A deep reinforcement learning control method for multi-zone precooling in commercial buildings

•We propose a Markov Decision Process framework for optimal precooling control, enabling the precise allocation of limited cooling capacity across multiple thermal zones and solving the energy consumption problem caused by precooling discrepancies between zones.•Thermal transmission between adjacent...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	Applied thermal engineering Ročník 260; s. 124987
Hlavní autori:	Fan, Yuankang, Fu, Qiming, Chen, Jianping, Wang, Yunzhe, Lu, You, Liu, Ke
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	Elsevier Ltd 01.02.2025
Predmet:	Deep reinforcement learning HVAC systems Multi-zone commercial buildings Parametric analysis Precooling control Deep reinforcement learning Multi-zone commercial buildings HVAC systems Parametric analysis Precooling control
ISSN:	1359-4311
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Popis
Shrnutí:	•We propose a Markov Decision Process framework for optimal precooling control, enabling the precise allocation of limited cooling capacity across multiple thermal zones and solving the energy consumption problem caused by precooling discrepancies between zones.•Thermal transmission between adjacent zones in the model and introduce precooling time correction to achieve a more accurate simulation of the precooling process.•By combining precooling control with the Deep Q-Network algorithm, the method effectively handles complex environmental changes, significantly improving precooling performance and showing strong energy-saving potential.•Comprehensive comparative experiments confirm the superiority of this method, with the study indicating that weather conditions have the most significant impact on short-term precooling performance, followed by house thermal properties and cooling conditions. In commercial buildings, implementing precooling measures before office hours in summer can effectively meet the thermal comfort needs of employees. However, in multi-zone environments, differences in the cooling rates between regions often exacerbate the heat transfer interference between zones, increasing the complexity of the precooling system and leading to energy waste with limited cooling capacity. To overcome these challenges, we have developed a novel multi-zone precooling control method, which integrates deep reinforcement learning (DRL) to optimize the heat transfer process by adjusting the Air Handling Units (AHUs) valve openings, thus achieving uniform precooling across the building. Comparisons with traditional precooling control methods demonstrate the effectiveness of the proposed method. The results show that, under conventional conditions, compared with the rule-based control (RBC) and proportional integral derivative (PID) methods, the precooling time is reduced by 11.4% and 5.8%, respectively, the complexity of heat transfer is reduced by 77.6% and 64.1%, and energy consumption is reduced by 14.5% and 9.3%. In addition, the study analyzes the influence of environmental parameters on precooling optimization. The findings indicate that weather conditions have the most substantial impact on short-term precooling performance, followed by building thermal performance and cooling conditions.
ISSN:	1359-4311
DOI:	10.1016/j.applthermaleng.2024.124987