Open AccessSupport Vector Machine and Long Short-term Memory using Multivariate Models for Wind Power Forecasting
Author(s) -
Eunju Kang,
Nam-Rye Son
Publication year - 2020
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.d2036.029420
Subject(s) - wind power , support vector machine , wind speed , wind power forecasting , computer science , term (time) , volatility (finance) , renewable energy , power (physics) , electric power system , artificial intelligence , meteorology , engineering , econometrics , mathematics , geography , electrical engineering , physics , quantum mechanics
Renewable energy has recently gained considerable attention. In particular, interest in wind energy is rapidly increasing globally. However, the characteristics of instability and volatility in wind energy systems also have a significant on power systems. To address these issues, numerous studies have been carried out to predict wind speed and power. Methods used to forecast wind energy are divided into three categories: physical, data-driven (statistical and artificial intelligence methods), and hybrid methods. In this study, among artificial intelligence methods, we compare short-term wind power using a support vector machine (SVM) and long short-term memory (LSTM). The method using an SVM is a short-term wind power forecast that considers the wind speed and direction on Jeju Island, whereas the method using LSTM does not consider the wind speed and direction. As the experiment results indicate, the SVM method achieves an excellent performance when considering the wind speed and direction.