Over the past decade, additive manufacturing (AM) has gained considerable traction in the gas turbine industry. Manufacturers now look to AM for the production of critical components central to the combustor architecture. However, AM struggles to achieve satisfactory surface finishes, with its increased surface roughness shown to impact axial velocities, heat release, NOx emissions, and stability limits. Correlations between increased roughness and boundary layer flashback (BLF) resistance are of significant interest for lean premixed combustion of high hydrogen content (HHC) fuels. With the development of innovative burner configurations designed to address static stability issues in HHC fuels, and the growing use of AM in their fabrication, studying roughness effects on these designs is essential. This study aims to quantify the impact of surface roughness on an industry-relevant jet burner configuration, building on previous research by the authors using premixed swirl geometries. A premixed jet burner was developed with an interchangeable section to accommodate varying surface textures. The burner was characterized under atmospheric hydrogen-fired conditions at thermal powers ranging from 10 to 25 kW. Both smooth and rough inserts representing traditional machining and selective laser melting manufacturing techniques were utilized. The findings highlight how surface roughness influences flame topology, static stability, and emissions, providing an experimental basis for future numerical studies on roughness sensitivity in advanced burner designs.

A new optical modular staged combustor has been developed to facilitate the study of the staged combustion of alternative fuels, both liquid and gaseous, at elevated inlet pressures and temperatures. Initial testing to characterize the combustor was carried out at pressures of up to 6 bara, for both premixed and non-premixed methane/air swirling flames with a combustor inlet temperature of 450 K. Exhaust gas analysis together with OH* and CH* chemiluminescence was used to evaluate the operability of the combustor and the quenching effect of the secondary air flow for a range of primary and global equivalence ratios. Laser Induced Grating Spectroscopy (LIGS) was used to obtain instantaneous local temperatures of the mixture across the secondary air mixing zone. A previously developed analytic approach was used to extract temperature from the measured signal frequencies within the primary flame brush and secondary mixing region, where both burnt and unburnt gases were present. The development of such modular apparatus specifically to allow for the application of novel optical techniques, such as LIGS, enables the future evaluation of different combustor architectures for unconventional fuels. Additionally, this work further demonstrates the development of the LIGS technique for obtaining measurements within complex high-pressure, turbulent swirling flames.

Parametric experiments were performed using CH4-H2 blends to fuel a model gas turbine combustor at elevated inlet conditions, unabated and whilst employing exhaust gas recirculation (EGR) as a NOx-abatement technique. The purpose was to quantify the performance of conventional dry, 15% O2 NOx correction (ISO) against mass by heat input (MBHI) normalisation to experimentally verify simulated work from the literature, while expanding on this work to cover NOx-abated combustion. NOx was found to be consistently inflated in H2 combustion when compared to CH4 by the ISO method, whereas the MBHI method was found to closely approximate the mass by work output baseline emissions. Multipliers proposed to correct this inflation were shown to be impractical. A fundamental problem in assessing the effectiveness of EGR using the ISO method was identified as it does not account for the additional volumetric dilution introduced, whereas the MBHI normalisation method avoids this issue.

Interest in using renewably produced, partially cracked ammonia in gas turbines is gaining traction, but challenges relating to emissions of NOx and unburned ammonia remain. The present work progresses existing research on using hydrogen stratification to reduce NOx from ammonia/hydrogen flames by experimentally and numerically investigating the effects of also injecting nitrogen from the cracking process. It additionally assesses the NOx reduction capability of a recently developed novel swirl burner by adding hydrogen to the stratified flow to maintain the diffusive equivalence ratio at two high NO production conditions, slightly lean and stoichiometric. At slightly globally rich conditions, maintaining the diffusive equivalence ratio at 0.9 resulted in an order of magnitude reduction in NO emissions with only a 33% increase in unburned NH3, compared to a fully premixed flame with the same fuel and air flow rates. This stratified configuration was found to increase consumption of NO by NH2, likely due to flame morphology effects, while NO production from OH and HNO pathways was reduced. The reduced OH intensity was posited as the cause for increased NH3 emission. A strong emissions sensitivity to diffusive equivalence ratio was found, as the case with a stoichiometric diffusive equivalence ratio did not show such marked improvements over its corresponding premixed condition. Both stratified and premixed flames were found to be stable, however stratification has potential to trigger instabilities at different frequencies to premixed.

Using renewably produced ammonia as a zero-carbon fuel is gaining momentum due to its ease of transportation and storage as a hydrogen vector. This is particularly true for partially cracking ammonia immediately prior to use, injecting a blend of NH3, H2 and N2. Challenges with this fuel combination relate to the emissions of NOx and unburned NH3, as well as understanding flame stability for practical applications. In this study, a 20 %(vol.) cracked ammonia blend was investigated using a fully premixed swirl burner, operating at a thermal power of 10 kW with steam injection of 30 %(vol.) of the fuel and preheating inlet temperatures of up to 390 K, for a range of equivalence ratios from lean to rich. Emissions of NO, NO2, N2O, NH3, H2, O2 and H2O were recorded, along with OH*, NH* and NH2* chemiluminescence. Additionally, a numerical investigation was conducted using CHEMKIN-PRO to elucidate the main reactions responsible for reducing emissions by providing a rate of production analysis. The 20 %(vol.) cracked ammonia blend was found to reduce NO, NO2 and N2O significantly, with an increase in NH3 emissions at rich conditions and instabilities at both lean and rich extremes, compared to the widely investigated 70/30(vol %) ammonia/hydrogen blend. Humidification reduced NO and NO2 emissions due to a reduction in HNO production via OH and NH but caused an increase in N2O by reducing the flame temperature and unburned NH3 emissions at rich, low power conditions due to combustion instabilities. Unburned H2 emissions however were reduced, likely relating to a reduction in exhaust temperature thermally cracking less unburned NH3 into H2 and N2.

The use of additive manufacturing (AM) has seen increased utilization over the last decade, thanks to well-documented advantages such as lower startup costs, reduced wastage, and the ability to rapidly prototype. The poor surface finish of unprocessed AM components is one of the major drawbacks of this technology, with the research literature suggesting a measurable impact on flow characteristics and burner operability. For instance, surface roughness has been shown to potentially increase resistance to boundary layer flashback—an area of high concern, particularly when utilizing fuels with high hydrogen content. A more detailed understanding of the underlying thermophysical mechanisms is, therefore, required. Computational fluid dynamics can help elucidate the impact of these roughness effects by enabling detailed data interrogation in locations not easily accessible experimentally. In this study, roughness effects on a generic gas turbine swirler were numerically modeled using a low-y+ detached eddy simulation (DES) approach. Three DES models were investigated utilizing a smooth reference case and two rough cases, the latter employing a literature-based and novel equivalent sand-grain roughness (ks) correlation developed for this work. Existing experimental isothermal and CH4 data were used to validate the numerical simulations. Detailed investigations into the effects of roughness on flow characteristics, such as swirl number and recirculation zone position, were subsequently performed. The results show that literature-based ks correlations are unsuitable for the current application. The novel correlation yields more promising outcomes, though its effectiveness depends on the chosen turbulence model. Moreover, it was demonstrated that, for identical ks values, while trends remained consistent, the extent to which they manifested differed under reacting and isothermal conditions.

Ammonia (NH3) is a zero-carbon fuel which can be produced renewably, so has been gaining significant research interest in recent years. It is also found in the waste streams of all industries handling material of organic origin (e.g. agriculture, sewage, oil refining, coal coking, etc.). As ammonia is toxic, it is often destroyed rather than recovered. The Phosam process enables recovery of high purity anhydrous ammonia from industrial waste streams and has been used commercially in the steel industry. Ammonia combustion challenges include low reactivity and a high propensity for NOx emissions. Low reactivity can be overcome by blending with more reactive gases (e.g. hydrogen) and NOx emissions can be reduced via staged combustion. Fortunately, hydrogen-rich process gases are often a byproduct of the same industries. This work progresses previous work that successfully combusted a 15%(vol) hydrogen-rich industrial process gas (coke oven gas) with both anhydrous ammonia and humidified (30 %(vol) H2O) ammonia using a premixed swirl burner in a model GT combustor under fuel-rich conditions and ambient pressure. This campaign introduces secondary air at two different locations (15 and 25 cm from the burner exit) and at two pressures (1.1 and 1.3 bara), to demonstrate relative chemical kinetic effects on emissions formation with complete combustion of these relatively complex, multicomponent fuel blends. The equivalence ratio (Phi) was varied, both in the primary zone and globally to evaluate the relative parametric effects on exhaust emissions. The global Phi was manipulated by substitution of a portion of the air with nitrogen of equivalent thermal capacity. When secondary air-staging, the anhydrous blend achieved consistently lower minimized NO than the humidified blend, despite the blends having equivalent minimized emissions in the absence of secondary air. Modest pressure elevation of similar to 17%, reduced NO entering the second stage by similar to 25%. Staging further downstream produced similar to 24% lower NO, when operating at near atmospheric pressure. Observations of OH* and NH2* flame chemiluminescence indicated that flame structure was modestly influenced as the secondary staging moved upstream. Both staging locations enabled adequate secondary air mixing prior to gas sampling (CO reduced from similar to 6000 to < 5 ppm). Increases in global F (above 0.7) had no measured effect on NO production.

Contemporary research into decarbonized fuels such as H-2/NH3 has highlighted complex challenges with applied combustion, with marked changes in thermochemical properties leading to significant issues such as limited operational range, flashback, and instability, particularly when attempts are made to optimize emissions production in conventional lean-premixed systems. Non-premixed configurations may address some of these issues but often lead to elevated NOx production, particularly when ammonia is retained in the fuel mixture. Optimized fuel injection and blending strategies are essential to mitigate these challenges. This study investigates the application of a 75 %/25 %(mol) H-2/NH3 blend in a swirl-stabilized combustor, operated at elevated conditions of inlet temperature (500 K) and ambient pressure (0.11-0.6 MPa). A complex, nonmonotonic relationship between swirl number and increasing ambient combustor pressure is demonstrated, highlighting the intricate interplay between swirling flow structures and reaction kinetics, which remains poorly understood. At medium swirl (SN = 0.8) an increase in pressure initially reduces NO emissions, diminishing past similar to 0.3 MPa, with an opposing trend evident for high swirl (SN = 2.0) as NO emissions fall rapidly when combustor pressure approaches 0.6 MPa. High-fidelity numerical modeling is presented to elucidate these interactions in detail. Numerical data, generated using Detached Eddy Simulations (DES), were validated against experimental results to demonstrate a change in flame anchoring on the axial shear layer and marked change in recirculated flow structure, successfully capturing the features of higher swirl number flows. Favorable comparisons are made with optical data and a reduction in NO emissions with increasing pressure is demonstrated to replicate changes to the swirling flame chemical kinetics. Findings provide valuable insights into the combustion behavior of hydrogen-rich ammonia flames, contributing to the development of cleaner combustion technologies.

As alternative fuels are designated for future energy applications, flexible combustor designs require considerable development to ensure stable operation with reduced NOx emissions. A non-premixed variable swirl burner was used to experimentally appraise changes in NO production pathways, with CH4 NH3, and H2 flames, alongside intermediate fuel blends. Maintaining an equivalent thermal power and flame temperature between fuels, preheated reactants (500 K) were supplied to the burner, with parametric changes made to pressure (1–6 bara) and swirl number (0.8–2.0). NO production was characterized, alongside variations in flame structure and topology, with a correlation demonstrated for exhaust emissions. NO production was shown to be sensitive to combustor pressure, providing an expected increase for CH4 and H2 flames. Emission profiles from both NH3 and H2 flames are shown to be significantly augmented by a change in swirl number. As NH3 fractions were increased in the H2 blend, a decaying trend in NO emissions was observed with an increase in pressure, and as a function of mixture ratio. However, this behaviour was markedly augmented by a change in swirl number and suggests that further reductions may be possible at increased pressure. At the low swirl/high pressure condition the NH3/H2 blend outperformed pure H2, providing lower NO concentrations. Emissions data were normalised using the traditional dry/O2 correction, alongside mass scaled by thermal power, with a comparison provided. The corresponding differences in emission formation pathways were investigated, alongside high-speed OH* chemiluminescence to further elucidate findings.

An Artificial Intelligence-based method has been developed to achieve Non-Intrusive Load Monitoring (NILM) at 11kV/400V electrical substations, accomplished using the aggregate load measured at the substation and separating it by property. This allows Utility & Power Distribution companies better electricity demand forecasts and aids demand-side response. There has been extensive research of NILM at household-level using household "smart meter" data. However, substation-level disaggregation offers a cost-effective alternative and streamlined approach as only the substation requires monitoring devices rather than every house in the distribution region. A deep learning eventless NILM approach was taken, using Recurrent Neural Networks with Long Short-Term Memory and Gated Recurrent Unit layers. Disaggregation was achieved using a synthetic dataset for Low-Carbon Technology devices such as Electric vehicles (EVs), Heat Pumps (HP) and Photo-Voltaic Solar (PV) with the relationship with substation load. The accuracies for unseen data compared to the ground truth were 99.20% (EV), 99.39% (PV), 81.39% (HP) and 92.16% for other household loads. This method proposed in this paper is foundational to future substation-level NILM research, allowing for an evaluation of a scenario-based uptake in low-carbon technologies and an analysis of the required change of the power distribution network.

Renewably generated NH3 has the potential to support future energy demand, however combustor designs and strategies require considerable development to ensure reduced NOx emissions. Expanding on previous work, a turbulent swirl burner was used to appraise potential NOx reduction pathways, both experimentally and numerically, with a premixed NH3/H2/air flame. With a 100-year global warming potential ∼265 times that of CO2, particular emphasis was given to N2O. Maintaining a constant thermal power, reactants were supplied at elevated temperature, with parametric changes made to pressure and humidity. Favourable agreement was demonstrated between exhaust emission measurements and simulations performed using a chemical reactor network model. NO and N2O emissions were shown to be sensitive to operational equivalence ratio, increasing by several orders of magnitude across the experimental range. An increase in combustor pressure was experimentally shown to reduce exhaust N2O concentrations with this globally lean fuel mixture, alongside NO. Steam injection was also explored in detail for the first time and shown to provide contrasting trends, with a reduction in NO, and a rise in N2O, as water loading was increased. Both pressure increase and steam injection were combined to give optimal NOx performance for the evaluated dataset. Changes in chemical kinetic pathways were investigated in detail, and compared to high-speed OH*, NH2*, and NH* chemiluminescence.

Ammonia (NH3) has been suggested as a fuel to attain zero carbon emissions. However, dealing with ammonia needs careful studies to reveal its limits as a suitable and promising fuel for broad applications within large power requirements. Chemical reaction mechanisms, widely employed in the modeling of these applications, are still under development. Therefore, this review is aimed to shed light on the current mechanisms available in the literature, highlighting modeling parameters that directly affect reaction rates which in turn govern the per-formance of each reaction mechanism. The key findings denote that most of the reaction mechanisms have poor performance when predicting combustion characteristics of ammonia flames such as laminar flame speed, ignition delay time, and nitrogen oxide emissions (NOx). In addition, none of the mechanisms have been opti-mised efficiently to predict properly experimental measurements for all these combustion characteristics. For example, Duynslaegher's mechanism perfectly predicted the laminar flame speed at lean and stoichiometric conditions, while Nakamura's reaction mechanism worked properly at rich conditions for the estimation of laminar flame speed. Although the aforementioned mechanisms achieved good estimation in terms of laminar flame speed, they showed poor performance against NO mole fractions. Similarly, Glarborg's (2018) mechanism properly estimated NO mole fractions at lean and stoichiometric flames while Wang's mechanism performed well in rich conditions for such emissions. Other examples are presented in this manuscript. Finally, the prediction performance of the assessed mechanisms varies based on operating conditions, mixing ratios, and equivalence ratios. Most mechanisms dealing with blended NH3 combinations gave good predictions when the concentration of hydrogen was low, while deteriorating with increasing hydrogen concentrations; a result of the shift in re-actions that require more research.

Ammonia/hydrogen blends have received some attention toward the development of new technologies focused on gas turbine combustion systems, as doping of hydrogen in ammonia enhances flame speed and stability while decreasing ignition energy. One of the challenges of these blends relies on the appropriate computational modeling of their combustion properties in combination with the complex hydrodynamics inherent to flow control techniques such as swirling flows, which are known to be the main method of flame stabilization in current gas turbines. Moreover, it is well-known that large reaction kinetic models are difficult to employ in these computational analyses, thus increasing the difficulty of obtaining reliable methods for the design of new combustors. Therefore, this research analyses a reduced chemical reaction mechanism, namely Okafor’s mechanism, comparing its performance and accuracy against obtained experiments. Emission measurements and non-intrusive laser techniques (LDA) were employed to validate models running on CHEMKIN-PRO flow reactors and RANS Complex Chemistry. Once validated, the study identified the main contributors and reaction kinetics of NH2 and NO consumption, hence evaluating the process via production rates and sensitivity analysis of various important reactions. The results depicted positive correlation between NH2 formation and heat release, N2, H2O, N2O, NH, NNH, NO, O formation, whereas NH3, N2H3 and NO2 have shown negative correlation. Statistical correlations supported these findings but unfortunately were inconclusive to the impacts of vorticity and turbulence over the production/consumption of amidogen. Sensitivity analysis has shown NH2 radicals and atomic N to be the main contributors of NO formation in the flame zone, although most of the NO formed in the flame zone shown to be consumed at the post-flame zone due to the presence of NHx radicals and atomic N.

Variation in natural gas composition, alongside the potential for H2 enrichment, creates the potential for significant changes to premixed flame behaviour. To strengthen fundamental understanding of lean multi-component alternative fuel blends, an outwardly propagating spherical flame was employed to measure the flame speeds and Markstein lengths of C1C4 hydrocarbons, alongside precisely mixed blends of CH4/C2H6, CH4/C3H8 and CH4/H2. Theoretical relationships between Markstein length and Lewis Number are explored alongside effective Lewis number formulations. Under lean conditions, equal volumetric additions of H2 and C3H8 (30% vol.) to CH4 resulted in similar augmentation of burning velocity, however, opposite susceptibility to preferential diffusional instability was noted. At a fixed equivalence ratio of 0.65, limited changes in composition provide a marked change in the premixed flame response with the addition of C2H6 and C3H8 to CH4. For lean CH4/H2 mixtures, a diffusional based Lewis Number formulation yielded a favourable correlation, whilst a heat-release model resulted in better agreement for lean CH4/C3H8 blends. Modelling work suggests that measured enhancement of lean CH4 flames upon H2 or C3H8 is strongly correlated to changes in volumetric heat release rates and production of H radicals. Furthermore, a systematic analysis of the flame speed enhancement effects (thermal, kinetic, diffusive) of H2 and C3H8 addition to methane was undertaken. Augmented flame propagation of CH4/H2 and CH4/C3H8 was demonstrated to be principally an Arrhenius effect, predominantly through reduction of associated activation energy. Finally, plausible short-term variations in composition with hydrogen-enriched multi-component natural gas flames were investigated experimentally and numerically. At the leanest conditions, small variations in CH4:C3H8 content at a fixed H2 fraction resulted in discernible changes in stretch related behaviour, a reflection of the thermo-diffusive behaviour of each fuel's response.

Non-volatile Particulate Matter (nvPM) from aircraft gas turbine engines are harmful to both human health and the environment, but can be significantly reduced by using low aromatic Sustainable Aviation Fuel (SAF). As part of the Horizon 2020 funded JETSCREEN (JET fuel SCREENing and optimisation platform for alternative fuels) project, nvPM and gaseous emissions were characterised using regulatory compliant sampling and measurement methodologies for a small-scale (<250kW) non-proprietary RQL combustion rig, at pressures ranging from 1.0 to 2.4 bara. The impact of flow conditions, air to fuel ratio and fuel composition was investigated for a selection of conventional aviation Jet-A1 fuels, SAFs, and blended fuels. Measured concentrations were corrected for particle size-dependant system losses using particle size measurements, to be representative of combustor exit concentrations.Across the range of fuels (hydrogen contents 13.51%-15:31%), system-loss-corrected nvPM mass, number, and size were shown to decrease with increasing fuel hydrogen content, in agreement with previous studies. Inverse power law correlations are proposed as the best descriptors of these trends. Average reductions in nvPM mass, number, and sizes of 73%, 54% and 17% respectively, were observed for a near-zero aromatic ATJ fuel compared to a reference Jet A1 fuel, with minimal changes to measured gaseous pollutants. It is noted that without size dependant system loss corrections, nvPM number reductions were overreported (~6% for the ATJ fuel) due to the smaller particle sizes with increasing fuel hydrogen content. It was hypothesised that observed nvPM deviations from the fuel hydrogen content trends were due to fuel physical properties affecting atomisation, however no correlations were found greater than the measurement uncertainty and combustor rig variability.This study provides a unique dataset intended to facilitate combustion model validation, providing full details of combustor geometry, flow conditions and rig conditions, along with the representative combustor exit nvPM and gaseous data.

In the development of ammonia -hydrogen blends as potential substitutes for fossil fuels, the retrofitting of existing devices running at very lean condition is one of the promising solutions for decarbonisation of the power sector. However, little is known about the impact of these conditions on the production of NOX, particularly N2O that is a potent greenhouse gas. Therefore, the influence of varying thermal power and Reynolds numbers on the flame and emission characteristics, especially N2O, of ammonia-hydrogen -air swirling flames has been evaluated for the first time through the use of spatially resolved OH*, NH* and NH2* chemiluminescence, spectrometry analyses and advanced emissions characterisation at a fixed lean equivalence ratio, Phi = 0.65, representative of the Dry Low NOX (DLN) approach in traditional stationary gas turbines. NO and NO2 emissions were found to be decreasing (from similar to 5000 ppmv to similar to 1000 ppmv; NO and from similar to 150 ppmv to similar to 50 ppmv; NO2) with increasing ammonia content (from 50% to 90%) in the fuel while N2O followed reverse trends (from similar to 50 ppmv to similar to 200 ppmv). More than 80% ammonia content in the fuel blends exhibited high amounts of unreacted ammonia fractions (similar to 100 to similar to 1200 ppmv), which can be potentially linked to flame instability and/or low temperatures. Furthermore, any increasing or decreasing trends in NOX with ammonia fraction were made more extreme by increas-ing thermal power or Reynolds number due to the differences in relevant radicals (NH, OH, NH2 etc.) formation in the flames. Experimental results suggest the unviability of these blends at the conventional lean conditions utilised at the DLN power applications due to excessive NOX emissions. Detailed sensitivity analyses of N-2 O concentration at the flame and post flame zone has been carried out utilising Ansys Chemkin-PRO to identify and investigate the reactions responsible for N2O formation/consumption in the experimental flames. Results have identified the reaction NH + NO ? N-2 O + H as the major source of N2O production in the flame, while the reactions N2O + H ? N-2 + OH and N-2 O( + M ) ? N-2 + O ( + M ) are responsible for N-2 O consumption at the post flame zone, with higher reactivity for the latter reaction at longer residence time and relatively lower temperatures. (C) 2022 The Author(s). Published by Elsevier Inc. on behalf of The Combustion Institute.

Steelmaking is energy intensive, with manufacturing facilities representing some of the biggest point-source carbon dioxide (CO2) emitters in the UK. Efficiency improvements are essential with rising energy costs, driving significant investment from the UK iron and steel sector. However, the industry still finds it difficult to justify waste heat recovery (WHR) projects, as individual schemes incorporating waste heat capture and an end-use for the waste heat often incur high capital costs, resulting in long payback times. This paper describes the conceptual and numerical development of a strategy for the deployment of WHR using a large integrated steel works as a case study. An existing asset was utilised to link individual waste heat schemes together with a single end-user; thereby reducing the capital requirement for each subsequent project. The proposed strategy and its development is discussed, followed by the resultant carbon dioxide and energy savings (estimated to be 2.3 Mt and equivalent to £45 million), over the 6-year period since its implementation.