A hybrid neural network model for short-term wind speed forecasting based on decomposition, multi-learner ensemble, and adaptive multiple error corrections,Renewable Energy

当前位置： X-MOL 学术 › Renew. Energy › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

A hybrid neural network model for short-term wind speed forecasting based on decomposition, multi-learner ensemble, and adaptive multiple error corrections
Renewable Energy ( IF 8.7 ) Pub Date : 2021-03-01 , DOI: 10.1016/j.renene.2020.11.002
Hui Liu , Rui Yang , Tiantian Wang , Lei Zhang

Abstract Under the dual stimulus of the new energy demand and the increasing competitiveness of wind energy, the construction of wind speed prediction models began to be placed in a position that cannot be ignored. To overcome the challenges brought by wind speed fluctuations to wind speed forecasting, this paper proposes a novel hybrid wind speed forecasting deep model. The model has three modules, including data preprocessing, multi-learner ensemble, and adaptive multiple error correction. We used four real wind series in Xinjiang, China to verify the performance of the model. The results of the case study show that: (a) The proposed hybrid deep model for wind speed forecasting is superior to several state-of-the-art models in terms of both forecasting stability and forecasting accuracy; (b) The proposed hybrid deep model is excellent in multi-step forecasting, taking the site #1 as an example, the MAEs of the proposed model are 0.0250 m/s, 0.0417 m/s, and 0.0570 m/s, respectively.

更新日期：2021-03-01

点击分享查看原文

点击收藏

阅读更多本刊最新论文本刊介绍/投稿指南

全部期刊列表>>