Please use this identifier to cite or link to this item:
Title: Information fusion for HVAC application
Authors: Leong, Wei Jie
Keywords: DRNTU::Engineering
Issue Date: 2014
Abstract: The developments of pervasive sensors for Heating, Ventilation and Air Conditioning (HVAC) process has been attracting increasing attention over the years. Information from these sensors carries complex interrelationships between the thermodynamic parameter of HVAC. Therefore, the extraction of key relationships from these pervasive sensors that drive the heat exchange phenomena, including their interactions with the environment and users are to be discovered. Information fusion from multi-modal data source, ASHRAE RP-884 Adaptive Model of Thermal Comfort and Performance with simulations using MATLAB® is proposed to analyse the heat exchange phenomena. Feed forward neural networks model, Extreme Learning Machines (ELM) is used to analysis the testing accuracy based on the target values and its variables. Correlations of variables with PMV and ASHRAE shows different outcome with additional variables through ELM testing accuracy. This study shows how correlations were established using ELM with given data variables and randomly generated variables. Six primary variables of predicted mean vote (PMV) index calculation by Fanger were used from RP-884 data and randomly generated variables. Addition of available data variables from RP-884 is used to discover the relationship of PMV and ASHRAE thermal sensation scale. The target values used for the ELM technique are PMV and ASHARAE thermal sensational scale. ELM results shown that the testing accuracy does not necessary improve with additional variables
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
FYP Report A4147-131.pdf
  Restricted Access
565.55 kBAdobe PDFView/Open

Page view(s)

Updated on Jan 18, 2021


Updated on Jan 18, 2021

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.