Skip to the main content

Original scientific paper

https://doi.org/10.17559/TV-20221007130139

Improved Electricity Portfolio Prediction Based on Optimized Ant Colony Algorithm

Li Zemin ; Inner Mongolia Power (Group) Company Limited Electricity Marketing Service and Operation Management Branch


Full text: english pdf 1.250 Kb

page 458-464

downloads: 350

cite


Abstract

In order to overcome the shortcomings of the traditional single power forecasting method, the article uses LSTM network, GM model and SVR support vector machine regression model to forecast electricity, and also uses ant colony optimization algorithm to build a new combined forecasting model for the three forecasting methods, which takes into account the factors affecting power forecasting more comprehensively and helps to improve the accuracy of power forecasting. The paper also uses the ant colony algorithm to optimize the weights of the single forecasting method, which can effectively avoid the problem of the traditional algorithm falling into the local optimal point, and obtain a more accurate power combination forecasting model. Through application examples, it is verified that the combined forecasting model can effectively improve the accuracy of power forecasting and provide reference for power system planning and operation. The research results show that the combined prediction has a greater improvement in accuracy compared with the single Gray, LSTM network and other predictions.

Keywords

combinatorial prediction; gray prediction; neural networks; support vector machines

Hrčak ID:

294353

URI

https://hrcak.srce.hr/294353

Publication date:

26.2.2023.

Visits: 926 *