Siberian Federal University 
- Issue
- Journal of Siberian Federal University. Engineering & Technologies. 2024 17 (4)
- Authors
- Engel, Ekaterina A.; Engel, Nikita E.
- Contact information
- Engel, Ekaterina A.: Khakas State University Abakan, Russian Federation; ; Engel, Nikita E.: Khakas State University Abakan, Russian Federation
- Keywords
- UML; forecasting of a solar power plant production; recurrent neural networks; attention mechanism; modified fuzzy neural network; UML
- Abstract
The forecast of the generated electricity of a solar power plant allows efficient and safe management of electrical grid that include solar power plants. The day-ahead market buys at penalty rates electricity from solar power plants, deviating by more than 5 % of the maximum solar power plant capacity from the provided hourly day-ahead market layout of electricity generated by the solar power plant. An analysis of existing software showed the lack of available software for effectively forecasting the production of a solar power plant. In this study, an indirect forecasting intelligent technology of a solar power plant production based on a modified fuzzy neural network with an attention mechanism was developed, tested and implemented. The UML class diagram and block-modular architecture of the indirect forecasting intelligent technology of a solar power plant production have been developed. This block-modular architecture provides flexibility and easy modification of the indirect forecasting intelligent technology of a solar power plant production. The approval of the indirect forecasting intelligent technology of a solar power plant production reflects its effective, robust results and the feasibility of its use for automatic generation of day-ahead market layouts
- Pages
- 464–473
- EDN
- JKWKAW
- Paper at repository of SibFU
- https://elib.sfu-kras.ru/handle/2311/153186
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).