期刊
ENERGY AND BUILDINGS
卷 285, 期 -, 页码 -出版社
ELSEVIER SCIENCE SA
DOI: 10.1016/j.enbuild.2023.112899
关键词
Artificial intelligence; Buildings; Deep learning; Occupant-centric control; HVAC temperature setpoint control; Thermal comfort; Indoor air quality
Occupant-Centric Control strategies have gained interest in adjusting building systems, but its application to natural ventilation systems and its impact on air quality have been overlooked. This study proposes an Occupant-Centric Heating and Natural Ventilation Control strategy that utilizes real-time occupant behavior data to improve thermal comfort, reduce energy consumption, and enhance indoor air quality. The strategy showed significant improvements in energy consumption, thermal comfort, and CO2 concentration compared to conventional control strategies.
Recently, Occupant-Centric Control (OCC) strategies have gained mounting interest. Previous studies made use of OCC strategies for adjusting the operation of heating/cooling systems, improving indoor thermal comfort and governing mechanical ventilation systems. However, a very limited number of stud-ies have applied OCC strategies to natural ventilation systems. Further, the feasibility of establishing OCC strategies for controlling indoor thermal comfort, energy use and specifically air quality has received much less attention and investigation. This paper presented an Occupant-Centric Heating and Natural Ventilation Control (OCHNVC) strategy for enhancing indoor thermal comfort, building energy perfor-mance and indoor air quality. Firstly, real-time profiles of occupant behavior and window opening in a case study building were collected using artificial intelligence (AI)-powered cameras and deep vision algorithms. Secondly, shallow artificial-neural-networks predictive models were established for forecast-ing the responses of the studied building to different levels of occupant behavior and window opening behavior. Thirdly, an OCHNVC strategy tailored to the studied room was proposed and applied to the studied room. The strategy could lower heating energy consumption by between 0.6 % and 29.0 % and improve the level of indoor thermal comfort by between 0 % and 58.8 %, relative to a conventional control strategy. Moreover, the conventional window control strategy only maintained indoor CO2 concentra-tions below 1000 ppm for 59.7 % of the period that occupants were within the studied room, while the proposed controller could do so for 89.2 % of the period. Future works shall focus on experimentally deploying the strategy to real buildings and evaluating its performance.(c) 2023 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据