is converted into a probability density model conforming to Gaussian distribution through Gaussian distribution assumption of loads and a least square method. A probability density prediction integration method is adopted, a probability density prediction optimal integration model is constructed based on the trained probability density prediction model and result, the weights of different probability density prediction methods are determined, and therefore the probability loss of the continuous grade of the final integration prediction model is minimum. The method is finally converted into a quadratic programming problem, the global optimal integrated weight is quickly searched by utilizing mature commercial software, the probability density short-termload prediction precision is improved, and the dispatching operation cost of the power system is further reduced.

Description

technical field [0001] The invention relates to an integration method of multiple prediction results of electric load probability density, belonging to the technical field of power system analysis. Background technique [0002] Power load forecasting is an important part of power system planning and the basis of power system economic operation. In order to assist the power system to make optimal decisions to effectively reduce the planning and operating costs of the power system, high-precision load forecasting is essential. In recent years, with the continuous increase in the scale of the power system, the addition of intermittent energy sources such as wind power and solar energy, and the continuous and rapid growth of distributed renewable energy have made the load of the power system show strong uncertainty. The importance of probabilistic load forecasting research to power system planning and operation has begun to emerge. Compared with traditional point forecasting, ...

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): G06Q10/04G06Q50/06G06N3/00G06N3/02
CPCG06N3/006G06N3/02G06Q10/04G06Q50/06
Inventor 王毅李天一张宁康重庆
Owner TSINGHUA UNIV
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