Award Date

December 2023

Degree Type

Thesis

Degree Name

Master of Science in Engineering (MSE)

Department

Civil and Environmental Engineering and Construction

First Committee Member

Jeehee Lee

Second Committee Member

Pramen . Shrestha

Third Committee Member

Jee Woong Park

Fourth Committee Member

Seungman Park

Number of Pages

71

Abstract

Building occupants experience varying thermal comforts at static indoor set-point temperatures due to variances in preferences, needs, and activities, both within and among individuals. Despite advancement in smart building control systems, many building occupants remain dissatisfied with indoor temperatures. This study developed and created an optimization model to predict the optimized indoor temperature preferences in multiple-occupancy settings. To achieve this, personalized thermal comfort models for each occupant in a multiple occupancy space were developed. The personal comfort models serve as the foundation for predicting their individual thermal preference and developing indoor temperature optimization model. When predicting the individual thermal preference, a personalized thermal comfort model incorporates both physiological signals (such as heart rate, skin temperature, and skin conductance) and environmental parameters (such as air temperature, relative humidity, and CO2 levels). Data was collected through an experiment conducted in an office building and was used to train the model using different machine learning algorithms to identify the most reliable personalized predictive model. The results show that each person possesses an effective classification model that accurately predicts their preferences in personal comfort models. The developed personal comfort models served as the crucial inputs for optimization model. After developing these personal comfort models, optimization methods were employed using three different probability approaches, including average of the three subjects predicted probabilities, majority values selection, and posterior probability approach in Naive Bayes classifier to determine the optimized indoor temperature and thermal preferences based on three different thermal preferences predictions from personal comfort models. The proposed indoor optimization model will contribute to providing a thermally comfortable indoor environment while minimizing unnecessary energy consumption.

Keywords

Classification model; Multiple occupancy; Optimization model; Personal thermal comfort; Thermal comfort; Thermal preference

Disciplines

Civil Engineering

Degree Grantor

University of Nevada, Las Vegas

Language

English

Rights

IN COPYRIGHT. For more information about this rights statement, please visit http://rightsstatements.org/vocab/InC/1.0/

Download


Share

COinS