• 2013-2022close
• UK Research and Innovation close
• UKRI|EPSRC close
• 2018 close

Computational fluid dynamics (CFD) is fundamental to modern engineering design, from aircraft and cars to household appliances. It allows the behaviour of fluids to be computationally simulated and new designs to be evaluated. Finding the best design is nonetheless very challenging because of the vast number of designs that might be explored. Computational optimisation is a crucial technique for modern science, commerce and industry. It allows the parameters of a computational model to be automatically adjusted to maximise some benefit and can reveal truly innovative solutions. For example, the shape of an aircraft might be optimised to maximise the computed lift/drag ratio. A very successful suite of methods to tackle optimisation problems are known as evolutionary algorithms, so-called because they are inspired by the way evolutionary mechanisms in nature optimise the fitness of organisms. These algorithms work by iteratively proposing new solutions (shapes of the aircraft) for evaluation based upon recombinations and/or variations of previously evaluated solutions and, by retaining good solutions and discarding poorly performing solutions, a population of optimised solutions is evolved. An obstacle to the use of evolutionary algorithms on very complex problems with many parameters arises if each evaluation of a new solution takes a long time, possibly hours or days as is often the case with complex CFD simulations. The great number of solutions (typically several thousands) that must be evaluated in the course of an evolutionary optimisation renders the whole optimisation infeasible. This research aims to accelerate the optimisation process by substituting computationally simpler, dynamically generated "surrogate" models in place of full CFD evaluation. The challenge is to automatically learn appropriate surrogates from a relatively few well-chosen full evaluations. Our work aims to bridge the gap between the surrogate models that work well when there are only a few design parameters to be optimised, but which fail for large industry-sized problems. Our approach has several inter-related aspects. An attractive, but challenging, avenue is to speed up the computational model. The key here is that many of these models are iterative, repeating the same process over and over again until an accurate result is obtained. We will investigate exploiting partial information in the early iterations to predict the accurate result and also the use of rough early results in place of the accurate one for the evolutionary search. The other main thrust of this research is to use advanced machine learning methods to learn from the full evaluations how the design parameters relate to the objectives being evaluated. Here we will tackle the computational difficulties associated with many design parameters by investigating new machine learning methods to discover which of the many parameters are the relevant at any stage of the optimisation. Related to this is the development of "active learning" methods in which the surrogate model itself chooses which are the most informative solutions for full evaluation. A synergistic approach to integrate the use of partial information, advanced machine learning and active learning will be created to tackle large-scale optimisations. An important component of the work is our close collaboration with partners engaged in real-world CFD. We will work with the UK Aerospace Technology Institute and QinetiQ on complex aerodynamic optimisation, with Hydro International on cyclone separation and with Ricardo on diesel particle tracking. This diverse range of collaborations will ensure research is driven by realistic industrial problems and builds on existing industrial experience. The successful outcome of this work will be new surrogate-assisted evolutionary algorithms which are proven to speed up the optimisation of full-scale industrial CFD problems.

Diffusion-NMR is a powerful technique with applications that span from material science to medicine. With the characteristic non-invasiveness and non-harmfulness of Nuclear Magnetic Resonance (NMR), the technique can infer information on molecular diffusion in various media. Diffusion-NMR has applications that range from analytic sciences where it can be used for example to sort out complex molecular mixtures according to different diffusion coefficients up to medicine where it is used to obtain contrast between biological tissues within which molecules have different diffusion properties. Porous media, which are ubiquitous in nature with examples including rocks, bones, wood etc. are perhaps the most suitable systems to be characterised through diffusion-NMR. And indeed scientific literature contains numerous examples of such investigations. However, because conventional NMR signals last typically for only up to a few seconds, diffusion-NMR studies have some limitations. The measurements of diffusion are done on a microscopic level by registering the changes in the intensity of an NMR signal as molecules diffuse in solution. The longer the diffusion time the farther the molecules diffuse, so that the the registered change in signal is more dramatic and the more accurate the measurement. Molecular diffusion is affected by the microscopic structure of the material a molecule diffuses within, therefore diffusion-NMR is a very sensitive tool to probe micro-structures, hence its great utility in porous media investigations. However its sensitivity to dimensions is directly linked to the timescale explored i.e. to the available diffusion time. Limitations to diffusion time due to the lifetime of conventional NMR signals therefore restrict the technique to geometries within 100 micrometers. Since many interesting porous structures have pore larger than 100 micrometers the technique cannot probe pore connectivity in those systems hence cannot provide a measure of tortuosity which is of instrumental importance in many areas including oil engineering and battery development. In the past 10 years I have been investigating the topic of long-lived spin states which are particular configurations of nuclear spin states displaying very long lifetimes that can reach in some cases even an hour length. This lifetime extension can be used in diffusion-NMR to prolong the diffusion time and obtain a better accuracy in diffusion measurement plus the possibility to access information on pore connectivity and hence measure tortuosity. This proposal deals with the development and assessment of methodology that exploit long-lived states to expand the accessible diffusion time in diffusion-NMR experiments thus giving access to measurement of tortuosity, macroscopic compartmentation and diffusion anisotropy in porous media. The main outcomes of this research are: 1. molecular probes of diffusion that support long-lived states to give access to very long diffusion times 2. NMR methodology to measure diffusion by encoding positional information on long-lived spin states 3. a simulation procedure for simulation of complex NMR experiments on porous systems 4. measurements of tortuosity, anisotropic diffusion and macrostructures in porous media The proposed methodology is expected to benefit laboratories and industries with interests in characterising porous material and/or developing new materials (an increase in the tortuosity of lithium batteries' electrodes during electrochemical cycling is thought to be partially responsible for the observed reduction of performances, for example). Diffusion anisotropy is of particular interest in MRI where it is exploited in diffusion-tensor-imaging, a technique that uses diffusion anisotropy to map the direction of fibres in the body: methods and procedures developed in this project have the potential to impact this area too.

The use of ultrasound as a diagnostic imaging tool is well known, particularly during pregnancy where ultrasound is used to create images of the developing foetus. In recent years, a growing number of therapeutic applications of ultrasound have also been demonstrated. The goal of therapeutic ultrasound is to modify the function or structure of the tissue, rather than produce an anatomical image. This is possible because the mechanical vibrations caused by the ultrasound waves can affect tissue in different ways, for example, by causing the tissue to heat up, or by generating internal radiation forces that can agitate the cells or tissue scaffolding. These ultrasound bioeffects offer a huge potential to develop new ways to treat major diseases such as cancer, to improve the delivery of drugs while minimising side-effects, and to treat a wide spectrum of neurological and psychiatric conditions. The fundamental challenge shared by all applications of therapeutic ultrasound is that the ultrasound energy must be delivered accurately, safely, and non-invasively to the target region within the body. This is difficult because bones and other tissue interfaces can severely distort the shape of the ultrasound beam. This has a significant impact on the safety and effectiveness of therapeutic ultrasound, and presents a major hurdle for the wider clinical acceptance of these exciting technologies. In principle, any distortions to the ultrasound beam could be accounted for using advanced computer models. However, the underlying physics is complex, and the scale of the modelling problem requires extremely large amounts of computer memory. Using existing software, a single simulation running on a supercomputer can take many days to complete, which is too long to be clinically useful. The aim of this proposal is to develop more efficient computer models to accurately predict how ultrasound waves travel through the human body. This will involve implementing new approaches that efficiently divide the computational problem across large numbers of interconnected computer cores on a supercomputer. New approaches to reduce the huge quantity of output data will also be implemented, including calculating clinically important parameters while the simulation runs, and optimising how the data is stored to disk. We will also develop a professional user interface and package the code within the regulatory framework required for medical software. This will allow end-users, such as doctors, to easily use the code for applications in therapeutic ultrasound without needing to be an expert in computer science. In collaboration with our clinical partners, the computer models will then be applied to different applications of therapeutic ultrasound to allow the precise delivery of ultrasound energy to be predicted for the individual patient.

The world faces a major challenge, namely, how to supply energy to a growing population reliably, economically and without causing severe environmental damage. Looking forward, it is inevitable that our reliance on electricity will increase, as transport and heating become increasingly electrified, and that this increase will be largely met by renewable sources. Such facilities will be constructed at locations where prevailing conditions are appropriate and, in the UK, an example relates to plans to develop major offshore wind resources in the North Sea. However, the construction of large offshore facilities and the transmission of the resulting electricity back to shore is still very expensive and, therefore, it is imperative that this is done efficiently. All electrical plant relies upon electrical insulation and, today, this is primarily based upon polymers. While these materials are excellent electrical insulators, they are also poor conductors of heat, such that heat dissipation is a major issue. There would therefore be massive technological, environmental and societal benefits from the availability of commercially viable material systems that were excellent electrical insulators and good thermal conductors. Although it is intuitively appealing to think that thermal conductivity can be increased by adding a good thermal conductor to a thermally insulating material, this is not generally true, because the resulting boundaries give rise to phonon scattering which, effectively, offsets the anticipated gains. While this can be overcome if the thermally conducting additives form percolating paths through the material, the consequences of this have inevitably been an unacceptable reduction in the electrical breakdown strength of the material. However, recent results obtained at the University of Southampton appear to overthrow this paradigm. Specifically, a 20% INCREASE in breakdown strength has been accompanied by a 60% INCREASE in thermal conductivity in a system based upon hexagonal boron nitride (h-BN) dispersed in a polyethylene matrix. Since these preliminary results were obtained from a totally non-optimised system, we believe that further improvements in both technical performance and economic attractiveness (i.e. reduced cost from adding less h-BN) are attainable. The results of our preliminary work are contrary to accepted understanding, so the PROJECT AIM is to determine how simultaneous improvement can be optimised for use in two key materials that are particularly relevant to power cable applications. The key challenges are: to understand how to optimse the exfoliation of h-BN particles into their constituent layers and, subsequently, to disperse them within the matrix, such that the required combination of electrical and thermal characteristics result; to ensure scale-ability, such that laboratory results are technologically viable. In this project, we will consider two matrix systems, due to their technological relevance. First, we will examine crosslinked polyethylene (XLPE), since this is currently the most important cable insulation material. The work programme will progressively build from improving solvent dispersion, polymer blending methods and surface functionalisation, to scale-up with masterbatch production through combined solution and melt-process methods. Characterisation of the microstructure and dielectric testing will ensure consistent dispersion and distribution of the hBN filler, as well as optimal electrical properties. In this way, quantitative structure-property-process relationships will be established that will enable the resulting material systems to be used reliably in the electrical cable industry. While the focus of this project is on electrical properties, the knowledge about structure-property-process relationships will affect much wider technology areas, which employ advanced materials for improved mechanical or thermal properties.

This project is concerned with socially integrated mitigation of multiple structural risks in the urban environment, with a focus on the linked risks of earthquake and fire. Fire is the largest contributor to building damage following earthquakes. To date, this research area has largely been ignored as it crosses the boundaries between the knowledge areas of earthquake and fire safety engineering. The combination of factors adds to the challenges in risk estimation already existing in each distinct area. There is currently no universally accepted method for accounting for the effect of strengthening practices on building vulnerability to earthquakes (let alone earthquakes followed by fire). In the case of fire safety engineering, few credible techniques for damage estimation or risk-based design currently exist due to a lack of requisite input data. This project will develop, through large scale structural testing and computational analysis, new technical engineering solutions to these problems. And, for the first time, these technical engineering solutions will be developed explicitly accounting for the social context within which they are to be enacted. 21st Century engineering must provide effective and practical multi-hazard risk mitigation solutions against a background of growing urbanisation, overstretched services and vulnerable infrastructure. Thus, Challenging RISK will develop and holistically optimise strategies for the mitigation of multi-hazard risks, leading to resilient physical infrastructure within local communities. It proposes fundamental engineering research leading to new understanding of the response of existing concrete structures to earthquake and fire, with a long-term view to extend the lessons learned to other hazards. Simultaneously, it will design and implement effective mitigation promotion campaigns by extending existing research on individual and societal risk perception to community level through adoption of participatory citizen science techniques. The interaction with communities, local authorities and construction industry, will also feed back into the experimental programme, such that only structural strengthening options that meet the requirements of these stakeholders will be investigated. Our ultimate goal is to increase the uptake of structural and non-structural mitigation measures, resulting in reduced life and economic losses. We will produce new knowledge on the performance of existing reinforced concrete structures subjected to earthquake and fire hazards (individually and in sequence), with the aim of developing an integrated framework for performance-based assessment and structural mitigation. We will engage with communities to define "acceptable" performances, inform the technical programme and the means for effective implementation. Large-scale experimental studies will be carried out on structural elements and sub-assemblages representing non-seismically designed buildings that have been strengthened for improved seismic response using fibre reinforced polymers (FRP). The elements will be subjected to seismic loads and then to fire (and vice-versa). They will provide information for the design of FRP strengthening to achieve multiple performance states under the fire and earthquake loads. In parallel, action-oriented research, where interventions are carried out at both the technical and social levels, will be used to develop an understanding of risk perception and representation of fire and seismic risk at the individual and community levels, with specific interventions carried out in London, Seattle, and Osaka. This will be done, for example, through seismic monitoring combined with perceptions surveys using smartphones, and recording fire risks and incidents using social media. The results of this part of the study will assist in the development of effective communication to raise awareness, leading to action.

Conventionally designed wind turbines only operate efficiently in steady, uninterrupted air. However, most users want to access wind in urban areas or near industrial units where the nature of the wind is more turbulent and swirling. Conventional designs do not work efficiently with the swirling, variable nature of wind at such sites. In this project Swift Energy present a radical re-design of a vertical axis wind turbine, with key technological improvements that will allow efficient operation in small-footprint, urban sites. Such sites have the added advantage that they are close to consumers, minimising transmission losses. WindSurf is a vertical axis, active pitching wind turbine. Swift's patented control technology uses servomotors to continually alter blade pitch, which allows self-starting in wind speeds as low as 3m/s, and optimised energy capture in free and turbulent wind streams. Edinburgh's role in this project is to produce an optimised design of the electrical generator for the WindSurf rated at 16kW, taking into account the environment in which it will be operating. A direct drive generator will be used to eliminate the gearbox, which will improve reliability and efficiency. Both of these contribute to LCOE: reliability through increased availability and reduced OPEX; and improved efficiency will enhance annual energy yield. An air-cored permanent magnet generator will be designed and built that is optimised for the structure of the Swift wind turbine. In order to achieve such an optimised design an integrated design approach is required, which links electromagnetic design, with structural design and thermo-fluid design. Edinburgh has built up 10 years of experience in the integrated design of direct drive permanent magnet air-cored generators for wind and marine renewable energy applications. Air-cored machines eliminate undesirable magnetic attraction forces that try to close the gap, and thus this topology benefits manufacture, assembly and structural design. A vertical axis wind turbine allows the electromagnetic design of the machine to have a large diameter, out near the blades. A large diameter will result in high airgap velocity and thus have a positive impact on torque density (Nm/kg), reducing the amount of active material, which is the most expensive part of the machine. A novel structural arrangement will be developed for integration into the turbine, which where possible makes best use of the existing structural material, again to minimise material usage and thus cost. A modular design approach will be adopted to ease manufacture and assembly of the generator, but also to make O&M easier. By positioning the generator close to the blades, we will investigate we will investigate methods of "scooping" air from the turbine onto the generator to assist with cooling. Effective cooling will benefit the torque density and the overall performance of the machine. Numerical modelling tools will be used in the design process, such as ANSYS for structural analysis, StarCCM for thermo-fluid analysis, and Infolytica for electromagnetic design. An existing analytical design tool will be refined based on the structural and CFD modelling in order to assist SWIFT in the future design and production of their turbine. Multi-body modelling using SIMPACK will be combined with structural modelling to investigate the impact of environmental loads on the generator in terms of airgap deflection. Once the design is finalised, the machine will be built under subcontract to Fountain Design Ltd, with whom we have worked in the past to build prototype generators. The machine will be tested at the University of Edinburgh on its wind-emulator test rig to verify performance and the design tools developed. A thorough integrated design approach with manufacturing and production techniques in mind supported by laboratory testing will ensure that SWIFT can move towards commercialisation.

Marine Class rules require the performance of an integrity survey every five years involving dry-docking the vessel, with external hull inspection, maintenance carried out as required (cleaning and painting). Safety societies demand that 100% ultrasonic testing should be performed on these welds. This represents the welding and inspection of some 0.5 km of weld line between the hull sections on the external surface of the hull. The length of time taken for human operators to perform all of this welding and inspection has obvious implications for high labour costs and high incidence of fatigue-induced mistakes and safety risks for personnel performing the inspections. An automated process would reduce costs to shipbuilder, reduce the need for hazardous working at height and improve consistency, reliability and availability of inspection data. However, current automated solutions involve large gantry robots that are unwieldy and expensive and for these reasons are not widely used. The shipping industry in the UK is undergoing a renaissance with specialization in high specification specialist and luxury vessels. The UK also has many companies, such as project lead NDT Consultants providing specialist inspection services to the global market. The aim of AWI is to develop mobile robots that can climb vertical and curved hull surfaces to deploy inspection tool. The system will be autonomous, and require no external scaffolding or gantry structures and relatively low cost. AWI will provide the project lead NDT Consultants Ltd. with new inspection services that they will use to expand their business in the global inspection industry. The supply chain consortium also includes a robotics systems SME manufacturer who will bring new robotic solutions to market.

The development of new and efficient catalysts for the synthesis of complex molecules is an ongoing need in the chemical industry, with both bio- and chemo-catalysts being applied at scale. This research is concerned with the design of a novel, hybrid chemo-enzymatic catalyst, which combines the best features of both chemical and enzymatic catalysis, by introducing metal catalysts into natural enzymes. Enzymes have evolved in nature and can be modified by scientists to perform many useful transformations in a very efficient way. However, the modifications are never too far from the natural world, in terms of functionality (i.e, mutation of amino acids to other amino acids). Recently, specific chemical modification of proteins has been developed as an exciting new tool to introduce chemo-catalytic functionalities within a biomolecule, thus expanding the functionality space. The objective of this project is to implement such a modification procedure into an existing enzyme, to change its reactivity and improve a key reaction needed for industrial exploitation of enzymes. The novelty of this design is the use of an enzyme specifically evolved for substrate binding and transition state stabilisation. The longer term vision is to apply hybrid catalysts to completely unnatural catalytic reactions. The use of enzymes that stabilise hydride transfer transition states in conjunction with suitable chemical catalysts could be envisaged for expanding the enzyme functionality to the reduction of more challenging substrates, rarely occurring in nature (e.g. nitriles, amides). Therefore, this research addresses a timely and fascinating question for enzyme study and application: can we design enzymes containing ANY desired catalytic functionality? The hybrid catalysts prepared here will be used for the synthesis of molecules containing alcohol functions. Chiral alcohols are important constituents of pharmaceuticals and agrochemicals. Therefore, catalysts that facilitate their preparation are important tools for these industries. The use of enzymes as catalysts is very efficient and sustainable, due to their high activities and selectivities, benign reaction conditions and their renewable origin. However, many of the enzymes that can make chiral alcohols enantioselectively have one major shortcoming: they employ the expensive and sensitive nicotinamide adenine dinucleotide phosphate, NADPH cofactor (£1085 per g), which is an essential component of the reaction mixture. The high cost precludes its use as a disposable reagent and continuous regeneration of NADPH in the reaction vessel (in situ) is necessary for economically acceptable transformations using isolated enzymes. Existing regeneration systems are based on either enzymes or chemical catalysts, but they have limited applicability, because they generate by-products, have a low stability or activity, or are inactivated by the presence of the enzyme. Therefore, we will engineer a chemical cofactor regeneration system that will be protected from deactivation, by incorporating it into a protein. We will use the same protein to incorporate the regeneration catalyst and to perform the alcohol synthesis, and in this way we will avoid complex mixtures containing two enzymes. The system will be of relevance to the industries involved in the synthesis of active pharmaceutical ingredients (pharma, custom manufacturing organisations etc), which could potentially apply it as a simple technology for recycling NADPH.

A primary goal of organic chemists is the construction of molecules for applications as diverse as medicines, new materials and biomolecules. The field is constantly driven by the need for new, more efficient methods as well as ways to access molecules which may have previously been impossible. The most important tool at an organic chemist's disposal is undoubtedly catalysis, whereby the use of a small amount of a custom-designed catalyst can permit a reaction to occur under much milder conditions than otherwise, or opens up new chemical pathways altogether. For this reason, innovation in catalysis is central to innovation in organic chemistry. Nature's catalysis is performed by enzymes; evolution has made them phenomenally efficient. Often playing a leading role in enzymatic processes are 'hydrogen bonds', special types of electrostatic attraction which are important in facilitating the chemical reaction between two molecules by bringing them into close proximity with one another or by stabilising the pathway leading to product formation. My research seeks to employ these same interactions, but in the context of small molecules which we can readily synthesise and handle in the lab. This approach to catalysis is very exciting as it is still in its infancy yet offers exciting opportunities for both activation and control. This project will seek to take inspiration from a distinct field within chemistry called Supramolecular Chemistry, which explores the behavior of large molecules which are assembled from smaller ones using multiple weak 'temporary' interactions working in tandem. Hydrogen bonds are very important in this regard but there are a number of other key interactions such as ion pairs and pi-cation interactions which have been shown to be powerful in building up molecular structures. It is our aim to apply several of these interactions together in tandem to design new catalysts that will bind with our reactant in a very well defined orientation. The catalyst will also induce the substrate to react with another molecule, allowing the selective synthesis of one mirror image of a molecule over the other (so-called enantiomers). This is a very important pursuit in science, since the inherent 'handedness' of biological systems means that the different mirror image forms of chiral molecules often have very different effects in the body. This is of particular importance in pharmaceutical applications.

The product rule of the all familiar operation of taking derivatives of real valued functions has a plethora of generalisations and applications in algebra. It leads to the notion of derivations of algebras - these are linear endomorphisms of an algebra satifying the product rule. They represent the classes of the first Hochschild cohomology of an algebra. The first Hochschild cohomology of an algebra turns out to be a Lie algebra, and more precisely, a restricted Lie algebra if the underlying base ring is a field of positive characteristic. The (restricted) Lie algebra structure extends to all positive degrees in Hochschild cohomology - this goes back to pioneering work of Gerstenhaber on defornations of algebras. Modular representation theory of finite groups seeks to understand the connections between the structure of finite groups and the associated group algebras. Many of the conjectures that drive this area are - to date mysterious - numerical coincidences relating invariants of finite group algebras to invariants of the underlying groups. The sophisticated cohomological technology hinted at in the previous paragraph is expected to yield some insight regarding these coincidences, and the present proposal puts the focus on some precise and unexplored invariance properties of certain groups of integrable derivations under Morita, derived, or stable equivalences between indecomposable algebra factors of finite group algebras, their character theory, their automorphism groups, and the local structure of finite groups.