Particle Swarm Optimization-Enhanced Twin Support Vector Regression for Wind Speed Forecasting
Wind energy is considered one of the renewable energy sources that minimize the cost of electricity production. This article proposes a hybrid approach based on particle swarm optimization (PSO) and twin support vector regression (TSVR) for forecasting wind speed (PSO-TSVR). To enhance the forecasti...
Main Author: | |
---|---|
Format: | Article |
Language: | English |
Published: |
De Gruyter
2017-11-01
|
Series: | Journal of Intelligent Systems |
Subjects: | |
Online Access: | https://doi.org/10.1515/jisys-2017-0378 |
id |
doaj-c72895c51cd6471aafa0d66af551afc5 |
---|---|
record_format |
Article |
spelling |
doaj-c72895c51cd6471aafa0d66af551afc52021-09-06T19:40:38ZengDe GruyterJournal of Intelligent Systems0334-18602191-026X2017-11-0128590591410.1515/jisys-2017-0378Particle Swarm Optimization-Enhanced Twin Support Vector Regression for Wind Speed ForecastingHoussein Essam H.0Faculty of Computers and Information, Minia University, El Minya, EgyptWind energy is considered one of the renewable energy sources that minimize the cost of electricity production. This article proposes a hybrid approach based on particle swarm optimization (PSO) and twin support vector regression (TSVR) for forecasting wind speed (PSO-TSVR). To enhance the forecasting accuracy, TSVR was utilized to forecast the wind speed, and the optimal settings of TSVR parameters were optimized by PSO carefully. Moreover, to estimate the performance of the suggested approach, three wind speed benchmark data of OpenEI were used as a case study. The experimental results revealed that the optimized PSO-TSVR approach is able to forecast wind speed with an accuracy of 98.9%. Further, the PSO-TSVR approach has been compared with two well-known algorithms such as genetic algorithm with TSVR and the native TSVR using radial basis kernel function. The computational results proved that the proposed approach achieved better forecasting accuracy and outperformed the comparison algorithms.https://doi.org/10.1515/jisys-2017-0378wind speed predictingtwin support vector regression (tsvr)particle swarm optimization (pso)genetic algorithm (ga)68t20 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Houssein Essam H. |
spellingShingle |
Houssein Essam H. Particle Swarm Optimization-Enhanced Twin Support Vector Regression for Wind Speed Forecasting Journal of Intelligent Systems wind speed predicting twin support vector regression (tsvr) particle swarm optimization (pso) genetic algorithm (ga) 68t20 |
author_facet |
Houssein Essam H. |
author_sort |
Houssein Essam H. |
title |
Particle Swarm Optimization-Enhanced Twin Support Vector Regression for Wind Speed Forecasting |
title_short |
Particle Swarm Optimization-Enhanced Twin Support Vector Regression for Wind Speed Forecasting |
title_full |
Particle Swarm Optimization-Enhanced Twin Support Vector Regression for Wind Speed Forecasting |
title_fullStr |
Particle Swarm Optimization-Enhanced Twin Support Vector Regression for Wind Speed Forecasting |
title_full_unstemmed |
Particle Swarm Optimization-Enhanced Twin Support Vector Regression for Wind Speed Forecasting |
title_sort |
particle swarm optimization-enhanced twin support vector regression for wind speed forecasting |
publisher |
De Gruyter |
series |
Journal of Intelligent Systems |
issn |
0334-1860 2191-026X |
publishDate |
2017-11-01 |
description |
Wind energy is considered one of the renewable energy sources that minimize the cost of electricity production. This article proposes a hybrid approach based on particle swarm optimization (PSO) and twin support vector regression (TSVR) for forecasting wind speed (PSO-TSVR). To enhance the forecasting accuracy, TSVR was utilized to forecast the wind speed, and the optimal settings of TSVR parameters were optimized by PSO carefully. Moreover, to estimate the performance of the suggested approach, three wind speed benchmark data of OpenEI were used as a case study. The experimental results revealed that the optimized PSO-TSVR approach is able to forecast wind speed with an accuracy of 98.9%. Further, the PSO-TSVR approach has been compared with two well-known algorithms such as genetic algorithm with TSVR and the native TSVR using radial basis kernel function. The computational results proved that the proposed approach achieved better forecasting accuracy and outperformed the comparison algorithms. |
topic |
wind speed predicting twin support vector regression (tsvr) particle swarm optimization (pso) genetic algorithm (ga) 68t20 |
url |
https://doi.org/10.1515/jisys-2017-0378 |
work_keys_str_mv |
AT housseinessamh particleswarmoptimizationenhancedtwinsupportvectorregressionforwindspeedforecasting |
_version_ |
1717768062653956096 |