Robert Steven

Existing further and higher education (FHE) buildings urgently need effective retrofit strategies to meet net-zero targets, despite limited historical data. This research integrates LCA, techno-economic assessments, and modelling to identify optimal interventions balancing envelope improvements and low-carbon heating. It considers factors like cost, user comfort, climate change, and evolving energy grids to achieve significant carbon reductions. Considering the inherited occupant thermal comfort benefits of the proposed building retrofit options, the study outlines results that reduce not only operational carbon but also embodied carbon from retrofit work to the end-of-life of the building (60 Years). A fast carbon reduction approach, such as converting to heat pumps, can cut operational energy by 100-118% and GWP by 97%, but increase operational energy costs by 80-101%. Although choosing between specifying synthetic over natural materials can impact project costs, the differences in whole-life carbon emissions provide a 27% and 31% reduction in carbon without considering a change in heating system (and fuel type). Over time, during replacements and maintenance periods, these can play a more relevant role. Switching to lower carbon heat pump refrigerants has shown marginal carbon reductions; however, it is hoped that this technology can find other improvements in lowering its embodied carbon footprint. This research has shown that implementing combined retrofits by integrating heat pumps and envelope improvements, offer optimal cost-effective, emission-cutting solutions for university buildings, enhancing comfort and prioritising natural materials and optimised heating technology. The study provides a detailed comparison of retrofit solutions to inform holistic decarbonisation strategies replicable across different building archetypes.

Sustainable distribution networks are defined here as electricity distribution networks with the inclusion of distributed energy resources (DERs) and their support for the electrification of both transport and heating. Efforts to maximize the inclusion of these elements are not without their challenges, with non-dispatchable DERs such as inverter-connected photovoltaics requiring active control to ensure the safe operation of the network. This chapter briefly discusses these challenges and presents technologies and strategies to overcome them, broadly divided into hardware, software/control and energy markets. These include measurement and control hardware devices installed within the network as well as control strategies including demand response, market-based and inverter-based approaches.

This is the dataset used to generate results presented in "Distributed Energy System Design including Unbalanced AC Power Flow for Large LV Networks with ADMM". Data is presented for the main results section, comparison of central strict and relaxed solvers, constraint violation statistics, z and lambda calculation benchmarks and initial penalty value selection.

Modern power grids have become increasingly complex, with greater uncertaintydue to the widespread integration of renewable energy resources potentiallyleading to higher operating costs. The optimal operation of these networks canbe accomplished using optimal power flow (OPF), a fundamental optimisationtool for power networks with objectives including generation cost minimisation.Whilst the OPF problem itself is not new, quickly solving problems of a practicalscale remains an active research area. Two approaches here are distributedoptimisation and, more recently, machine learning (ML). Distributed optimisationimproves scalability, avoids single points of failure, and enhances userprivacy, whilst ML has the potential to provide solutions significantly fasterthan traditional optimisation methods.The goal of this review is to present approaches which overlap both areas, identifyingcomplementary aspects as well as areas for further exploration. For example,one drawback of the alternating direction method of multipliers (ADMM),a distributed optimisation algorithm, is that it has slow convergence. Several reviewedpapers have mitigated this, using ML to accelerate convergence throughthe prediction of consensus variable values, demonstrating improvements interms of convergence time. Challenges remain, including the generalisation ofresults across different network topologies, something with the potential to beaddressed with additional ML models such as graph neural networks (GNNs).Further areas to explore at the intersection of these two areas are identified, includingaugmented Lagrangian alternating direction inexact Newton (ALADIN)and overlapping Schwarz decomposition optimisation methods and ML modelssuch as GNNs and physics-informed neural networks (PINNs).