A deterministic and probabilistic framework based on corrected wind speed to improve Short-Term wind power forecasting accuracy

The safety and stability of power systems with high wind power penetration are challenged by the inherent variability and uncertainty of wind generation. To improve short-term wind power forecasting accuracy, this study proposes a novel deterministic and probabilistic forecasting framework based on...

Full description

Saved in:
Bibliographic Details
Main Authors: Nuttapat Jittratorn, Chao-Ming Huang, Hong-Tzer Yang
Format: Article
Language:English
Published: Elsevier 2025-09-01
Series:International Journal of Electrical Power & Energy Systems
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S0142061525004077
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849391490184249344
author Nuttapat Jittratorn
Chao-Ming Huang
Hong-Tzer Yang
author_facet Nuttapat Jittratorn
Chao-Ming Huang
Hong-Tzer Yang
author_sort Nuttapat Jittratorn
collection DOAJ
description The safety and stability of power systems with high wind power penetration are challenged by the inherent variability and uncertainty of wind generation. To improve short-term wind power forecasting accuracy, this study proposes a novel deterministic and probabilistic forecasting framework based on corrected wind speed, integrating a wind speed correction method, a bidirectional long short-term memory network, Markov chain-based integration, and jackknife resampling. The novelty lies in combining these components to enhance input data quality and predictive performance. Each component contributes as follows: the wind speed correction reduces input error, improving input accuracy; the bidirectional long short-term memory enhances time-series learning to better capture temporal dependencies; the Markov chain integration improves deterministic forecasts by modeling transition probabilities; and the jackknife resampling quantifies prediction uncertainty to improve probabilistic forecast reliability. Using real-world data from a 3.6 MW wind power plant in Changhua, Taiwan, the proposed model reduces the mean absolute error of forecasted wind speed from 3.93 m/s to 0.78 m/s. The model achieves a mean relative error of 5.89 %, representing a 49.27 % improvement over the model without wind speed correction and outperforming other benchmark deterministic forecasting models. In probabilistic forecasting, the model attains a prediction interval coverage probability of 95.14 %, improving by 11.90 % over the model without wind speed correction and surpassing other benchmark probabilistic models. These results confirm the effectiveness of the proposed approach in enhancing both deterministic and probabilistic forecasting accuracy, thereby supporting more reliable power system operations.
format Article
id doaj-art-1fe7316858f942d78c85aeb45b1efc50
institution Kabale University
issn 0142-0615
language English
publishDate 2025-09-01
publisher Elsevier
record_format Article
series International Journal of Electrical Power & Energy Systems
spelling doaj-art-1fe7316858f942d78c85aeb45b1efc502025-08-20T03:41:04ZengElsevierInternational Journal of Electrical Power & Energy Systems0142-06152025-09-0117011085910.1016/j.ijepes.2025.110859A deterministic and probabilistic framework based on corrected wind speed to improve Short-Term wind power forecasting accuracyNuttapat Jittratorn0Chao-Ming Huang1Hong-Tzer Yang2Department of Electrical Engineering, National Cheng Kung University, Tainan 701, TaiwanDepartment of Electrical Engineering, Kun Shan University, Tainan 710, TaiwanDepartment of Electrical Engineering, National Cheng Kung University, Tainan 701, Taiwan; Corresponding author.The safety and stability of power systems with high wind power penetration are challenged by the inherent variability and uncertainty of wind generation. To improve short-term wind power forecasting accuracy, this study proposes a novel deterministic and probabilistic forecasting framework based on corrected wind speed, integrating a wind speed correction method, a bidirectional long short-term memory network, Markov chain-based integration, and jackknife resampling. The novelty lies in combining these components to enhance input data quality and predictive performance. Each component contributes as follows: the wind speed correction reduces input error, improving input accuracy; the bidirectional long short-term memory enhances time-series learning to better capture temporal dependencies; the Markov chain integration improves deterministic forecasts by modeling transition probabilities; and the jackknife resampling quantifies prediction uncertainty to improve probabilistic forecast reliability. Using real-world data from a 3.6 MW wind power plant in Changhua, Taiwan, the proposed model reduces the mean absolute error of forecasted wind speed from 3.93 m/s to 0.78 m/s. The model achieves a mean relative error of 5.89 %, representing a 49.27 % improvement over the model without wind speed correction and outperforming other benchmark deterministic forecasting models. In probabilistic forecasting, the model attains a prediction interval coverage probability of 95.14 %, improving by 11.90 % over the model without wind speed correction and surpassing other benchmark probabilistic models. These results confirm the effectiveness of the proposed approach in enhancing both deterministic and probabilistic forecasting accuracy, thereby supporting more reliable power system operations.http://www.sciencedirect.com/science/article/pii/S0142061525004077Wind power forecastingWind power prediction intervalsWind speed correction
spellingShingle Nuttapat Jittratorn
Chao-Ming Huang
Hong-Tzer Yang
A deterministic and probabilistic framework based on corrected wind speed to improve Short-Term wind power forecasting accuracy
International Journal of Electrical Power & Energy Systems
Wind power forecasting
Wind power prediction intervals
Wind speed correction
title A deterministic and probabilistic framework based on corrected wind speed to improve Short-Term wind power forecasting accuracy
title_full A deterministic and probabilistic framework based on corrected wind speed to improve Short-Term wind power forecasting accuracy
title_fullStr A deterministic and probabilistic framework based on corrected wind speed to improve Short-Term wind power forecasting accuracy
title_full_unstemmed A deterministic and probabilistic framework based on corrected wind speed to improve Short-Term wind power forecasting accuracy
title_short A deterministic and probabilistic framework based on corrected wind speed to improve Short-Term wind power forecasting accuracy
title_sort deterministic and probabilistic framework based on corrected wind speed to improve short term wind power forecasting accuracy
topic Wind power forecasting
Wind power prediction intervals
Wind speed correction
url http://www.sciencedirect.com/science/article/pii/S0142061525004077
work_keys_str_mv AT nuttapatjittratorn adeterministicandprobabilisticframeworkbasedoncorrectedwindspeedtoimproveshorttermwindpowerforecastingaccuracy
AT chaominghuang adeterministicandprobabilisticframeworkbasedoncorrectedwindspeedtoimproveshorttermwindpowerforecastingaccuracy
AT hongtzeryang adeterministicandprobabilisticframeworkbasedoncorrectedwindspeedtoimproveshorttermwindpowerforecastingaccuracy
AT nuttapatjittratorn deterministicandprobabilisticframeworkbasedoncorrectedwindspeedtoimproveshorttermwindpowerforecastingaccuracy
AT chaominghuang deterministicandprobabilisticframeworkbasedoncorrectedwindspeedtoimproveshorttermwindpowerforecastingaccuracy
AT hongtzeryang deterministicandprobabilisticframeworkbasedoncorrectedwindspeedtoimproveshorttermwindpowerforecastingaccuracy