Name
Title | ||
|---|---|---|
Study of Features and Number of Time Steps to Improve Hourly Solar Power Forecasting via LSTM. |
Abstract | ||
|---|---|---|
The number of time steps used for machine learning methods on the hourly prediction of solar photovoltaic (PV) generation is typically set to one. In this paper, we investigate the larger-than-one time steps for long short-term memory (LSTM)-based method to improve the accuracy of prediction with a limited dataset. Besides, we analyze the contribution of all available features toward a different number of time steps and select proper features for better training. By using real-world dataset, the experimental results show that the proposed method can effectively outperform the conventional one whilst using the same dataset. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1109/GCCE46687.2019.9015344 | GCCE |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
0 | 6 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Li Cho | 1 | 0 | 0.34 |
| Chau-Yun Hsu | 2 | 0 | 0.34 |
| Chun-Wei Li | 3 | 0 | 0.34 |
| Wei-Chieh Chen | 4 | 0 | 0.34 |
| Hong-Ren Wang | 5 | 0 | 0.34 |
| Chang-Ping Lin | 6 | 0 | 0.34 |