Arida, M A (2024) Enhancing thermal comfort in commercial buildings: Contributions of machine learning algorithms for optimizing predictive thermal comfort indices. PhD thesis, North Carolina Agricultural and Technical State University, USA.
Abstract
In recent years, there has been a dramatic increase in the number of energy-efficient structures built in the United States. Optimizing thermal comfort and energy efficiency represents a critical challenge and opportunity within the field of energy-efficient building design. Modern control systems based on thermal comfort models, such as the Predicted Mean Vote (PMV), offer a promising avenue for improving occupant comfort while simultaneously reducing energy consumption. Leveraging machine learning for predictive modeling further advances this field, enabling more precise control and forecasting of indoor environmental conditions. Focusing on these innovations, this research examines the efficacy of PMV-based comfort controls compared to traditional thermostatic controls in commercial buildings through detailed energy modeling. A sensitivity analysis, grounded in the Design of Experiments methodology and augmented by real-world energy consumption data, assesses how variations in building envelope characteristics influence the balance between indoor thermal comfort and energy efficiency. The findings reveal that PMV-based systems significantly improve this balance, outshining conventional temperature-based controls under a variety of conditions.Furthermore, the investigation extends to the predictive modeling of thermal comfort indices, specifically PMV and PPD, employing advanced machine learning techniques such as Artificial Neural Networks (ANN) and Support Vector Machines (SVM). Through the careful collection and preprocessing of data on environmental variables, clothing insulation, metabolic rates, and personal preferences, this research fine-tunes the predictive accuracy of these models. The comparative analysis underscores the superior performance of ANN and SVM in forecasting thermal comfort, highlighting the transformative potential of machine learning in the domain of building environmental comfort.Contributing to both theoretical and practical dimensions, this research explains the advantages of integrating comfort-based control systems with machine learning predictive models. It supports a concept shift towards more sustainable and occupant-centric building management practices, offering valuable insights for the design of energy-efficient, responsive indoor environments. Through this dual approach, the work sets a new benchmark for achieving optimal thermal comfort and in the built environment.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Thesis advisor: | Megri, A |
| Uncontrolled Keywords: | United States; accuracy; building design; built environment; energy consumption; energy efficiency; forecasting; learning; machine learning; thermal comfort; variations |
| Index terms: | predictive modelling, sensitivity analysis, energy modelling, United States, building management, artificial neural network, accuracy, design of experiment, comparative analysis, control system, variation, methodology, commercial building, environmental conditions, machine learning, preference, forecasting, clothing, building design, energy efficiency, indoor environment, comfort, investigation, thermal comfort, built environment, energy consumption, insulation, building envelope, dimension, energy-efficient building |
| Subjects: | sustainable design, Geography, architectural elements, decision-making and reasoning, research methods, environmental hazards, construction type, environmental science, architectural design, monitoring and control, contractual condition, professional development, management, infrastructure and transport systems, environmental engineering, occupational health and safety management, data collection methods, artificial intelligence, energy systems, data analysis and analytics, health monitoring assessment and metrics, prediction and forecasting, building materials, sustainability and energy, modelling and simulation |
| Topics: | Construction Technology, Business Strategy, Information Management, Research Practice, Contract Administration, Digital Applications, Urban Studies, Design Practice, Sustainability, Health and Safety, Geographical Context |
| Descriptive scope: | 4 PCTA |
N.B. Descriptive scope is a count of how many of the five facets of empirical research are indicated by the words used in title, abstract and keywords. It is not intended as a judgement on the research; merely a count of the kind of word we would expect to indicate Phenomenon, Concepts, Theoretical framing, Empirical techniques, Analytical techniques. If all five are present, then a code of “5 PCTEA” will indicate this. If you feel the coding for this record is questionable, we welcome discussion around the terms we matched or the way we categorized them. The facet you would expect may not be coded, or a facet may be coded inappropriately. This can also bear on a larger question, of which facets should be treated as defining in construction management research. Please get in touch, and we will look at it. More details here