Large-scale Computational Fluid Dynamics Research Articles

CFD simulation of a multi-rotor system using diffuser augmented wind turbines by lattice Boltzmann method

A diffuser-augmented wind turbine (DAWT) achieves greater power generation efficiency by increasing wind speed through the diffuser. Nevertheless, scaling up this technology is difficult because of the considerable amount of wind drag on the diffusers. To overcome this difficulty, a multi-rotor system with two or more wind turbines on the same structure is one approach to increasing wind turbine power output. This research proposes a computational fluid dynamics (CFD) method to evaluate the hydrodynamic performance of a large-scale multi-rotor system of DAWTs. Compared to conventional wind turbines, CFD simulations of DAWTs necessitate higher computational costs because of the need of high-resolution meshes for the diffuser. Furthermore, the computational cost of a multi-rotor system increases with the number of rotors. To address the issue of high computational cost, we use the Lattice Boltzmann Method (LBM), which is well suited to large-scale CFD simulations. A wind turbine is modeled as an actuator line model and a diffuser as a wall boundary. An adaptive mesh refinement approach generates higher resolution meshes near the rotor and diffuser. LBM simulations were conducted for a single DAWT and a multi-rotor system with five DAWTs. The LBM results of the wake velocity and pressure distributions were in agreement with those obtained from wind tunnel experiments and general CFD methods in earlier studies. To investigate the diffuser gap effects, we simulated five DAWTs with diffuser gaps of 5%–25% of the diffuser diameter. The power gain of each DAWT was assessed. Great performance improvements were found with diffuser gaps of 20% and 25% of the diffuser diameter. On average, the five DAWTs achieved a power gain of more than 10%. These findings confirmed the accurate evaluation capability of the proposed CFD method for hydrodynamic characteristics of multi-rotor systems using DAWTs.

Automated surgery planning for an obstructed nose by combining computational fluid dynamics with reinforcement learning

Septoplasty and turbinectomy are among the most common interventions in the field of rhinology. Their constantly debated success rates and the lack of quantitative flow data of the entire nasal airway for planning the surgery necessitate methodological improvement. Thus, physics-based surgery planning is highly desirable. In this work, a novel and accurate method is developed to enhance surgery planning by physical aspects of respiration, i.e., to plan anti-obstructive surgery, for the first time a reinforcement learning algorithm is combined with large-scale computational fluid dynamics simulations. The method is integrated into an automated pipeline based on computed tomography imaging. The proposed surgical intervention is compared to a surgeon’s initial plan, or the maximum possible intervention, which allows the quantitative evaluation of the intended surgery. Two criteria are considered: (i) the capability to supply the nasal airway with air expressed by the pressure loss and (ii) the capability to heat incoming air represented by the temperature increase. For a test patient suffering from a deviated septum near the nostrils and a bony spur further downstream, the method recommends surgical interventions exactly at these locations. For equal weights on the two criteria (i) and (ii), the algorithm proposes a slightly weaker correction of the deviated septum at the first location, compared to the surgeon’s plan. At the second location, the algorithm proposes to keep the bony spur. For a larger weight on criterion (i), the algorithm tends to widen the nasal passage by removing the bony spur. For a larger weight on criterion (ii), the algorithm’s suggestion approaches the pre-surgical state with narrowed channels that favor heat transfer. A second patient is investigated that suffers from enlarged turbinates in the left nasal passage. For equal weights on the two criteria (i) and (ii), the algorithm proposes a nearly complete removal of the inferior turbinate, and a moderate reduction of the middle turbinate. An increased weight on criterion (i) leads to an additional reduction of the middle turbinate, and a larger weight on criterion (ii) yields a solution with only slight reductions of both turbinates, i.e., focusing on a sufficient heat exchange between incoming air and the air-nose interface. The proposed method has the potential to improve the success rates of the aforementioned surgeries and can be extended to further biomedical flows.

Reduced-order equivalent source modeling of aeroacoustic sources

The acoustic far-field prediction of high-fidelity aerodynamic simulation data is still not accurately possible when accounting for the surface reflections from an aircraft’s fuselage or wings. By presenting a general method for order-reduced equivalent source modeling of aeroacoustic sound sources, this work is bridging the gap between high-fidelity aerodynamic simulations and its integration to auralization frameworks. At the example of a numerically simulated jet, this study shows the computation of an equivalent source model with minimal complexity by means of spherical harmonic (SH) coefficients. The minimal source extension, in which all acoustic sources of a flow field are confined, is computed by using spherical Hankel extrapolation of sound pressure data from virtual concentric microphone arrays. The result of the SH transform shows that the dominant energy contribution can be associated to nine elementary sound sources. The resulting equivalent source model of jet noise provides a convenient date interface between large-scale computational fluid dynamics and acoustics simulations.

Quasi-Global Sensitivity Analysis of Supersonic Computational Fluid Dynamics Simulations

Determining the sensitivity of model outputs to input parameters is an important precursor to developing informative parameter studies, building surrogate models, and performing rigorous uncertainty quantification. Determining parameter sensitivities over a range of parameter values, termed global sensitivity analysis, requires many model evaluations sampled over the parameter space, which is intractable for many large-scale computational fluid dynamics (CFD) applications. For moderate parameter dimensions, we propose the use of Morris screening, a one-at-a-time quasi-global method for estimating parameter sensitivities over a range of parameter values for CFD simulations. The Morris method is implemented within a CFD framework that utilizes adaptive grid refinement, thereby enabling its application to state-of-the-art production-level problems. The method is shown to be viable for a model problem of supersonic flow over an axisymmetric capsule geometry with both physical and nonphysical (i.e., simulation-informing) input parameters. It is shown that Morris screening identifies relative parameter sensitivities consistent with those of more costly experimental and computational studies that were previously performed on this geometry.

Family of Skeletal Reaction Mechanisms for Methane–Oxygen Combustion in Rocket Propulsion

Analyzing methane–oxygen rocket propellant combinations requires suitable modeling of the major chemical reaction processes. Although several detailed kinetic mechanisms for methane oxidation in air exist, most do not reproduce the reaction pathways of high-pressure methane–oxygen combustion, typical of liquid rocket engines. Moreover, when large-scale computational fluid dynamics simulations are pursued, detailed reaction schemes are not computationally viable. In the present study, we identify a reliable detailed kinetic scheme for liquid rocket applications, and then we perform a wide reduction campaign leveraging computational singular perturbation theory. Enforcing various reduction targets, we obtain a family of seven skeletal schemes, including 11–39 species. Each mechanism targets different combustion modes, namely, homogeneous ignition, complex flows and flame extinction, premixed burning, reaction processes under intense turbulent mixing, and largely off-stoichiometric mixtures, typical of rocket engine preburners. We test the skeletal mechanisms against meaningful validation targets, attaining appreciable predictive accuracy compared with the detailed parent scheme. We expect the proposed family of skeletal schemes to offer a wide and flexible range of solutions—in terms of size, accuracy, and dominant combustion mode—for performing large-scale yet cost-affordable computational fluid dynamics of methane–oxygen flames under rocket-engine-relevant conditions.

Interactive steering on in situ particle-based volume rendering framework

The development of supercomputers and multi-scale computational fluid dynamics (CFD) models based on adaptive mesh refinement (AMR) enabled fast, large-scale, and high fidelity CFD simulations. Interactive in situ steering is an effective tool for debugging, searching for optimal solutions, and analyzing inverse problems in such CFD simulations. We propose an interactive in situ steering framework for large-scale CFD simulations on GPU supercomputers. This framework employs in situ particle-based volume rendering (PBVR), in situ data sampling, and a file-based control that enables interactive and asynchronous communication of steering parameters, compressed visualization particle data, and sampled monitoring data between supercomputers and user PCs. The parallelized PBVR is processed on the host CPU to avoid interference with CFD simulations on the GPU. We apply the proposed framework to a real-time plume dispersion analysis code CityLBM, which computes the lattice Boltzmann method on the block AMR grid using GPU supercomputers. In the numerical experiment, we address an inverse problem to find a pollutant source from the observation data at monitoring points and demonstrate the effectiveness of the human-in-the-loop approach via the in situ steering framework.Graphical abstract

CFD Model Verification and Aerodynamic Analysis in Large-Scaled Venlo Greenhouse for Tomato Cultivation

To address the challenges of climate change and food security, the establishment of smart farm complexes is necessary. While there have been numerous studies on the productivity and environmental control of individual greenhouses, research on greenhouse complexes is considerably limited. Conducting environmental studies during the design phase of these complexes poses financial constraints and practical limitations in terms of on-site experiments. To identify potential issues that may arise when developing large-scale greenhouse complexes, it is possible to utilize modeling techniques using computational fluid dynamics (CFD) to assess environmental concerns and location issues before constructing the facilities. Consequently, simulating large-scale CFD models that incorporate multiple greenhouses and atmospheric conditions simultaneously presents significant numerical challenges. The objective of this study was to design and verify the 3D CFD model for a large-scale Venlo greenhouse, where acquiring field data before construction is not feasible for designing a greenhouse complex. The verification of the CFD models was conducted using the improved grid independence test (GIT) and wall Y+ approaches. The findings revealed that a grid resolution of 0.8 m and a first-layer height of 0.04 m were suitable for developing large-scale greenhouse models, resulting in a low Root Mean Square Error (RMSE) of 3.9% and a high coefficient of determination (R2) of 0.968. This process led to a significant reduction of 38% in the number of grid cells. Subsequently, the aerodynamic characteristics and regional ventilation efficiency were analyzed in a 3D greenhouse model for developing a new large-scale greenhouse complex.

Acoustic source characterization of simulated subsonic jet noise using spherical harmonics

As subsonic jets remain one of the major contributions to aircraft noise emissions, near-field flow simulations should be included in aircraft design at an early stage using quantitatively predicted sound pressure levels and the time-domain signal properties of the noise data. In this regard, the interface from the near-field data to the far-field radiation—under consideration of acoustic reflections from objects such as fuselage and wings—remains as bottleneck. This study presents the computation of a spherical equivalent source model of jet noise with minimal complexity by means of spherical harmonic (SH) coefficients. Using spherical Hankel extrapolation of sound pressure data from virtual, concentrical microphone arrays, the results of the determination for the radius, in which all acoustic sources of a flow field are confined, indicate the source radius around the end of the potential core to be equivalent to five times the nozzle diameter. The result of the SH transform shows the dominant energy contribution to be related to nine elementary sources. The resulting equivalent source model of jet noise provides a convenient format for further use in large-scale computational fluid dynamics simulations.

Cloud benchmarking and performance analysis of an HPC application in Amazon EC2

Cloud computing platforms have been continuously evolving. Features such as the Elastic Fabric Adapter (EFA) in the Amazon Web Services (AWS) platform have brought yet another revolution in the High Performance Computing (HPC) world, further accelerating the convergence of HPC and cloud computing. Other public clouds also support similar features further fueling this change. In this paper, we show how and why the performance of a large-scale computational fluid dynamics (CFD) HPC application on AWS competes very closely with the one on Beskow—a Cray XC40 supercomputer at the PDC Center for High-Performance Computing - in terms of cost-efficiency with strong scaling up to 2304 processes. We perform an extensive set of micro and macro benchmarks in both environments and conduct a comparative analysis. Until as recently as 2020 these benchmarks have notoriously yielded unsatisfactory results for the cloud platforms compared with on-premise infrastructures. Our aim is to access the HPC capabilities of the cloud, and in general to demonstrate how researchers can scale and evaluate the performance of their application in the cloud.

Experimental and numerical study of CO2 plume diffusion in a confined space

In this paper, heavy gas diffusion in a confined space has been investigated. The effects of barrier and source intensity on CO2 diffusion are explored by the small-scale experiments and computational fluid dynamics (CFD) methods. Six different turbulence models are selected to predict the gas concentrations. By comparing these experimental values with the simulated ones, it is found that all models can effectively predict the concentration variation with time, and SST k-ω model is most close to the ideal model compared with others. Three source-barrier distances and three CO2 flow rates have been set up for the study. In this confined space, the main flow is concentrated in the region near the ground. The existence of barriers in the space will have a dilution effect on the high-concentration plume near the ground of the near-source area and a barrier effect on the low-concentration plume in the far-source area. The changes in source intensity have notable impact on the gas concentrations. This study can provide an experimental basis for the risk assessment in the confined spaces, as well as an experimental and data reference for large-scale CFD simulations.

Flow Field Reconstructions With GANs Based on Radial Basis Functions

Nonlinear sparse data regression and generation have been a long-term challenge in the field of aerodynamics, flow field reconstruction is an area of specific interest in this article. The high computational costs of computational fluid dynamics (CFD) make large scale CFD data production expensive, which is the reason why cheaper methods are needed. Traditional reduced-order models were promising but they cannot generate a large amount of full domain flow field data (FFD) to execute high-precision flow field reconstructions. Motivated by the problem of existing approaches, and inspired by the success of generative adversarial networks (GANs) in the field of computer vision, we prove a theorem that shows the optimal approximation to a GAN discriminator is a radial basis function neural network when engaged in with nonlinear sparse FFD regression and generation. Based on this theorem, a radial basis function-based GAN (RBF-GAN) and a RBF cluster-based GAN (RBFC-GAN) are proposed for regression and generation purposes. Three different datasets are applied to verify the feasibility of our models. The results show that the performance of RBF-GAN and RBFC-GAN are better than that of GANs and conditional GANs (cGANs) by both the mean square error and the MSPE measurements. In addition, compared with GANs/cGANs, the stability of the RBF-GAN and the RBFC-GAN are improved by 34.62 and 72.31%, respectively. Consequently, our proposed models can be used to generate full domain FFD from limited and sparse datasets to meet the requirements of high-precision flow field reconstructions.

Development of machine learning models for the prediction of laminar flame speeds of hydrocarbon and oxygenated fuels

Laminar flame speed (LFS) is a key physicochemical property of a premixed fuel/oxidizer mixture, and is critical in the description of complex combustion phenomena. Accurate experimental measurements of LFSs for various fuels have been performed to develop and validate detailed kinetic mechanisms, which in turn are used to predict LFSs under various combustion conditions. However, such procedure is inefficient, especially in large-scale turbulent combustion modeling studies. Based on previous experimental studies of LFSs for various fuels, this work aims to develop a data-driven machine learning (ML) model for the prediction of LFSs of hydrocarbon and oxygenated fuels. Descriptors computed from semi-empirical quantum chemistry methods are used as input in ML models due to the simplicity and computational-efficiency. Pearson correlation analysis is used to select important features, and 5 descriptors are screened as the input features for ML model development. The accuracies and interpretabilities of existing 16 ML algorithms in the prediction of LFSs are compared through systematically evaluated the errors based on the differences between experimental data and model prediction. These ML models include regression trees, support vector machine regression, gaussian process regression, and ensemble trees. An efficient ML model for predicting LFSs of hydrocarbon and oxygenated fuels based on gaussian process regression algorithm is proposed, which exhibits good accuracy in predicting of LFSs for variable pressure, temperature, and equivalence ratio. The dependency of LFSs on the descriptors are also analysed. The developed ML model is fast enough for integration into large-scale computational fluid dynamics for combustion studies.

Genetic Optimisation of a Free-Stream Water Wheel Using 2D Computational Fluid Dynamics Simulations Points towards Design with Fully Immersed Blades

A large-scale two-dimensional computational fluid dynamics study is conducted in order to maximise the power output and smoothness of power delivery of a free-stream water wheel, a low-impact hydropower device. Based on models and methods developed in previous research, the study uses a genetic algorithm to optimise the geometry of a wheel with a given radius and depth, maximising two objective functions simultaneously. After convergence and suitable post-processing, a single optimal design is identified, featuring eight shortened blades that become fully immersed at the nadir point. The design results in a 71% reduction in blade material and a 113% increase in the work ratio while improving the hydraulic power by 8% compared to the previous best design. These characteristics are applied retroactively to a broad family of designs, resulting in significant improvements in performance. Analysis of the resulting designs indicates that when either the hydraulic power coefficient, rotor power coefficient, or work ratio is considered, free-stream water wheels with fully immersed blades, whose power mechanisms are shown to rely on lift, as well as drag, outperform all other designs studied so far.

Patient-Specific Computational Fluid Dynamics Reveal Localized Flow Patterns Predictive of Post-Left Ventricular Assist Device Aortic Incompetence.

Progressive aortic valve disease has remained a persistent cause of concern in patients with left ventricular assist devices. Aortic incompetence (AI) is a known predictor of both mortality and readmissions in this patient population and remains a challenging clinical problem. Ten left ventricular assist device patients with de novo aortic regurgitation and 19 control left ventricular assist device patients were identified. Three-dimensional models of patients' aortas were created from their computed tomography scans, following which large-scale patient-specific computational fluid dynamics simulations were performed with physiologically accurate boundary conditions using the SimVascular flow solver. The spatial distributions of time-averaged wall shear stress and oscillatory shear index show no significant differences in the aortic root in patients with and without AI (mean difference, 0.67 dyne/cm2 [95% CI, -0.51 to 1.85]; P=0.23). Oscillatory shear index was also not significantly different between both groups of patients (mean difference, 0.03 [95% CI, -0.07 to 0.019]; P=0.22). The localized wall shear stress on the leaflet tips was significantly higher in the AI group than the non-AI group (1.62 versus 1.35 dyne/cm2; mean difference [95% CI, 0.15-0.39]; P<0.001), whereas oscillatory shear index was not significantly different between both groups (95% CI, -0.009 to 0.001; P=0.17). Computational fluid dynamics serves a unique role in studying the hemodynamic features in left ventricular assist device patients where 4-dimensional magnetic resonance imaging remains unfeasible. Contrary to the widely accepted notions of highly disturbed flow, in this study, we demonstrate that the aortic root is a region of relatively stagnant flow. We further identified localized hemodynamic features in the aortic root that challenge our understanding of how AI develops in this patient population.

A non-linear transform approach for conduction-radiation heat transfer in the extended thermal discrete element method

Conduction-radiation heat transfer widely exists in nuclear pebble beds, packed beds and powder materials under moderate or high temperatures. Because of the considerable computational cost, it is unrealistic to trace all possible obstructed view factors between spheres in transient particle-scale simulations. With a non-linear transform approach, an extended thermal discrete element method (TDEM) with almost the same computational cost as that required for conduction alone is proposed here. Particle-particle conduction in the contact area, particle-fluid-particle conduction near the contact point and the Knudsen effect under low gas pressures are considered in the determination of the contact thermal resistance (CTR). A particle conduction-radiation scalar is applied in the particle energy balance equation, which is a function of the particle temperature, CTR and radiation exchange factor. The numerical results of the pebble bed are in good agreement with the continuum model and measured results under high temperatures. The temperature transport process in conduction-radiation heat transfer involves anomalous diffusion by discrete and continuum methods. The proposed extended TDEM is proven to be efficient for conduction-radiation heat transfer and feasible to be applied in large-scale computational fluid dynamics (CFD)-DEM simulations.

Experimental and numerical investigations of water entry in subsea modules with porous structures

Slamming impacts on subsea modules used for oil and gas exploration and production have become an increasingly important structural safety issue. There are many aspects of the subsea module water entry problem that have not yet been sufficiently explored, and these include the shape of the module and its impact velocity. In this study, a large-scale model and computational fluid dynamics (CFD) simulations were used to explore these aspects, with a focus on their role in porous plates. The numerical simulation results for vertical forces showed good agreement with the experimental results. The CFD simulations, numerical methods, and computing grids can be used to accurately and efficiently determine the slamming loads both in the time domain and at specified points. The critical vertical speed snap load was 0.093 m/s for the studied module. The air layer under the porous plates plays an important role in the slamming impact. The maximum delay in slamming caused by the porous plates was 79% at a vertical speed of 0.1 m/s.

Model Reduction for Steady Hypersonic Aerodynamics via Conservative Manifold Least-Squares Petrov–Galerkin Projection

High-speed aerospace engineering applications rely heavily on computational fluid dynamics (CFD) models for design and analysis. This reliance on CFD models necessitates performing accurate and reliable uncertainty quantification (UQ) of the CFD models, which can be very expensive for hypersonic flows. Additionally, UQ approaches are many-query problems requiring many runs with a wide range of input parameters. One way to enable computationally expensive models to be used in such many-query problems is to employ projection-based reduced-order models (ROMs) in lieu of the (high-fidelity) full-order model (FOM). In particular, the least-squares Petrov–Galerkin (LSPG) ROM (equipped with hyper-reduction) has demonstrated the ability to significantly reduce simulation costs while retaining high levels of accuracy on a range of problems, including subsonic CFD applications. This allows LSPG ROM simulations to replace the FOM simulations in UQ studies, making UQ tractable even for large-scale CFD models. This work presents the first application of LSPG to a hypersonic CFD application, the Hypersonic International Flight Research Experimentation 1 (HIFiRE-1) in a three-dimensional, turbulent Mach 7.1 flow. This paper shows the ability of the ROM to significantly reduce computational costs while maintaining high levels of accuracy in computed quantities of interest.

Software Prefetching for Unstructured Mesh Applications

This article demonstrates the utility and implementation of software prefetching in an unstructured finite volume computational fluid dynamics code of representative size and complexity to an industrial application and across a number of modern processors. We present the benefits of auto-tuning for finding the optimal prefetch distance values across different computational kernels and architectures and demonstrate the importance of choosing the right prefetch destination across the available cache levels for best performance. We discuss the impact of the data layout on the number of prefetch instructions required in kernels with indirect addressing patterns and show how to best implement them in an existing large-scale computational fluid dynamics application. Through this, we show significant full application speed-ups on a range of processors and realistic test cases in both single core/tile and full socket configurations, such as 1.14× on the Intel Xeon Sandy Bridge, 1.09× on the Intel Xeon Broadwell, 1.29× on the Intel Xeon Skylake, 1.99× on the in-order Intel Xeon Phi Knights Corner coprocessor, and 1.51× on the out-of-order Intel Xeon Phi Knights Landing many-core processor.

Concentration profiling of a horizontal sedimentation tank utilising a bespoke acoustic backscatter array and CFD simulations

The performance of a pilot-scale horizontal sedimentation tank was characterised utilising computational fluid dynamics (CFD) and a bespoke ultrasonic backscatter array, for both spherical glass and flocculated calcite separation. The CFD simulation was developed in OpenFOAM, using algebraic slip and hindered settling models, in order to solve the transport of multiple particle size classes, enabled through a population balance approach. Simulations of concentration compared closely to samples for the glass dispersions, but under-predicted concentration with flocculated calcite (likely due to complexities from modelling floc break-up in the mixer). In comparison, the acoustic array measured the calcite concentration with a high degree of resolution, where in particular, evidence suspension mobilisation near the outlet was observed, due to recirculation. Overall, we demonstrate the performance and current limitations of large-scale CFD for complex floc systems, as well as the use of ultrasonics to significantly aid process understanding, through online monitoring of solid-liquid separations.

Multiscale Parallelized Computational Fluid Dynamics Modeling Toward Resolving Manufacturable Roughness

Abstract Typical turbomachinery aerothermal problems of practical interest are characterized by flow structures of wide-ranging scales, which interact with each other. Such multiscale interactions can be observed between the flow structures produced by surface roughness and by the bulk flow patterns. Moreover, additive manufacturing (AM) may sooner or later open a new chapter in the way components are designed by granting designers the ability to control the shape and patterns of surface roughness. As a result, surface finish, which so far has been treated largely as a stochastic trait, can be shifted to a set of design parameters that consist of repetitive, discrete micro-elements on a wall surface (“manufacturable roughness”). Considering this prospective capability, the question would arise regarding how surface microstructures can be incorporated in computational analyses during designing in the future. Semi-empirical methods for predicting aerothermal characteristics and the impact of manufacturable roughness could be used to minimize computational cost. However, the lack of element-to-element resolution may lead to erroneous predictions, as the interactions among the roughness micro-elements have been shown to be significant for adequate performance predictions (Kapsis and He, 2018, “Analysis of Aerothermal Characteristics of Surface Micro-Structures,” ASME J. Fluids Eng., 140(5), p. 051104). In this paper, a new multiscale approach based on the novel block spectral method (BSM) is adopted. This method aims to provide efficient resolution of the detailed local flow variation in space and time of the large-scale microstructures. This resolution is provided without resorting to modeling every single ones in detail, as a conventional large-scale computational fluid dynamics (CFD) simulation would demand, but still demonstrating similar time-accurate and time-averaged flow properties. The main emphasis of this work is to develop a parallelized solver of the method to enable tackling large problems. The work also includes a first of the kind verification and demonstration of the method for wall surfaces with a large number of microstructured elements.