Multi-variable optimization of thermal energy efficiency retrofitting of buildings using static modelling and genetic algorithms – A case study

• Published in: Building and environment 2014-05, Vol.75, p.98-107
• Main Authors: Murray, Sean N., Walsh, Brendan P., Kelliher, Denis, O'Sullivan, D.T.J.
• Format: Article
• Language: English
• Subjects: Applied sciences; Building technical equipments; Buildings; Buildings. Public works; Degree-days simulation; Education and research facilities; Energy efficiency; Energy modelling; Environmental engineering; Exact sciences and technology; Existing building retrofitting; Genetic algorithms; Housing improvement. Restoration; Multi-variable optimization; Space heating; Types of buildings
• ISSN: 0360-1323; 1873-684X
• DOI: 10.1016/j.buildenv.2014.01.011

The retrofitting of existing buildings is an area of research that requires development in order to overcome the ‘rule of thumb’ based approach currently being undertaken. Simulation-based optimization is one approach that can assist consultant engineers, architects and other professionals who undertake retrofit projects. This paper presents a degree-days simulation technique coupled with a genetic algorithms optimization procedure to propose optimal retrofit solutions. The research is applied to a recently retrofitted case-study building. A comparison between the implemented retrofit solution and the simulation-based optimal solution is included to demonstrate the applicability of the research to real-world situations. This research demonstrates the necessity to carry out analysis of a project before retrofit works commence to ensure an optimal approach is taken in accordance with the project specific criteria. •Simulation based optimisation using a static energy modelling approach and the genetic algorithms optimization technique.•Multi-variable optimization in the area of energy efficiency using discrete data sets.•Application of the optimal retrofit decision support tool to a real non-domestic building.•Optimal retrofit solutions identified for differing objective functions and constraint criteria.