Thesis

Towards an improved representation of building occupant's thermal interaction, integrating detailed occupant thermal models within building simulation

Creator
Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2020
Thesis identifier
  • T15739
Person Identifier (Local)
  • 201563474
Qualification Level
Qualification Name
Department, School or Faculty
Abstract
  • This thesis is concerned with advancing the modelling of the building occupant thermal interaction in building simulation tools. A detailed multi-segment human thermal model has been implemented within the building simulation tool ESP-rand its integrated computational fluid dynamic CFD module. The improvement of ESP-r's building occupant representation in building simulation has been done in three stages. With the complexity of the integration increasing in each stage. In the first stage, a responsive occupant heat load model has been developed and implemented in ESP-r. In this model, the sensible and latent heat loads are regression equations derived from the literature and are a function of operative temperature and metabolic rate. In the second stage, a two-node thermo-physiology model has been developed and implemented that dynamically simulate with the thermal building model.;This ensures that occupant thermal models are responsive to the prevailing conditions and secondly, improves the resolution modelling of occupants and their environment. In addition, clothing adaptation has been considered by implementing a dynamic clothing algorithm. The third stage involved implementing a multi-segment human thermal model within ESP-r and its integrated CFD module. The integration of all three levels of occupant model has been validated with published experimental data. Moreover, each of the three approaches has been demonstrated using example applications. It is hoped that these fully-integrated models of occupant thermo-physiology help advance the modelling of the indoor environment, occupant thermal comfort and building performance prediction within a whole-building simulation.
Advisor / supervisor
  • Kelly, Nicolas James
Resource Type
DOI
Date Created
  • 2020
Former identifier
  • 9912927492902996

Relações

Itens