Open AccessA Wind Speed Prediction Method Based on Improved Empirical Mode Decomposition and Support Vector Machine
Author(s) -
Shibo Wang,
Yun Guo,
Yanzhuo Wang,
Qinghua Li,
Nan Wang,
Shumin Sun,
Yan Cheng,
Peng Yu
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/680/1/012012
Subject(s) - hilbert–huang transform , wind speed , support vector machine , mode (computer interface) , series (stratigraphy) , computer science , noise (video) , algorithm , term (time) , time series , subsequence , pattern recognition (psychology) , artificial intelligence , mathematics , machine learning , white noise , meteorology , telecommunications , paleontology , mathematical analysis , physics , bounded function , quantum mechanics , image (mathematics) , biology , operating system
Based on the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and bat algorithm (BA) to optimize the support vector machine, this paper proposed a combined model for short-term wind speed forecasting to predict the wind speed more accurately. Firstly, CEEMDAN was used to decompose the original wind speed time series into a series of subsequences with different frequencies. Secondly, the decomposed subsequences were forecasted by combined model of BA-SVM. Finally, the wind speed forecasting results was achieved by superposing each predicted subsequence. The simulation results suggest that the model improves the prediction accuracy and reduces the error.