Open AccessTidal current prediction based on fractal theory and improved least squares support vector machine
Author(s) -
Zhang Anan,
Lin Yu,
Sun Yangfan,
Yuan Haiyun,
Wang Min,
Liu Guofeng,
Hu Yiru
Publication year - 2021
Publication title -
iet renewable power generation
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.005
H-Index - 76
eISSN - 1752-1424
pISSN - 1752-1416
DOI - 10.1049/rpg2.12335
Subject(s) - current (fluid) , fractal , algorithm , autocorrelation , time series , mathematics , computer science , statistics , engineering , mathematical analysis , electrical engineering
An accurate short‐term tidal current prediction is essential to optimize the operation mode of offshore microgrids. According to the correlation characteristics of tidal current speed and direction in time series, a tidal current prediction model is proposed by using fractal theory, dragonfly optimization algorithm and least squares support vector machine (DA‐LSSVM). First, the Rescaled range analysis (R/S analysis) in fractal theory is used to construct the tidal current analytic model with moving Hurst index of time series characteristics and Von‐Mises statistics (V statistic). The model evaluates the stability and autocorrelation of the tidal current and reveals the persistence of tidal current fluctuation. It also shows the orderliness hidden behind the randomness, indicating that the tidal current characteristics can be extended. Second, the least squares support vector machine is used to train the samples in high‐dimensional space, and the dragonfly algorithm is used to optimize the parameters of the least squares support vector machine to obtain the best prediction model. Finally, based on the empirical data recorded in the straits in North America, the model was used for the analysis and prediction of tidal current speed and tidal current direction, which verified the validity and accuracy of the proposed method.