Please use this identifier to cite or link to this item: https://idr.l1.nitk.ac.in/jspui/handle/123456789/15866
Title: An efficient online sequential extreme learning machine model based on feature selection and parameter optimization using cuckoo search algorithm for multi-step wind speed forecasting
Authors: Shetty R.P.
Sathyabhama A.
Pai P.S.
Issue Date: 2021
Citation: Soft Computing , Vol. 25 , 2 , p. 1277 - 1295
Abstract: Accurate wind speed forecasting (WSF) has become increasingly important to overcome the adverse effects of stochastic nature of the wind on wind power generation. This paper proposes a multi-step hybrid online WSF model by combining online sequential extreme learning machine (OSELM), optimized variational mode decomposition (OVMD) and cuckoo search optimization algorithm (CSO). OVMD decomposes the wind speed series into subseries, and CSO selects the input features for each subseries. Multi-step forecasting for each subseries is performed using OSELM model optimized by CSO. Finally, the forecasting results are obtained by the aggregate calculations. The proposed model has been examined by using 10-min average wind speed data collected in monsoon and winter seasons from a supervisory control and data acquisition system of a 1.5 MW wind turbine situated in central dry zone of Karnataka, India. The results reveal that the model proposed captures the nonlinear characteristics of the wind speed in a better manner in comparison with the batch learning approach, giving accurate wind speed forecasts. This can help wind farms to estimate the wind power in a location efficiently. © 2020, Springer-Verlag GmbH Germany, part of Springer Nature.
URI: https://doi.org/10.1007/s00500-020-05222-x
http://idr.nitk.ac.in/jspui/handle/123456789/15866
Appears in Collections:1. Journal Articles

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.