Current and future energy policies are increasingly aiming to reduce carbon emissions from the propulsion and power sector. The combustion of fossil fuels releases carbon, in the form of carbon dioxide (CO2), and there is consensus that the rapid anthropogenic emission of fossil bound carbon is resulting in global climate change. Co-currently, there is growing awareness of the negative impacts of toxic exhaust pollutants from fossil fuel combustion, such as nitrogen oxides (NOx) and carbonaceous soot or particulate matter (PM), on the health of urban populations. While electrification offers a potential replacement for fossil fuels, the electric powertrain is currently only suitable for light duty applications, such as passenger vehicles. There are several high energy requirement applications (aircraft, off-road vehicles in military and construction, thermal power generation) for which currently no appropriate alternative to combustion engines exists. Hydrogen (H2) has the potential of emerging as the leading energy carrier for the next generation of zero-carbon emission combustion systems. H2 fuelled gas turbines are potentially capable of providing very efficient energy conversion with no carbon emissions, and will be able to span the power and weight requirements of land-based power generation and aero-propulsion. H2 can offer significant benefits over hydrocarbon fuels; its wide flammability range allows very lean combustion, low ignition energy ensures prompt ignition and high diffusivity facilitates efficient air-fuel mixing. However, the utilisation of H2 for combustion is hindered by considerable challenges. Its high flame speed can intensify risks of flame instability and flashback, adversely affecting operation, and high rates of heat release (leading to high thermal loading), combined with H2's corrosive properties, can lead to combustor damage. This means that current gas turbine combustors are not suitable for pure H2 combustion and will have to be re-designed. Complex reactions, turbulent conditions and complicated geometries means that conventional design techniques (such as simulation tools) need to be revised for H2 combustion. Comprehensive experimental campaigns are required to fulfil the gaps in our understanding of fundamental H2 combustion, and to identify regimes for high efficiency and near-zero emission operation in practical H2 combustion systems. In order to set out new design and operation principles for H2 combustors, the research proposed will (a) identify strategies for H2 injection and efficient mixing with air to create a uniformly distributed H2-air mixture, (b) identify suitable operating conditions that result in favourable flame behaviour with suppressed NOx emissions, (c) identify suitable materials for use with H2 at elevated pressures and temperatures, (d) understand the influence of acoustic boundary conditions on combustion instabilities and (e) investigate the effects of translating concepts studied in a-d vary from lab-scale to large-scale systems operating at practical conditions. The fundamental principles associated with H2 combustion will be developed and evaluated through rigorous experimentation at laboratory scale, and then implemented in two different types of semi-industrial scale combustion systems, (i) representative of industrial small gas turbine for power generation, and (ii) scaled down version of the pre-burner component of the SABRE rocket engine. The experiments performed on these semi-industrial systems will lay the foundations for the follow-on research (beyond the 4 years of this fellowship) to integrate H2-fuelled combustors in full-scale industrial multi-cannular gas turbines and in full-scale rocket engines. The research outcomes will provide underpinning scientific knowledge on H2 combustion for the project partners, Siemens Industrial Turbomachinery Ltd. and Reaction Engines Ltd. (REL), giving them a direct uptake route for this research.

Magnetic Resonance Imaging (MRI) scans play a vital role in helping many ill children, by finding out what the problem is and helping plan their treatment. MRI is safe because it does not use radiation. MRI scans produce good-quality pictures or images of many parts of the body, including the brain, heart, spine, joints and other organs. The main problem is they take a long time - often over an hour. During the scan, the child has to keep very still and may even need to hold their breath many times. This is especially hard for children and unwell patients. Hence, younger children under 8 years old need a general anaesthetic, to put them to sleep during the scan. In many childhood diseases, for example in cancer, children may need many MRI scans to follow up disease progression and treatment. Being put to sleep for all of these scans is not pleasant for the child and may occasionally cause problems. It also puts a lot of pressure on hospitals who need to find the doctors, beds, equipment and funds for this. One way of overcoming these problems would be to speed up the MRI scans so the children do not have to keep still or hold their breath. The simplest way of doing this is to collect less data for each image, but this causes so much distortion in the images that they cannot be used. There are some ways of converting these into useful images, but these are complicated and take too long to use in a hospital. Machine Learning is an upcoming way of teaching computers to find complicated patterns in large amounts of information. Recent advances mean that computers are now so powerful that they can learn effectively. Machine Learning has been successfully used for analysing many types of images, for example to perform de-noising, interpolation, image classification and border identification. Despite its popularity, only a few recent studies have shown its potential for reconstruction of MRI images. This is partly due to the greater complexity of the problem and importantly, the large amounts of data required to 'learn' the solution. At Great Ormond Street Hospital, we have MRI images from over 100,000 children and scan an additional 10,000 children each year, all of which we could use to help train and test Machine Learning technologies. I have already shown that basic Machine Learning techniques can remove distortions from MRI scans of the heart, so I am well placed to develop Machine Learning techniques to reconstruct MRI images from other children's diseases, as well as developing more advanced Machine Learning techniques. I showed Machine Learning to be faster than existing reconstruction methods and the images were of better quality than more conventional state-of-the-art techniques. However, much more work is needed to get Machine Learning working reliably in children's scans and to make the most of the possible benefits. If we can use fast scanning with Machine Learning we could shorten scan times from 1 hour to about 10 minutes for children having MRI scans. They would not have to keep completely still for the scan and would not have to hold their breath, therefore reducing the need to put patients to sleep. This would make MRI scanning far less difficult and daunting for children, and would eliminate the cost and side effects from the anaesthetic. Quicker scans would help reduce waiting lists and costs for the NHS. It would also mean that MRI scanning would be used far more often, so it could help many more children. Additionally, these techniques could enable MRI scans to become affordable in some countries for the first time.

Understanding, predicting and controlling turbulent flows is of central importance and a limiting factor to a vast range of industries: naval, aeronautical, automotive, power generation, process, pharmaceutical, meteorological and environmental. Our view is that the key to advances in turbulence is by sustaining and stimulating interaction among researchers. It is essential that a diverse range of viewpoints, opinions, strategies and methods are brought together in an efficient and constructive manner. The essence of the consortium is to provide the central core of a needed critical mass activity considering the big challenges posed by turbulence. The consortium brings together complementary expertise/experience/knowledge and coordinate activities to look at coherent, rational and strategic ways of understanding, predicting and controlling turbulent flows using High Performance Computing. The consortium is crucial for the UK in order to coordinate, augment and unify the research efforts of its participants and to communicate its expertise and findings to a wider audience. Firstly funded in 1995, the UKTC has been through five highly successful iterations. It has seen significant growth since its inception, from 5 original members to 46 members over 21 UK institutions for the present bid, and is continuously receiving requests from academics to join (20 new members for the present bid). In the last 22 years, the UKTC has (i) demonstrated its ability to convert access to national High-End Computing (HEC) resources into internationally leading research (hundreds published papers since 1995 with thousands non-self citations), (ii) established its international competiveness, (iii) helped its members to leverage and secure multi -million £ grants from governmental funding bodies and industries, (iv) allowed the discovery of new fluid flow phenomena which have led to new ways of improving beneficial effects and reducing negative effects of turbulent flows and (v) facilitated the design of more sophisticated turbulence models redefining industry standards. The member of the consortium are (in alphabetic order): Pavlos Aleiferis (Imperial College London); Eldad Avital (Queen Mary London); Angela Busse (University of Glasgow); Yongmann Chung (University of Warwick); Dimitris Drikakis (University of Strathclyde); David Emerson (Daresbury Lab); Jian Fang (Daresbury Lab); Gerard Gorman (Imperial College London); Shuishen He (University of Sheffield); Yongyun Hwang (Imperial College London); Richard Jefferson-Loveday (University of Nottingham); Xi Jiang (Queen Mary London); Robert Kerr (University of Warwick); Jae-Wook Kim (University of Southampton); Sylvain Laizet (Imperial College London); Michael A. Leschziner (Imperial College London); Kai Luo (University College London); Xuerui Mao (University of Nottingham); Olaf Marxen (University of Surrey); Joanne Mason (University of Exeter); Aimee S. Morgans (Imperial College London); Charles Moulinec (Daresbury Lab); Gary Page (Loughborough University); George Papadakis (Imperial College London); Matthew Piggott (Imperial College London); Alfredo Pinelli (City University London); Alistair Revell (University of Manchester); Pierre Ricco (University of Sheffield); Aldo Rona (University of Leicester); Neil Sandham (University of Southampton); Mark Savill (University of Cranfield); Peter Schmid (Imperial College London); Mehdi Seddighi (University of Liverpool); Spencer Sherwin (Imperial College London); John S. Shrimpton (University of Southampton); Vassilios Theofilis (University of Liverpool); Emile Touber (Imperial College London); Paul Tucker (University of Cambridge); Maarten van Reeuwijk (Imperial College London); J. Christos Vassilicos (Imperial College London); Peter Vincent (Imperial College London); Andy Wheeler (Univer- sity of Cambridge); Beth Wingate (University of Exeter); Jun Xia (Brunel University London); Yufen Yao (University of Bristol).

The primary users of gas turbines are being impacted by rising fossil fuel prices and stringent government targets for reducing carbon-dioxide emissions. This is putting increasing pressure on gas turbine manufacturers to improve engine efficiencies so that their products remain competitive. One way of improving the efficiency of a gas turbine is to raise the turbine entry temperature (TET). Present-day engines operate with TETs as high as 2000K, which is well above the melting point of the alloys from which first-stage turbine blades are made. Two cooling techniques are employed to prevent damage to the blades from high TETs: film cooling, where a thin film of coolant introduced to the external surface of the blade reduces the driving-temperature for heat transfer; and internal cooling, where coolant is passed through a series of passages within the blade to convect heat from the internal surfaces. The air for this cooling is taken from the compressor at a penalty to engine efficiency: for every 1% of air drawn from the compressor a 1% drop in isentropic efficiency follows. Relatively few experimental studies have investigated coupled film and internal cooling; consequently there are insufficient published data for validation of the models used to predict blade metal temperatures. There is little margin for error in these predictions: the life of a blade can be reduced by half if the temperature at which it operates is 10K higher than predicted. As a result, blades are often superfluously cooled at the expense of engine efficiency. Validated models would enable blade cooling schemes to be designed with more confidence. This would reduce design conservatism, enabling more efficiently cooled designs with an associated improvement in engine efficiency. It would also reduce the costly risk of re design or in-service replacement of inadequately cooled blades. The proposed project will design and build a highly-modular rig for obtaining fluid dynamic and heat transfer information on test pieces subjected to coupled film and internal cooling. The rig will make use of the University of Bath's state-of-the-art EPSRC funded Versatile Fluid Measurement System (VFMS), enabling high-precision measurements of heat transfer coefficients and temperatures on the surface of the test pieces, and the concentration field and three component velocities in the fluid volume above the film cooling holes. The flexibility of the facility combined with the unparalleled fidelity of measurement techniques offered through the VFMS will make it a highly novel and extremely useful platform for studying combined film-internal cooling. Findings from the project will provide unique insight into the fundamental science of the research problem and will supply Siemens - the industrial partner in this proposal - with data to validate their models and inform design methodology. The data will also be made available to workers in the wider gas turbine technical community and academia.

UK industries are facing a growing problem - a lack of experts! Multiple sectors of the UK's economy, especially in Engineering, are increasingly dependent on older workers, leaving employers exposed to a massive need for skilled staff when they retire. While the UK attempts to provide more quality vocational training to young people so they can replace skilled older workers when they retire, there remains years of knowledge gap to be filled. Hence, a technological solution becomes increasingly attractive - i.e. assisting humans with "Virtual Expert" (VE) systems and complementing them while they acquire experience. Many UK companies in industry have a range of automation and digitalisation challenges, such as automatic remote condition monitoring tools and engine test automation, which this project seeks to address. The main concept behind this new project is to build and train an Evolutionary Virtual Expert System (EVES) to assist current and future industrial fault diagnostic engineers. These "virtual apprentices" (diagnostic agents, including knowledge-based rules, signal processing algorithms and model-based approaches) will be trained by human experts, through coaching, examining and refining processes. After a number of subject matter tests, the successful "virtual apprentices" are promoted to become VEs and their weightings (rankings) will be updated using a genetic algorithm. Over generations of evolution, EVES will be able to find a suitable population of VEs (rules/algorithms/models), and produce a heuristically best decision through a weighted voting process, with reasoning mechanisms and possible solutions made transparent to users. EVES integrates the strengths of precision, learning ability, adaptability and knowledge representation from all the VEs that conform to the population, aiming to provide an automated and digitalised fault diagnostic system, to match or possibly outperform human experts working without such support. The EVES project will have a big impact on areas of industrial application. This proposal is timely, as the proportion of experts in UK industries are getting older, while at the same time more modern technologies involve longer learning curves for young people. To be ready for the industries of the future, these VEs, when fully trained, will provide critical support for existing experts, and also act as good trainers for the younger workers. As the future generation is based on high technologies, good virtual assistants and virtual trainers will become increasingly important. The proposal is important, as the structure of EVES is widely applicable to all industrial sectors, for example, from fault diagnostics of machines and plants, to remote condition monitoring for railway applications, agriculture precision, water quality monitoring, and even to diagnostics for human health.