An indoor thermal comfort model for group thermal comfort prediction based on K-means++ algorithm

•K-means++ algorithm was first used to build a thermal comfort model without occupants’ feedback.•Even with small dataset, the proposed model achieved high accurancy.•Mahalanobis distance algorithm was used for outlier test, reducing errors and avoiding excessive reduction of data set.•The model cou...

Full description

Saved in:
Bibliographic Details
Published in:Energy and buildings Vol. 327; p. 115000
Main Authors: Liu, Ying, Li, Xiangru, Sun, Cheng, Dong, Qi, Yin, Qing, Yan, Bin
Format: Journal Article
Language:English
Published: Elsevier B.V 15.01.2025
Subjects:
Indoor thermal comfort
K-means++ clustering algorithm
Machine learning
Prediction model
ANN
K-means++ clustering algorithm
CLO
Prediction model
KNN
WGBT
HR
DBI
SVM
DT
SC
Indoor thermal comfort
CC
SSE
TOA
IQR
EEM
TA
TSK
TC
NB
RF
RH
RI
Machine learning
VEL
TP
MET
TR
TS
ISSN:0378-7788
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•K-means++ algorithm was first used to build a thermal comfort model without occupants’ feedback.•Even with small dataset, the proposed model achieved high accurancy.•Mahalanobis distance algorithm was used for outlier test, reducing errors and avoiding excessive reduction of data set.•The model could effectively predict group thermal comfort by using only three parameters: Clo, Ta, and RH. Predicting indoor thermal comfort plays an essential role in controlling energy consumption in buildings. Existing studies have used supervised machine learning to predict thermal comfort, which were more accurate than traditional models. However, these models required occupants’ subjective feedback for model training, which reduced the accuracy of the model. In this study, a prediction model that didn’t require feedback was proposed for the first time using the K-means++ algorithm based on the ASHRAE Global Thermal Comfort Database II. Firstly, the data quality was improved through feature selection, dimensional processing, and feature weighting. Then the influence of different outlier judgment methods, feature weight and data set size on model accuracy were compared. Finally, the K-means++ algorithm was applied for thermal comfort clustering analysis. The result showed that the model with an accuracy higher than 90 % could be constructed using only three factors (CLO, TA, RH), and the proposed model could predict indoor group thermal comfort reliably, and provide a foundation for the indoor thermal sensation evaluation.
ISSN:0378-7788
DOI:10.1016/j.enbuild.2024.115000