A Self-Adaptive Multikernel Machine Based on Recursive Least-Squares Applied to Very Short-Term Wind Power Forecasting
Wind power has contributed significantly to the increase in electricity generation, but a decision-making tool capable of dealing with its variability and limited predictability is necessary. For this purpose, a novel self-adaptive approach for kernel recursive least-squares machines named multiple...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9495822/ |
| _version_ | 1830357313641250816 |
|---|---|
| author | Erick C. Bezerra Pierre Pinson Ruth P. S. Leao Arthur P. S. Braga |
| author_facet | Erick C. Bezerra Pierre Pinson Ruth P. S. Leao Arthur P. S. Braga |
| author_sort | Erick C. Bezerra |
| collection | DOAJ |
| description | Wind power has contributed significantly to the increase in electricity generation, but a decision-making tool capable of dealing with its variability and limited predictability is necessary. For this purpose, a novel self-adaptive approach for kernel recursive least-squares machines named multiple challengers is introduced in this work, which is successfully used to produce very short-term wind power forecasts at eight wind farms in Australia. The proposed method is based on a competitive tracking method, and the algorithm deals with some common difficulties of kernel methods, e.g., the increasing kernel matrix size associated with time and memory complexities and the overfitting problem. The proposed method always considers the new information received by the model, thus identifying changes in the time series, avoiding abrupt loss of information and maintaining a controlled number of examples since there is an adaptive selection of the active kernel. It works with the smallest dictionary possible, reducing the probability of overfitting. Five minute-ahead wind power forecasts are produced and evaluated in terms of point forecast skill scores and calibration. The results of the proposed method are compared with those provided by other kernel-based versions of the recursive least-squares algorithm, an online version of the extreme learning machine method, and the persistence time series model. An increase in the number of kernels used in the ensemble system can lead to better results when compared with those of single-kernel models. |
| first_indexed | 2024-12-20T02:21:34Z |
| format | Article |
| id | doaj.art-de49795fe5a14911a0a6b21d96561985 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-20T02:21:34Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-de49795fe5a14911a0a6b21d965619852022-12-21T19:56:48ZengIEEEIEEE Access2169-35362021-01-01910476110477210.1109/ACCESS.2021.30999999495822A Self-Adaptive Multikernel Machine Based on Recursive Least-Squares Applied to Very Short-Term Wind Power ForecastingErick C. Bezerra0https://orcid.org/0000-0003-1074-9704Pierre Pinson1Ruth P. S. Leao2Arthur P. S. Braga3Electrical Engineering Department, Technical University of Denmark, Lyngby, DenmarkElectrical Engineering Department, Technical University of Denmark, Lyngby, DenmarkElectrical Engineering Department, Federal University of Ceará, Fortaleza, Ceará, BrazilElectrical Engineering Department, Federal University of Ceará, Fortaleza, Ceará, BrazilWind power has contributed significantly to the increase in electricity generation, but a decision-making tool capable of dealing with its variability and limited predictability is necessary. For this purpose, a novel self-adaptive approach for kernel recursive least-squares machines named multiple challengers is introduced in this work, which is successfully used to produce very short-term wind power forecasts at eight wind farms in Australia. The proposed method is based on a competitive tracking method, and the algorithm deals with some common difficulties of kernel methods, e.g., the increasing kernel matrix size associated with time and memory complexities and the overfitting problem. The proposed method always considers the new information received by the model, thus identifying changes in the time series, avoiding abrupt loss of information and maintaining a controlled number of examples since there is an adaptive selection of the active kernel. It works with the smallest dictionary possible, reducing the probability of overfitting. Five minute-ahead wind power forecasts are produced and evaluated in terms of point forecast skill scores and calibration. The results of the proposed method are compared with those provided by other kernel-based versions of the recursive least-squares algorithm, an online version of the extreme learning machine method, and the persistence time series model. An increase in the number of kernels used in the ensemble system can lead to better results when compared with those of single-kernel models.https://ieeexplore.ieee.org/document/9495822/Multiple kernel learningonline trainingrenewable energywind power forecasting |
| spellingShingle | Erick C. Bezerra Pierre Pinson Ruth P. S. Leao Arthur P. S. Braga A Self-Adaptive Multikernel Machine Based on Recursive Least-Squares Applied to Very Short-Term Wind Power Forecasting IEEE Access Multiple kernel learning online training renewable energy wind power forecasting |
| title | A Self-Adaptive Multikernel Machine Based on Recursive Least-Squares Applied to Very Short-Term Wind Power Forecasting |
| title_full | A Self-Adaptive Multikernel Machine Based on Recursive Least-Squares Applied to Very Short-Term Wind Power Forecasting |
| title_fullStr | A Self-Adaptive Multikernel Machine Based on Recursive Least-Squares Applied to Very Short-Term Wind Power Forecasting |
| title_full_unstemmed | A Self-Adaptive Multikernel Machine Based on Recursive Least-Squares Applied to Very Short-Term Wind Power Forecasting |
| title_short | A Self-Adaptive Multikernel Machine Based on Recursive Least-Squares Applied to Very Short-Term Wind Power Forecasting |
| title_sort | self adaptive multikernel machine based on recursive least squares applied to very short term wind power forecasting |
| topic | Multiple kernel learning online training renewable energy wind power forecasting |
| url | https://ieeexplore.ieee.org/document/9495822/ |
| work_keys_str_mv | AT erickcbezerra aselfadaptivemultikernelmachinebasedonrecursiveleastsquaresappliedtoveryshorttermwindpowerforecasting AT pierrepinson aselfadaptivemultikernelmachinebasedonrecursiveleastsquaresappliedtoveryshorttermwindpowerforecasting AT ruthpsleao aselfadaptivemultikernelmachinebasedonrecursiveleastsquaresappliedtoveryshorttermwindpowerforecasting AT arthurpsbraga aselfadaptivemultikernelmachinebasedonrecursiveleastsquaresappliedtoveryshorttermwindpowerforecasting AT erickcbezerra selfadaptivemultikernelmachinebasedonrecursiveleastsquaresappliedtoveryshorttermwindpowerforecasting AT pierrepinson selfadaptivemultikernelmachinebasedonrecursiveleastsquaresappliedtoveryshorttermwindpowerforecasting AT ruthpsleao selfadaptivemultikernelmachinebasedonrecursiveleastsquaresappliedtoveryshorttermwindpowerforecasting AT arthurpsbraga selfadaptivemultikernelmachinebasedonrecursiveleastsquaresappliedtoveryshorttermwindpowerforecasting |