| Autor: |
Su An, Honglei Wang, Xufeng Yuan |
| Jazyk: |
angličtina |
| Rok vydání: |
2020 |
| Předmět: |
|
| Zdroj: |
IEEE Access, Vol 8, Pp 216566-216579 (2020) |
| Druh dokumentu: |
article |
| ISSN: |
2169-3536 |
| DOI: |
10.1109/ACCESS.2020.3041709 |
| Popis: |
Under the increasingly stringent environmental regulations, the installed capacity of distributed renewable energy is increasing rapidly. How to fully absorb renewable energy without affecting power grid security and ensuring power quality is the key problem to be solved. The rapid development of energy production, conversion and storage equipment of prosumer makes the operation of micro energy grid possible. Therefore, according to the energy characteristics of different prosumer, this article divides the micro energy grid into two types: micro energy grid of electricity selling and micro energy grid of energy supplying. The constrained nonlinear optimization method is used to solve the optimization problem of this high-dimensional nonlinear system with time delay. In this article, the comparative study of optimal operation control between the two kinds of micro energy grid is carried out. In order to deal with the uncertainty of distributed renewable energy output and load forecasting deviation, this article proposes a real time adaptive dynamic optimization control strategy based on deep learning. The strategy uses deep learning technology to pretrain the action network, so as to learn the optimal operation behavior of micro energy grid. Finally, the online simulation of micro energy grid for consecutive days is used to verify the correctness and real-time performance of the algorithm. |
| Databáze: |
Directory of Open Access Journals |
| Externí odkaz: |
|
