This study explores the macroscopic spray behaviour of methanol and its oxygenated blends MTBE5 (5 vol% methyl tert-butyl ether), A5 (5 vol% acetone), and E5 (5 vol% 2-ethylhexyl nitrate) at a fixed low ambient pressure of 40 mbar while varying liquid injection pressure from 40 to 100 bar. By keeping the flash-boiling environment and injector configuration fixed, the study provides a controlled comparison of how additive chemistry alters methanol spray development. High-speed Schlieren imaging and image-based analysis were used to examine, four key spray features; namely cone angle, spray penetration, spray area, and spray width. The results show that higher injection pressures generally promote broader and more uniform macroscopic spray development, however, the additive-specific response remained distinct under identical operating conditions. Among the blends, MTBE5 showed an approximate 8% increase in cone angle during injection, forming wide and stable sprays due to its favourable balance of low surface tension and sustained lateral plume growth. The A5 blend, by contrast, produced sprays with the strongest forward momentum and most consistent penetration, particularly at moderate pressures (40-80 bar), however, it showed reduced stability at the higher pressures. Pure methanol displayed narrower but moderately stable sprays, while E5 generated the widest sprays but with noticeable instability in the breakup region and spatial distribution. These results show that chemically distinct oxygenated additives do not modify methanol spray behaviour in the same way; rather, each additive produces a different balance of lateral expansion, axial penetration, and repeatability under the same flash-boiling condition. This provides a controlled basis for additive selection in methanol-fuelled injection systems.

Methanol and its oxygenated blends are credible low-carbon fuels for direct-injection engines, yet classical correlations predict their spray behaviour poorly across practical pressures. Penetration sets the risk of wall impingement and cone angle governs air entrainment, so the two must be predicted together. High-speed Schlieren imaging in a constant-volume vessel recorded sprays of pure methanol and of blends with methyl tert-butyl ether, 2-ethylhexyl nitrate or acetone at 1, 5 and 10 % by volume, at injection pressures of 2 to 10 MPa and ambient pressures of 4 to 200 kPa, spanning flash-boiling and non-flashing regimes. Averaging five repeats per condition gave 4320 time-resolved observations across 144 operating conditions. A multi-output backpropagation neural network predicted both descriptors simultaneously and was evaluated on entire withheld spray conditions rather than individual frames. On 21 unseen conditions the backpropagation model reached coefficients of determination of 0.994 for penetration and 0.989 for cone angle, with root-mean-square errors of 2.00 mm and 1.82 degrees, at or below measurement repeatability. A calibrated Hiroyasu-Arai correlation remained useful under non-flashing conditions but lost most of its skill under flash boiling, where the backpropagation model did not. Gradient-boosted trees predicted cone angle marginally better, although only the backpropagation model describes both descriptors at once. Optimising its weights with a genetic algorithm gave no gain in accuracy. Penetration was governed by injection and ambient pressure, cone angle by additive composition. Withholding an entire ambient pressure level or additive concentration drove the coefficient of determination below zero, defining the model's practical limits.

A methodology based on high-speed photography has been developed to measure local droplet size and velocity distributions across the entire flow domain of a continuous, quasi-steady flat-fan water spray. The motivation is to complement phase Doppler anemometry (PDA) point measurements, which are limited to the far-field, dilute region of sprays. The nozzle was systematically shifted to capture a series of non-overlapping images of the spray, which were subsequently 'stitched' to form composite images. To reconstruct distributions along the stiches, four grids were used, each offset vertically and/or horizontally. Image analysis was employed to quantitatively evaluate the primary breakup zone and reconstruct droplet size and velocity distributions at all locations. Results showed that droplet sizes and velocities decrease downstream, with larger droplets and higher velocities observed at the periphery, attributed to rim tearing. Droplet sizes ranged from 10 to 250 mu m, with velocities between 6 and 24 m/s. The local droplet size distributions were found to follow a lognormal distribution, consistent with previous observations. The local velocity distributions within the main spray region exhibited a Gaussian profile, whereas those along the spray periphery displayed a negatively skewed distribution, varying from monomodal to bimodal. At lower flow rates, the breakup zone contracted and shifted further from the nozzle. Rim tearing became dominant, leading to increased polydispersity and velocity fluctuations along the periphery.

This aim of this study is to investigate local droplet size and velocity distributions in a near two-dimensional quasi-steady flat-fan water spray injected under quiescent conditions. The objective is to provide accurate experimental data to validate the generalised Fully Lagrangian Approach (gFLA) model for polydisperse droplet flows. To achieve this, droplet size and velocity distributions were characterised using classical phase Doppler anemometry (PDA) and a recently-developed imaging framework based on high-speed shadowgraphy in conjunction with particle image velocimetry (PIV). The latter technique can also be referred to as high speed videography (HSV). Average horizontal velocity components obtained from both techniques are in good agreement within the main body of the spray. However, average vertical velocity components measured using HSV were consistently lower than those from PDA by up to 20%. This discrepancy is primarily attributed to the broader dynamic range of PDA, allowing it to capture a wider spectrum of velocities, whereas HSV is constrained by the temporal and spatial resolution of the optical setup. Both techniques showed that droplet number median diameter decreases downstream, due to secondary and tertiary atomisation, though PDA consistently reported larger values compared to HSV by up to 15%. These differences stem from the limited size range of each method, as HSV measures droplet diameters from 10-300 mu m, while PDA is restricted to 2-120 mu m. These findings highlight the importance of selecting an appropriate combination of measurement techniques for spray characterisation. PDA is well-suited for velocity measurements due to its high temporal resolution, while a hybrid approach incorporating both PDA and HSV for droplet size distributions offers a more comprehensive representation of the spray field.

Ammonia (NH3) is gaining recognition as a viable “green” transportation fuel due to its zero-carbon characteristic, its high energy density and its widespread availability. However, NH3 has a high auto-ignition temperature, resulting in potential emissions of NOx and unburned NH3. Addressing combustion challenges requires innovative solutions, such as the application of combustion promoters to enhance NH3 combustibility. This review article focuses on the compatibility of NH3 as a fuel for spark-ignition (SI) engines, examining its combustion under various modes including pure NH3 combustion, gasoline blends, NH3/hydrogen (H2) blends, and NH3/natural gas blends in single or dual-fuel configurations. The formation of nitrogen oxides (NOx) and slip-NH3 is explored to understand emissions species such as NO and N2O. Additionally, the article highlights the limitations of NH3 as a fuel for SI combustion. The comprehensive discussion provided in this review aims to fill a critical gap in the literature regarding NH3’s feasibility as a zero-carbon fuel for SI engines, particularly in the maritime sector. By offering insights into NH3 combustion characteristics and emissions profiles, the article seeks to provide a roadmap for leveraging NH3 as a suitable non-carbon fuel to decarbonize the marine sector and advance global sustainability goals.

Atomisation in sprays exhibits a stochastic nature, yet it underpins the functionality of technologies reliant on sprays. To accommodate this inherent randomness, it is important to assess all possible outcomes of droplet sizes and velocities at each location, which is crucial for advancing spray-based applications. Phase Doppler Anemometry (PDA) is typically employed to measure these properties, though its use is largely restricted to the far-field, dilute region of a spray where droplets are generally smaller and spherical compared to those in the breakup zone. A new approach involving high-resolution, high-speed, microscopic imaging was developed to complement PDA measurements. Firstly, in this method, a composite image of the spray was constructed by translating the small field of view across the entire spray region in steps, resulting in a set of non-overlapping images. Along the boundaries, the artefacts due to composition were taken care of by utilising four grids, each offset vertically and/or horizontally by half the height and/or width of the field of view to reconstruct data along the ’stitches’. Secondly, the image processing algorithm enabled identification of perforations in the liquid sheet, breakup region and local droplet size and velocity distributions in the spray. In total, 1,200 composite images were produced for each grid at a frame rate of 84 kHz. Droplet sizes greater than 10 μm could be measured, with PDA extending the measurable range down to 1-2 μm. In addition, particle image velocimetry was performed on temporally resolved composite images to derive velocities in the field. The characteristics of a steady-state flat fan water spray at a flow rate of 5 kg/h were analysed using both methodologies. The findings indicated a strong qualitative agreement between the techniques, showing consistent trends, though some quantitative discrepancies were noted. The velocity measurements varied by up to 20% between the two methods. These variations may stem from the exclusion of non-spherical droplets in PDA or from droplets within the depth of field in the images, which move slowly and lead to bias during image analysis.

In the case of a public health emergency such as the COVID-19 pandemic, access to large quantities of appropriate personal protection equipment (PPE) has presented a significant problem. A shortage of face masks and respirators has been widely reported across the world. A concerted effort to manufacture high volumes has not unsurprisingly put pressure on the supply chain and the important certification processes. PPE procured or donated as uncertified stock requires rigorous, expedient and scientifically informed evidence before decisions can be made regarding suitable deployment, expensive certification, return or possible destruction of stock. This paper reports a series of experiments devised in reaction to this situation. In this study, an experimental methodology for the assessment of the filtration performance of samples of real-world, uncertified, fluid resistant surgical masks (FRSM type IIR) was evaluated in the resource limited (lockdown) environment of the COVID-19 pandemic. A steady-state flow rig was adapted to incorporate a bespoke filter flow chamber for mounting face masks in order to evaluate the resistance to air flow as an indicator of mask inhalation performance. Pure air was drawn through a specified control surface area at known flow rate conditions; the resistance to the air flow through the masks was measured as the resulting pressure drop. Over 60 tests were performed from 4 different, randomly sampled batches and compared to a control sample of EN Type IIR certified FRSM masks. Steady-state volumetric airflow rates of 30 and 95 lmin -1 were chosen to represent deep breathing and vigorous exercise conditions respectively. The results showed that the sample masks produced a pressure drop of between 26% to 58% compared to the control batch at the lower flow rate and 22% to 55% at the higher rate. The results for each sample were consistent across both flow rates. Within the group of masks tested, two sets (between 48% and 58% of the reference set) showed the potential to be professionally assessed for appropriate deployment in a suitable setting. Although the absolute values of pressure drop measured by this method are unlikely to correlate with other testing approaches, the observed, indicative trends and relative performance of the masks is key to this approach. Critically, this method does not replace certification but it has enabled a public body to quickly make decisions; certify, re-assign, refund, thus saving time and resources. The total time spent conducting the tests was less than 8 hours and the low cost method proposed can be repurposed for low resource regions.

Nanofluids are metallic or nonmetallic, nanometer-sized particles dispersed in liquid. They can be used in various fields to increase heat transfer rates, as the thermal conductivity of nanofluids can be increased significantly. Nanofluids may be used as a good alternative coolant in spray cooling applications. This study conducted experiments to compare spray characteristics, such as droplet diameters and velocities, between water and alumina nanofluid sprays. The mass ratio of alumina nanoparticles was varied from 0.2 to 0.5 weight percentages (wt.%) and the spray injection pressure was varied between 0.2 and 0.3 MPa. The local distributions of droplet sizes and velocities along the spray axial and radial directions were measured by a laser doppler instrument. Generally, the spray characteristics of nanofluid sprays is significantly different from that of water sprays. The average droplet diameters of the fluids tested increased in an approximately linear manner with the increase in the mass ratio of nanoparticles up to 0.4 wt.%, whereas the average droplet velocities decreased. In the case of the nanofluid spray of 0.5 wt.%, the increase in droplet diameters and the decrease in droplet velocities were much more marked, departing from the linear relationship. This unusual behavior could also be observed in the local distributions of droplet diameters and velocities along the axial and radial directions. Further research studies are required to reveal how the addition of nanoparticles affects the atomization mechanism of nanofluids. The difference in the spray characteristics of nanofluid sprays from that of water sprays should be taken into consideration when the cooling effectiveness of nanofluids and water in spray cooling is compared.

We present an Adaptive Mesh Refinement (AMR) method suitable for hybrid unstructured meshes that allows for local refinement and de-refinement of the computational grid during the evolution of the flow. The adaptive implementation of the Discontinuous Galerkin (DG) method introduced in this work (ForestDG) is based on a topological representation of the computational mesh by a hierarchical structure consisting of oct- quad- and binary trees. Adaptive mesh refinement (h-refinement) enables us to increase the spatial resolution of the computational mesh in the vicinity of the points of interest such as interfaces, geometrical features, or flow discontinuities. The local increase in the expansion order (p-refinement) at areas of high strain rates or vorticity magnitude results in an increase of the order of accuracy in the region of shear layers and vortices. A graph of unitarian-trees, representing hexahedral, prismatic and tetrahedral elements is used for the representation of the initial domain. The ancestral elements of the mesh can be split into self-similar elements allowing each tree to grow branches to an arbitrary level of refinement. The connectivity of the elements, their genealogy and their partitioning are described by linked lists of pointers. An explicit calculation of these relations, presented in this paper, facilitates the on-the-fly splitting, merging and repartitioning of the computational mesh by rearranging the links of each node of the tree with a minimal computational overhead. The modal basis used in the DG implementation facilitates the mapping of the fluxes across the non conformal faces. The AMR methodology is presented and assessed using a series of inviscid and viscous test cases. Also, the AMR methodology is used for the modelling of the interaction between droplets and the carrier phase in a two-phase flow. This approach is applied to the analysis of a spray injected into a chamber of quiescent air, using the Eulerian–Lagrangian approach. This enables us to refine the computational mesh in the vicinity of the droplet parcels and accurately resolve the coupling between the two phases.

A transient axially symmetric two-phase vortex-ring flow is investigated using the one-way coupled, two-fluid approach. The carrier phase parameters are calculated using the approximate analytical solution suggested by Kaplanski and Rudi (Phys. Fluids vol. 17 (2005) 087101–087107). Due to the vortical nature of the flow, the mixing of inertial admixture can be accompanied by crossing particle trajectories. The admixture parameters are calculated using the Fully Lagrangian Approach (FLA). According to FLA, all of the dispersed phase parameters, including the particle/droplet concentration, are calculated from the solution to the system of ordinary differential equations along chosen particle trajectories; FLA provides high-accuracy particle number calculations even in the case of crossing particle trajectories (multi-valued fields of the dispersed media). Two flow regimes corresponding to two different initial conditions are investigated: (i) injection of a two-phase jet; and (ii) propagation of a vortex ring through a cloud of particles. It was shown that the dispersed media may form folds and caustics in these flows. In both cases, the ranges of governing parameters leading to the formation of mushroom-like clouds of particles are identified. The caps of the mushrooms contain caustics or edges of folds of the dispersed media, which correspond to particle accumulation zones.

A review of an extensive set of engine testing data was conducted in order to study low-speed, pre-ignition (LSPI) phenomena observed in turbo-charged, direct injection gasoline engines. A wide range of engine makes and sizes were tested by Ricardo Ltd. Analysis of the test results, further in-house experiments and inferences drawn from the wider literature, were used to estimate the impact of a range of variables upon LSPI. A table of weighted variables was developed in order to determine the conditions required to initiate LSPI on the engine test bed. This has led to shorter test times during durability testing as well as a better understanding of the mechanisms behind LSPI. Some challenges faced when trying to accurately and consistently measure this phenomenon are outlined. Finally, a method to identify LSPI is proposed along with reconunendations for suitable control responses to LSPI events encountered in a vehicle application.

A new approach to modelling the interaction between droplets and the carrier phase is suggested. The new model isapplied to the analysis of a spray injected into a chamber of quiescent air, using an Eulerian-Lagrangian approach. The conservative formulation of the equations for mass, momentum and energy transport is used for the analysis of the carrier phase. The dispersed phase is modelled using the Lagrangian approach with droplets represented by individual parcels.The implementation of the Discontinuous Galerkin method (ForestDG), based on a topological representation of the computational mesh by a hierarchical structure consisting of oct- quad- and binary trees, is used in our analysis. Adaptive mesh refinement (h-refinement) enables us to increase the spatial resolution for the computational mesh in the vicinity of the points of interest such as interfaces, geometrical features, or flow discontinuities. The local increase in the expansion order (p-refinement) at areas of high strain rates or vorticity magnitude results in an increase of the order of the accuracy of discretisation of shear layers and vortices.The initial domain consists of a graph of unitarian-trees representing hexahedral, prismatic and tetrahedral elements. The ancestral elements of the mesh can be split into self-similar elements allowing each tree to grow branches to an arbitrary level of refinement. The connectivity of the elements, their genealogy and their partitioning are described by linked lists of pointers. These are attached to the tree data structure which facilitates the on-the-fly splitting, merging and repartitioning of the computational mesh by rearranging the links of each node of the tree. This enables us to refine the computational mesh in the vicinity of the droplet parcels aiming to accurately resolve the coupling betweenthe two phases.DOI: http://dx.doi.org/10.4995/ILASS2017.2017.4671

The fully Lagrangian approach (FLA) to the calculation of the number density of inertial particles in dilute gas-particle flows is implemented into the CFD code ANSYS Fluent. The new version of ANSYS Fluent is applied to modeling dilute gas-particle flow around a cylinder and liquid droplets in a gasoline fuel spray. In a steady-state case, the predictions of the FLA for the flow around a cylinder and those based on the equilibrium Eulerian method (EE) are almost identical for small Stokes number, Stk, and small Reynolds number, Re, (Re = 1, Stk = 0.05). For the larger values of these numbers (Re = 10, 100; Stk = 0.1, 0.2) the FLA predicts higher values of the gradients of particle number densities in front of the cylinder compared with the ones predicted by the EE. For transient flows (Re = 200), both methods predict high values of the number densities between the regions of high vorticity and very low values in the vortex cores. For Stk >= 0.1 the maximal values predicted by FLA are shown to be several orders of magnitude higher than those predicted by the EE. An application of FLA to a direct injection gasoline fuel spray has focused on the calculation of the number densities of droplets. Results show good qualitative agreement between the numerical simulation and experimental observations. It is shown that small droplets with diameters d(p) = 2 mu m tend to accumulate in the regions of trajectory intersections more readily, when compared with larger droplets (d(p) = 10 mu m, d(p) = 20 mu m). This leads to the prediction of the regions of high number densities of small droplets.

This paper introduces a minimum viable software product to filter large datasets of engine data recorded during laboratory experiments of combustion engines. The aim is to support analysts in the identification and analysis of specific physical phenomenon within hours of recorded engine experimental data. Specifically, the tool has been designed considering the use case of identifying Low Speed Pre-Ignition events. This work describes the tool's graphical user interface and its scalable architecture based on mainstream web and big-data technologies as well as the practical application to pre-ignition events identification. The paper provides details on the architecture's performance, providing evidence of its scalability by increasing the number of available computing workers.

In this paper, a continuous-discrete time observer which simultaneously estimates the unmeasurable states and the uncertainties for the electro-hydraulic actuator (EHA) system is presented. The main feature of the proposed observer is the use of an intersample output predictor which allows the users to increase the frequency acquisition of the piston position sensor without affecting the convergence performance. The stability analysis of the proposed observer is proved using Lyapunov function adapted to hybrid systems. To show the efficiency of our proposed observer, numerical simulations and experimental validation involving a control application, which combines the designed observer and a proportional-integral (PI) controller for the purpose of piston position tracking problem, are presented.

The paper presents an overview of recent progress in the modelling of vortex rings and vortex ring-like structures with specific potential applications to direct injection gasoline engines. A new theoretical model for a confined axisymmetric vortex ring is described. The predictions of this model are shown to be in agreement with available experimental data and numerical simulations. Direct numerical simulations (DNS) are used to test the range of applicability of the model and to investigate new physical features of confined vortex rings recently reported in the experimental studies. A meshless method for modelling of 2D transient, non-isothermal, two-phase flows with phase transitions, based on a combination of the viscous-vortex and thermal-blob methods for the carrier phase with the Lagrangian approach for the dispersed phase, is briefly described. The flow parameters were found from the solution of three systems of ordinary differential equations, describing the motion of viscous-vortex blobs, thermal blobs and droplets. Potential applications of the results to modelling the processes in direct injection gasoline engines are discussed.