BCIT

Building Science Graduate Program

Sustainability + Innovation

  • Home
  • About Us
    • Our Team
  • Academic
    • MASc – Master of Applied Science in Building Engineering/Building Science
    • MEng – Master of Engineering in Building Science
    • Graduate Certificate in Building Energy Modelling
    • Part-time Master’s Program
    • Graduate Student Funding
  • Research
    • Building Science Centre of Excellence
    • Research Infrastructure
    • Student Research
  • News
  • Knowledge Base
  • Student Club
    • Events
    • Student Club Contacts
  • Contact

Abstract – Multi-Objective Optimization of High Performance Residential Buildings Using a Genetic Algorithm

Kelvin Y.H. Liu, M.A.Sc. 2014
Supervisor: Dr. Fitsum Tariku

Full thesis available in the BCIT Thesis Repository

Traditional methods of design and construction of residential buildings are common practice, and in most cases, are required by building codes. However, these design practices do not necessarily yield the most optimized designs in terms of cost, environmental impact, and occupant thermal comfort. Typically, the owner or investor hires an architect that designs the building based on the client’s requirements, and then technical designs, such as enclosure and HVAC systems, are tasked to construction and mechanical engineers to satisfy the original design without consideration to energy consumption and environmental impacts. Those who are energy and environmentally conscious rely on an iterative trial and error method using energy simulation tools, and this method consumes much time and resources. To address this problem, this research presents the development and implementation of a simulation-based optimization tool that relies on a genetic algorithm to systematically improve the building design at a conceptual stage based on a set of objective functions. For the purpose of this research, the objective functions include the life-cycle costs, life-cycle global warming potential, and occupant thermal comfort. More specifically, occupant thermal comfort (measured in PPD) acts that the constraint objective.

In this study, a multi-objective optimization genetic algorithm was implemented to find optimal residential building enclosure assemblies that minimizes the life-cycle costs, life-cycle global warming potential, and keeps occupant thermal comfort within check. Based on the design variables and objective functions, a software tool consisting of four modules is used for optimization: the input and input parameter database files; the genetic algorithm optimization software (jEPlus+EA); the energy simulation program (EnergyPlus) and the optimized output files. All required software and simulation programs can be acquired free of charge from the internet, with the exception of proprietary database files such as material and construction assembly libraries.

For validation, the optimization tool is implemented on a benchmark study, which demonstrates its application and capabilities. The benchmark study is based on ANSI/ASHRAE Standard 140-2001 BESTEST calibration and validation test case 600. The optimization results in multiple Pareto optimal solutions that gives the user a detailed look at the trade-off between the objective functions when high performance building systems are used. The optimization tool is then applied to a case study where an actual single family home (Harmony House) is modeled and important building design parameters are identified and discussed.

Join the Discussion

  • The Building Science Graduate Program
  • Scholarship opportunity
  • Student paper recognition by ASHRAE
  • 2019 Buildings XIV International Conference
  • Architect Magazine: The Case of the Missing Energy Model

RSS Journal of Building Physics

  • An error has occurred, which probably means the feed is down. Try again later.

Categories

  • building science
  • Indoor Air Quality
  • Indoor Environmental Quality
  • Scholarships
  • Uncategorized

Archives

  • August 2021
  • January 2019
  • October 2018
  • May 2016
  • April 2016
  • October 2015
  • April 2013
  • February 2013
  • January 2013
  • December 2012

Copyright © 2025 · BCIT · BCIT Commons