Multi-aspect renewable energy forecasting

TUCKER-CLUS: Multi-aspect renewable energy forecasting

A new method based on the Tucker tensor decomposition, capable of extracting a new feature space for the multi-plant energy forecasting task.

For evaluation purposes, we have investigated the performance of predictive clustering trees with the new feature space, compared to the original feature space, in three renewable energy datasets. The results are favorable for the proposed method, also when compared with state-of-the-art algorithms.

The performance of the method has been tested on the 24-hour ahead multi-plant energy forecasting task. Additional details can be found on the research paper referenced below.

Publications:

R. Corizzo, M. Ceci, H. Fanaee-T, J. Gama: Multi-aspect renewable energy forecasting, Information Sciences (DOI: 10.1016/j.ins.2020.08.003), https://www.sciencedirect.com/science/article/pii/S0020025520307611

Citation:

@article{corizzo2021multi,  title={Multi-aspect renewable energy forecasting},  author={Corizzo, Roberto and Ceci, Michelangelo and Fanaee-T, Hadi and Gama, Joao},  journal={Information Sciences},  volume={546},  pages={701--722},  year={2021},  publisher={Elsevier}}

数据集TUCKER-CLUS：多维度可再生能源预测

本研究提出一种基于Tucker张量分解（Tucker tensor decomposition）的创新方法，可针对多场站能源预测任务提取专属特征空间。

为开展模型验证，我们在三个可再生能源数据集上，对比了基于新特征空间与原始特征空间的预测聚类树（predictive clustering trees）模型性能。实验结果显示，所提方法不仅表现更优，且相较于当前前沿算法仍具备显著优势。

本方法的性能已在24小时前瞻多场站能源预测任务中完成验证，更多细节可参阅下文列出的研究论文。

发表文献：
R. Corizzo、M. Ceci、H. Fanaee-T、J. Gama：《多维度可再生能源预测》，刊载于《信息科学（Information Sciences）》（DOI: 10.1016/j.ins.2020.08.003），链接：https://www.sciencedirect.com/science/article/pii/S0020025520307611

引用格式：
@article{corizzo2021multi, title={Multi-aspect renewable energy forecasting}, author={Corizzo, Roberto and Ceci, Michelangelo and Fanaee-T, Hadi and Gama, Joao}, journal={Information Sciences}, volume={546}, pages={701--722}, year={2021}, publisher={Elsevier}}