期刊
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
卷 17, 期 1, 页码 300-310出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TII.2020.2987096
关键词
Task analysis; Predictive models; Autoregressive processes; Computational modeling; Data models; Information management; Prediction algorithms; Echo state neural network; multi-tasking model; renewable energy; smart grid; solar irradiance forecasting
类别
资金
- National Natural Science Foundation of China [61803054]
- Fundamental Research Funds for the Central Universities [2019CDQYZDH030]
- Graduate Scientific Research and Innovation Foundation of Chongqing, China [CYB18064, TII-19-4856]
This article investigates multi-timescale solar irradiance forecast using a multi-task learning mechanism, proposing a new MTS prediction framework and MTS-ESN model. Simulation results show that MTS-ESN achieves promising performance at both hourly and daily levels, outperforming the single-timescale ESN model.
Solar irradiance forecast is closely related with efficiency and reliability of renewable energy systems. Multi-timescale irradiance forecast is a new and efficient way to simultaneously predict solar energy generation on different timescales for hierarchical decision making. This article newly adopts the multi-task learning mechanism to study the multi-timescale forecast for improving accuracy and computational efficiency. A novel multi-timescale (MTS) prediction framework is presented to fulfill the multi-task application, and echo state network (ESN) is studied in the proposed MTS framework. The multi-timescale ESN (MTS-ESN) is proposed to enhance the information sharing among correlated tasks. Simulation results of hourly solar data demonstrate that the proposed MTS-ESN could achieve promising performance at both hourly and daily level in parallel. The MTS-ESN outperforms the single-timescale ESN (STS-ESN), which indicates the information sharing in the multi-task learning is effective in this application.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据