Collaborative optimal operation control of HVAC systems based on multi-agent
The HVAC system of public buildings, as a thermostatically controlled load, accounting for a relatively significant proportion of building energy consumption. Therefore, it is necessary to optimize energy efficient of HVAC systems in public buildings. Nevertheless, the complication of HVAC systems i...
Сохранить в:
| Главные авторы: | , , , , , , |
|---|---|
| Формат: | Статья |
| Язык: | английский |
| Опубликовано: |
Frontiers Media S.A.
2025-07-01
|
| Серии: | Frontiers in Energy Research |
| Предметы: | |
| Online-ссылка: | https://www.frontiersin.org/articles/10.3389/fenrg.2025.1609210/full |
| Метки: |
Добавить метку
Нет меток, Требуется 1-ая метка записи!
|
| Итог: | The HVAC system of public buildings, as a thermostatically controlled load, accounting for a relatively significant proportion of building energy consumption. Therefore, it is necessary to optimize energy efficient of HVAC systems in public buildings. Nevertheless, the complication of HVAC systems is on the rise. As a consequence, the computing efficiency of optimization algorithms is relatively low, posing challenges for real-time optimal operation control. Hence, there is an immediate requirement to boost both the energy efficiency of the system and the computing efficiency in order to strengthen the system’s robustness. In this paper, a collaborative optimization approach based on multi-agent is initially put forward to address the overall optimization issue (OOI) of a complicated HVAC system. The OOI is disintegrated into numerous sub-optimization issues within the multi-agent structure. These sub-issues take into account the interaction features among components. By doing so, the complication of the OOI within HVAC systems is effectively decreased. Secondly, the adaptive hybrid-artificial fish swarm algorithm (AH-AFSA) is proposed for solving optimization issues with mixed decision variables. Finally, the effectiveness of the proposed method is verified by an arithmetic example. The analysis reveals that the proposed approach is capable of reducing power consumption by 18.9% and the computation time for each operation condition is 12.2 s, which saves about 54% of time cost compared with the centralized method, and can enhance the computing efficiency of the optimization approach for a complicated HVAC system while reducing power consumption. |
|---|---|
| ISSN: | 2296-598X |