Multi-objective Optimisation for Existing Houses Retrofit Targeting Net Zero Energy: Framework and Application in Victoria, Australia

Energy efficiency retrofitting of existing buildings, including housing, plays a critical role in enhancing thermal comfort, reducing energy consumption, utility costs, and associated greenhouse gas (GHG) emissions. Despite these benefits, the energy efficiency of existing homes in Australia built before the mid 2000’s remains low, with an average rating of 1.7 stars (on a scale of 0- worst to 10- best) compared to the 7 star-rating required for new homes in states such as Victoria. While there have been attempts by policy makers and some homeowners to increase the level and depth of retrofit being undertaken in Australia, the progress falls short of what is required for a low carbon future and a number of barriers to increasing energy efficiency and adopting low-emission technologies persist. Part of the challenge for scaling up retrofit is the lack of clear information available about how to deliver retrofit across competing priorities such as investment cost and greenhouse gas emissions reduction.Residential buildings account for 8.6% of total energy consumption, 24% of electricity use, and over 10% of emissions in Australia. The majority of Australian homes were constructed before the implementation of minimum energy performance requirements, which means they were designed and built without adequate consideration for climate adaptation and energy efficiency. While recent regulations have introduced and raised minimum energy and performance requirements, existing homes remain exempt from any minimum requirements. Research indicates that retrofitting existing homes is more effective at reducing greenhouse gas emissions than demolition and reconstruction. With the majority of current Australian homes expected to remain in use in 2050, this presents an urgent national challenge requiring immediate attention if wider sustainability goals are to be achieved.Many previous energy retrofit projects have focused on individual measures or “quick” retrofits, which have been difficult to scale up to other retrofit activities or apply to different building archetypes, and few have targeted the ultimate goal of achieving net zero energy outcomes. This study addresses this challenge by developing a comprehensive framework to balance financial constraints with environmental goals with the overarching aim of achieving net zero energy consumption. The main question for this thesis focuses on how multi-objective optimisation can be used to support optimal selection strategies under multiple constraints such as minimising energy demand, investment cost, and greenhouse gas emissions for existing Australian housing.This research presents a 5-step, 3-phase framework that integrates simulation-based multi-objective optimisation using the active archive Non-dominated Sorting Genetic Algorithm II (aNSGA-II) with different decision-making methods to identify optimal residential retrofitting strategies and determine the final trade-off solutions. The framework simultaneously optimises thermal energy demand, investment cost, and greenhouse gas emissions. Developed in Python and coupled with EnergyPlus via the DesignBuilder interface, the framework evaluates a wide range of retrofit options, including building envelope upgrades, heating, ventilation, and air conditioning system improvements, indoor shading, and renewable energy integration. The model was applied to a dominant house type – detached housing – in Australia, testing 24 construction variations in the Melbourne climate zone. Best performance results were found to lift performance from 0 stars to 3.5-4 Star rating. The trade-off results reveal substantial performance gains, with an average 64% improvement in energy efficiency and up to a 78% reduction in greenhouse gas emissions achievable within a retrofit investment of AUD 20,000. The comprehensive multi-objective optimisation-based framework developed in this thesis can be used to evaluate energy demand savings, greenhouse gas emissions reduction potential under investment restrictions across various housing archetypes, as well as in other climate zones. Application of this framework to various housing types can provide policymakers with a robust basis for evaluating the potential for energy improvements and greenhouse gas emissions reduction and serve as a foundation for developing a bottom-up approach to estimate overall retrofit performance. This implication can be applied not only to Australia but also to other countries, contributing to global energy savings and greenhouse gas emissions reduction targets.The outcomes of all RBs in this thesis can help policymakers better understand the potential for performance improvement and define new energy codes and standards for existing housing retrofits in Victoria, Australia. The optimised results are better than the outcomes of previous retrofitting projects in Australia, however, the findings in this thesis remain well below the minimum requirement of 7 stars for new houses in Victoria, indicating that if the government aims to retrofit existing houses to an equivalent minimum star level for new houses, more efficient retrofitting measures should be considered. The results also support recommendations to classify different star-rating targets for retrofitted house energy performance according to the pre-retrofit performance of existing houses, rather than applying a one-size-fits-all regulation to all existing houses. Such approaches used in previous projects were found insufficient to improve all existing houses to the same energy performance level whilst original star ratings differed substantially. Given multiple optimal alternatives tailored to the needs of different stakeholders, this framework can provide the flexibility required by the “performance-based” energy code and benefit private homeowners specifically by offering more transparent information and a wider range of retrofitting package options. The outcomes also provide homeowners with clear insights into the value of their investment in retrofitting options. This creates a win–win situation to which environmental benefits align with government targets, while investment considerations address homeowners’ primary concerns. Such flexibility can further encourage the adoption of retrofitting measures in the residential sector.This thesis offers actionable insights for policymakers, and private homeowners, underscoring the economic feasibility and environmental benefits of retrofitting Australia's existing housing stock to support national carbon reduction targets.