Prediction and design optimization of indoor thermal environment and energy consumption in dental treatment space based on artificial neural network and genetic algorithm

Abstract

Studies on occupants’ thermal comfort in healthcare buildings have consistently shown that healthcare staff have a narrower thermal comfort temperature range compared to patients. Ensuring the thermal comfort of healthcare staff is crucial as it is one of the simplest and most effective approaches to improve their work efficiency and reduce medical errors. On the other hand, healthcare buildings are significant energy consumers, with energy consumption 1.6 to 2.0 times higher than typical public buildings. Achieving an optimal balance between thermal comfort and energy efficiency represents a significant challenge. Consequently, it is valuable to investigate this trade-off comprehensively and propose design and operation strategies with the dual objectives of thermal comfort and energy consumption in healthcare buildings.

In recent years, there has been a growing body of research focusing on the optimization of building indoor thermal comfort and energy consumption. Some studies have employed advanced machine learning algorithms to comprehensively optimize these conflicting objectives and identify optimal design and operation strategies. However, the majority of these studies have focused on educational, residential, and office buildings, with relatively limited attention given to healthcare buildings. Compared to other public buildings, healthcare facilities have unique thermal environment requirements and energy usage characteristics. Consequently, optimization results derived from general public buildings may not be directly applicable to healthcare settings. Overall, there is currently a dearth of research that addresses the unique features of healthcare buildings using advanced machine learning algorithms for optimizing thermal comfort and energy consumption.

This research takes Ningbo as the case study city, with six representative treatment spaces from four hospitals selected for investigation, simulation, and analysis of their thermal comfort and energy performance. Based on the analysis of healthcare building characteristics, the study further develops predictive models of design and operation parameters with thermal comfort and energy consumption, and identifies optimal strategies through multi-objective optimization.

The main finding of this study is the development of an adaptive predicted mean vote (aPMV) model specifically for healthcare staff, and the conduction of multi-objective optimization to identify optimal design and operation strategies.The results of the thermal comfort study indicate that healthcare staff tend to prefer cooler working environments. The model used to adjust their actual thermal sensation is as follows: aPMV = PMV / (1 - 0.34PMV) for cooler environments and aPMV = PMV / (1 + 0.128PMV) for warmer environments. Regarding energy consumption, the study emphasizes that the distinctive medical equipment present in healthcare facilities, while requiring a considerable amount of energy, also generates considerable internal heat gains that exert an influence on thermal comfort. The study further proposes specific design recommendations, including cooling, heating, and humidity setpoints, as well as suitable insulation and window-to-wall ratios, through the multi-objective optimization results. It is noteworthy that excessive insulation can hinder nighttime heat dissipation, which negatively impacts both thermal comfort and energy efficiency. Considering day lighting and view, using good insulation materials for roofs and external walls while maintaining relatively high window-to-wall ratio is a suitable approach. In addition, to balance thermal comfort and energy consumption, it is recommended that the cooling, heating, and relative humidity setpoints be set at 24-25 ℃, 17-18.5 ℃, and 50-58%, respectively.

The novelty of this study lies firstly in the development of an aPMV model specifically tailored for healthcare staff, quantifying the adjustments needed to accurately capture the unique thermal comfort requirements of healthcare staff in the Hot Summer Cold Winter climate zone. This adaptation is validated against existing standards for healthcare environment design and demonstrates significant deviations from conventional adaptive models for general public buildings, thereby highlighting the distinctive thermal demands in healthcare settings. Secondly, this research advances the field by conducting a comprehensive multi-objective optimization of thermal comfort and energy consumption specifically for healthcare buildings - a domain rarely addressed in previous studies. The integration of empirical data, predictive modeling and optimization within this specialized context offers practical design and operational strategies. The findings can provide customized recommendations for healthcare buildings in similar climatic zones and architectural forms.

Date of Award	13 Jul 2025
Original language	English
Awarding Institution	University of Nottingham
Supervisor	Wu Deng (Supervisor), Jun Lu (Supervisor) & Tongyu Zhou (Supervisor)