Two‐stage short‐term wind power forecasting algorithm using different feature-learning models
Two-stage ensemble-based forecasting methods have been studied extensively in the wind power forecasting field. However, deep learning-based wind power forecasting studies have not investigated two aspects. In the first stage, different learning structures considering multiple inputs and multiple ou...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co. Ltd.
2021-07-01
|
| Series: | Fundamental Research |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S266732582100100X |
| _version_ | 1797976326569721856 |
|---|---|
| author | Jiancheng Qin Jin Yang Ying Chen Qiang Ye Hua Li |
| author_facet | Jiancheng Qin Jin Yang Ying Chen Qiang Ye Hua Li |
| author_sort | Jiancheng Qin |
| collection | DOAJ |
| description | Two-stage ensemble-based forecasting methods have been studied extensively in the wind power forecasting field. However, deep learning-based wind power forecasting studies have not investigated two aspects. In the first stage, different learning structures considering multiple inputs and multiple outputs have not been discussed. In the second stage, the model extrapolation issue has not been investigated. Therefore, we develop four deep neural networks for the first stage to learn data features considering the input-and-output structure. We then explore the model extrapolation issue in the second stage using different modeling methods. Considering the overfitting issue, we propose a new moving window-based algorithm using a validation set in the first stage to update the training data in both stages with two different moving window processes. Experiments were conducted at three wind farms, and the results demonstrate that the model with single-input–multiple-output structure obtains better forecasting accuracy compared to existing models. In addition, the ridge regression method results in a better ensemble model that can further improve forecasting accuracy compared to existing machine learning methods. Finally, the proposed two-stage forecasting algorithm can generate more accurate and stable results than existing algorithms. |
| first_indexed | 2024-04-11T04:49:10Z |
| format | Article |
| id | doaj.art-bf1587cd1e1f4f8e8158ca25f10aed5d |
| institution | Directory Open Access Journal |
| issn | 2667-3258 |
| language | English |
| last_indexed | 2024-04-11T04:49:10Z |
| publishDate | 2021-07-01 |
| publisher | KeAi Communications Co. Ltd. |
| record_format | Article |
| series | Fundamental Research |
| spelling | doaj.art-bf1587cd1e1f4f8e8158ca25f10aed5d2022-12-27T04:42:03ZengKeAi Communications Co. Ltd.Fundamental Research2667-32582021-07-0114472481Two‐stage short‐term wind power forecasting algorithm using different feature-learning modelsJiancheng Qin0Jin Yang1Ying Chen2Qiang Ye3Hua Li4School of Computer Science and Technology, Harbin Institute of Technology, Heilongjiang 150000, ChinaSchool of Computer Science and Technology, Harbin Institute of Technology, Heilongjiang 150000, ChinaDepartment of Management Science and Engineering, School of Management, Harbin Institute of Technology, Heilongjiang 150000, China; Corresponding author.Department of Management Science and Engineering, School of Management, Harbin Institute of Technology, Heilongjiang 150000, ChinaDepartment of Mechanical and Industrial Engineering, Texas A&M University-Kingsville, Kingsville, Texas 78363, USATwo-stage ensemble-based forecasting methods have been studied extensively in the wind power forecasting field. However, deep learning-based wind power forecasting studies have not investigated two aspects. In the first stage, different learning structures considering multiple inputs and multiple outputs have not been discussed. In the second stage, the model extrapolation issue has not been investigated. Therefore, we develop four deep neural networks for the first stage to learn data features considering the input-and-output structure. We then explore the model extrapolation issue in the second stage using different modeling methods. Considering the overfitting issue, we propose a new moving window-based algorithm using a validation set in the first stage to update the training data in both stages with two different moving window processes. Experiments were conducted at three wind farms, and the results demonstrate that the model with single-input–multiple-output structure obtains better forecasting accuracy compared to existing models. In addition, the ridge regression method results in a better ensemble model that can further improve forecasting accuracy compared to existing machine learning methods. Finally, the proposed two-stage forecasting algorithm can generate more accurate and stable results than existing algorithms.http://www.sciencedirect.com/science/article/pii/S266732582100100XWind power forecastingDeep neural networksEnsemble learningExtrapolation |
| spellingShingle | Jiancheng Qin Jin Yang Ying Chen Qiang Ye Hua Li Two‐stage short‐term wind power forecasting algorithm using different feature-learning models Fundamental Research Wind power forecasting Deep neural networks Ensemble learning Extrapolation |
| title | Two‐stage short‐term wind power forecasting algorithm using different feature-learning models |
| title_full | Two‐stage short‐term wind power forecasting algorithm using different feature-learning models |
| title_fullStr | Two‐stage short‐term wind power forecasting algorithm using different feature-learning models |
| title_full_unstemmed | Two‐stage short‐term wind power forecasting algorithm using different feature-learning models |
| title_short | Two‐stage short‐term wind power forecasting algorithm using different feature-learning models |
| title_sort | two stage short term wind power forecasting algorithm using different feature learning models |
| topic | Wind power forecasting Deep neural networks Ensemble learning Extrapolation |
| url | http://www.sciencedirect.com/science/article/pii/S266732582100100X |
| work_keys_str_mv | AT jianchengqin twostageshorttermwindpowerforecastingalgorithmusingdifferentfeaturelearningmodels AT jinyang twostageshorttermwindpowerforecastingalgorithmusingdifferentfeaturelearningmodels AT yingchen twostageshorttermwindpowerforecastingalgorithmusingdifferentfeaturelearningmodels AT qiangye twostageshorttermwindpowerforecastingalgorithmusingdifferentfeaturelearningmodels AT huali twostageshorttermwindpowerforecastingalgorithmusingdifferentfeaturelearningmodels |