A simulation approach was established for the application of lattice -Boltzmann method (LBM) CFD to low pressure turbine (LPT) aerodynamics for 2.5D linear cascades of LPT airfoils at low Reynolds numbers. This methodology was evaluated for six LPT airfoil geometries developed at AFRL (four high -lift, two high-lift/high-work) over realistic Reynolds lapse ranges. Predicted loss, loading, and suction surface boundary layer state were compared against prior linear cascade wind tunnel measurements acquired in the AFRL Low Speed Wind Tunnel Facility. The chief computational strategy for the LBM simulations was to achieve direct numerical simulation (DNS) level fidelity across the Reynolds range using LBM-LES, as implemented in the PowerFLOW (R) commercial solver developed by Dassault Systemes SIMULIA Corporation. This DNS-level fidelity was achieved by maintaining isotropically-fine near-wall grid spacing ((x+) over bar, (y+) over bar, (z+) over bar <1) across the Reynolds number range, and was demonstrated with a grid sensitivity study and the calculation of established grid resolution quality metrics for scale -resolving simulations. Recent enhancements to PowerFLOW (R)'s subgrid turbulence model were leveraged to achieve accurate predictions of laminar separation/reattachment. Particular attention was paid to capturing the sensitivity of LPT airfoil cascade performance to freestream turbulence, accomplished via the introduction of resolved turbulence at the inlet of the LBM computational domain. Despite the DNS-level fidelity that was achieved, overall computational cost was compatible with industrial LPT airfoil design cycles, owing to careful scaling of the 2.5D span and simulation physical time as a function of Reynolds number. Several challenges were identified, including the occurrence of Reynolds lapse hysteresis for certain airfoils and relatively higher computational cost at the highest Reynolds numbers studied.

Assessing the ice accretion on aerodynamic surfaces due to aircraft operating at adverse weather conditions remains a challenging task. However, the impact on aerodynamic performance can be significant which makes ice accretion assessment an important aspect in the design and certification process. Simulation techniques can be employed to complement wind tunnel and flight tests, which are typically used to meet regulation requirements. This paper presents a novel, particle model based technique embedded in the Lattice Boltzmann Method (LBM) based flow solver PowerFLOW® to predict rime ice shapes. Validation results are presented for the NACA 0012 airfoil and for an inboard wing section of the NASA Common Research Model (CRM). Since realistic rime ice shapes can be obtained, scale-resolving simulations based on Very Large Eddy Simulations (VLES) can then be performed on the iced airfoil to assess the impact on aerodynamic performance and to analyze the turbulent flow field created by the ice topology.

Unsteady, scale-resolving aerodynamic simulations are presented that were conducted for the Fourth AIAA High-Lift Prediction Workshop on the high-lift version of the NASA Common Research Model using a Lattice Boltzmann Method. The method has been previously shown to capture the relevant flow physics of complex high-lift cases. Grid convergence studies for different angles of attack show the convergence behavior at distinct flow regimes and discuss challenges for time-accurate and scale-resolving simulations. Results of a flap deflection study show good qualitative agreement to experiments and demonstrate the capability to predict design changes, superior to standard RANS simulation tools as they are used in industry today. Dedicated maximum lift studies using free-air and wind tunnel boundary conditions, respectively, confirm the sensitivity of the aircraft model to wind tunnel installation effects. Free-air simulation results show good correlation to experiments within the linear range of the lift curve, but capture a critical break in the pitching moment curve later than seen in the experiments. The trend of an earlier pitch break, triggered by the wind tunnel floor boundary layer, is accurately captured in the simulations including the wind tunnel environment. This supports the notion that a precise representation of the experimental setup is important to accurately predict the high-lift performance of a test article.

This paper presents a sequential multi-physics/multi-scale simulation methodology for a landing gear that is retracting under aerodynamic loads. A main landing gear representative of a single-aisle aircraft has been designed and a mechanical model of the underlying kinematics has been established. This allows accessing the position of the landing gear at any intermediate position between fully deployed and fully retracted. Computational Fluid Dynamics (CFD) simulations are then executed for multiple landing gear positions at take-off flight conditions. The resulting aerodynamic loads are subsequently used inside a multi-body simulation of the landing gear. The engineering scenario considered here is to assess the actuator force that is required to retract the landing gear within a prescribed period of time. This force can then be used further in the design process, e.g. to dimension the actuator of the landing gear or the hydraulic system of the aircraft, which exemplarily shows the engineering value of the proposed workflow.

The flow through a 2.5D axial compressor cascade blade-row is simulated using a compressible Lattice-Boltzmann Method. The simulations are carried out for a nominal flow condition that yields a Reynolds number of 300,000 based on the axial chord and an inlet Mach number of 0.66. The simulations incorporate incoming wakes generated by moving bars with an equivalent reduced frequency of 1.9. A URANS simulation is carried out first, which reveals deficiencies in predicting time-averaged isentropic surface Mach numbers and wall shear stresses when compared against published DNS results. In order to increase the accuracy of the simulation, an implicit LES is run on a wall-resolved mesh. These results match the reference data because transition effects are correctly accounted for. A new, transition enabled hybrid RANS-LES approach is then carried out which shows that similar accuracy as the wall resolved LES can be obtained.

The NASA 2030 vision study illustrates the need for multi-physics simulations during the design process. Fluid-Structure Interaction (FSI) for airframes is an important example of multi-physics, multi-disciplinary simulation. The NASA Common Research Model (CRM) mock-up demonstrates the importance of FSI with displacements of up to 15.5mm, which is equivalent to 8.25% of the mean aerodynamic chord. The change in wing twist at cruise conditions yields values of up to -1.1° at the wingtip, which influences the local loading of the wing and the overall aerodynamic performance. The CRM mock-up serves as a validation case for the coupling of a Lattice-Boltzmann Method (LBM) based fluid solver with an implicit, structural mechanics solver. Since the coupling requires an iterative process to achieve the final wing deformation, an automated workflow was created that takes care of the run management, surface data mapping and geometry deformation. The predicted wing deformation is compared to the wing deformation measured in the wind tunnel for multiple angles of attack. The change in behavior of wing bending and twisting is captured correctly when transitioning from the linear to the non-linear aerodynamic regime. The accuracy lies within 0.25° for twisting and 5mm for bending with the largest differences occurring at low angles of attack.

In 2014 NASA published the outcome of the 2030 CFD (Computational Fluid Dynamics) Vision study: “CFD Vision 2030: A path to Revolutionary Computational Aerosciences”. The study provided a comprehensive review of the state of the art of CFD in 2014 for aerospace applications including, but not limited to, numerical algorithms, physics models, MDAO (Multidisciplinary Design Analysis and Optimization) and HPC (High Performance Computing) hardware. The study also proposed four conceptual ideas of Grand Challenge problems that would build on and benefit from advances outlined in the roadmap. The proposed challenges were meant to foster more detailed descriptions of grand challenge problems for specific disciplines. One of the proposed challenges was in the gas turbine propulsion area, focusing on transient full engine simulations. The current paper addresses detailed technical aspects of that challenge, and proposes a plan to approach it in a gradual manner, which includes high fidelity modeling of components, component coupling, and targeted experimental campaigns relying on common research models.

This paper investigates the capability of the Lattice-Boltzmann Method (LBM) based solver SIMULIA PowerFLOW® to predict the flow separation at a wing-fuselage junction. Turbulence is resolved using a Very Large Eddy Simulation (VLES) turbulence modeling approach in the juncture region. Geometrical trips are applied to the wing and the fuselage to create dynamic boundary layers. The juncture flow separation is investigated at a negative angle of attack of -2.5° and results are compared to high fidelity wind tunnel measurements. A comparison of velocity profiles in the separation region with the previously published results at an angle of 5.0° are included as well.

Fluidic thrust vectoring is an attractive concept for enhancing the maneuverability of advanced fighter aircraft. The design space for thrust vectoring can be large and physical testing is often limited in the level of detail that can be provided to enable design decisions. Continued advances in CFD can provide opportunities to understand not just global performance but the reason certain parameters impact the design space. Recent advances in Lattice-Boltzmann method (LBM)-based Very Large Eddy Simulation (VLES) for compressible flows enable opportunity to demonstrate feasibility of evaluating design space in an accurate and efficient manner. In this study, fluidic thrust vectoring simulations have been conducted using LBM-VLES for a fixed nozzle pressure ratio and different slot injection mass flow rates. The Mach number in the flow exceeded 2.0 which presented a challenge for LBM-VLES methodology. The simulations did not encounter any numerical stability issues at such Mach numbers and in comparison with the reference experimental data, thrust vectoring angles were predicted to within +/- 2 degs for all injection mass flows.

Non-linear turbulence closures were developed that improve the prediction accuracy of wake mixing in low-pressure turbine (LPT) flows. First, Reynolds-averaged Navier–Stokes (RANS) calculations using five linear turbulence closures were performed for the T106A LPT profile at exit Mach number 0.4 and isentropic exit Reynolds numbers 60,000 and 100,000. None of these RANS models were able to accurately reproduce wake loss profiles, a crucial parameter in LPT design, from direct numerical simulation (DNS) reference data. However, the recently proposed kv2¯ω transition model was found to produce the best agreement with DNS data in terms of blade loading and boundary layer behavior and thus was selected as baseline model for turbulence closure development. Analysis of the DNS data revealed that the linear stress-strain coupling constitutes one of the main model form errors. Hence, a gene-expression programming (GEP) based machine-learning technique was applied to the high-fidelity DNS data to train non-linear explicit algebraic Reynolds stress models (EARSM). In particular, the GEP algorithm was tasked to minimize the weighted difference between the DNS and RANS anisotropy tensors, using different training regions. The trained models were first assessed in an a priori sense (without running any CFD) and showed much improved alignment of the trained models in the region of training. Additional RANS calculations were then performed using the trained models. Importantly, to assess their robustness, the trained models were tested both on the cases they were trained for and on testing, i.e. previously not seen, cases with different flow features. The developed models improved prediction of the Reynolds stress, TKE production, wake-loss profiles and wake maturity, across all cases, in particular those trained on just the wake region.

In the context of wall-resolved industrial large-eddy simulation, a comparison is made between a high-order flux reconstruction (FR)/correction procedure via reconstruction (CPR) solver (hpMusic) with p refinement and a commercial second-order finite volume solver (Fluent) with mesh refinement (h refinement). A well-known benchmark problem in turbomachinery is employed: transonic flow over a von Karman Institute high-pressure turbine vane at a Reynolds number of 1.16 x 10(6). All of the meshes originated from the same coarse mesh, a mixed unstructured mesh, generated through global uniform refinement for the purpose of evaluating the solution dependence on mesh and polynomial order. Because the meshes used for hpMusic and Fluent belong to the same family, useful information about solution accuracy and efficiency can be obtained. Detailed comparisons are made in mean surface loading, heat transfer, power spectral density of pressure at selected monitor points, mean boundary-layer velocity and total temperature profiles, and wake loss. Numerical results are compared with experimental data, when available. The high-order FR/CPR method is shown to achieve a higher accuracy at a reduced cost than the second-order finite volume method.

The design of low-pressure turbines (LPTs) must account for the losses generated by the unsteady interaction with the upstream blade row. The estimation of such unsteady wake-induced losses requires the accurate prediction of the incoming wake dynamics and decay. Existing linear turbulence closures (stress–strain relationships), however, do not offer an accurate prediction of the wake mixing. Therefore, machine-learnt, nonlinear turbulence closures (models) have been developed for LPT flows with unsteady inflow conditions using a zonal-based model development approach, with an aim to enhance the wake mixing prediction for unsteady Reynolds-averaged Navier–Stokes calculations. High-fidelity time-averaged and phase-lock averaged data at a realistic isentropic Reynolds number and two reduced frequencies, i.e., with discrete incoming wakes and with wake “fogging,” have been used as reference data for a machine learning algorithm based on gene expression programing to develop models. Models developed via phase-lock averaged data were able to capture the effect of certain prominent physical phenomena in LPTs such as wake–wake interactions, whereas models based on the time-averaged data could not. Correlations with the flow physics lead to a set of models that can effectively enhance the wake mixing prediction across the entire LPT domain for both cases. Based on a newly developed error metric, the developed models have reduced the a priori error over the Boussinesq approximation on average by 45%. This study thus aids blade designers in selecting the appropriate nonlinear closures capable of mimicking the physical mechanisms responsible for loss generation.

Fast response pressure data acquired in a high-speed 1.5-stage turbine hot gas ingestion rig (HGIR) show the existence of pressure oscillation modes in the rim-seal-wheelspace cavity of a high pressure gas turbine stage with purge flow. The experimental results and observations are complemented by computational assessments of pressure oscillation modes associated with the flow in canonical cavity configurations. The cavity modes identified include shallow cavity modes and Helmholtz resonance. The response of the cavity modes to variation in design and operating parameters are assessed. These parameters include cavity aspect ratio (AR), purge flow ratio, and flow direction defined by the ratio of primary tangential to axial velocity. Scaling the cavity modal response based on computational results and available experimental data in terms of the appropriate reduced frequencies appears to indicate the potential presence of a deep cavity mode as well. While the role of cavity modes on hot gas ingestion cannot be clarified based on the current set of data, the unsteady pressure field associated with turbine rim cavity modal response can be expected to drive ingress/egress.

Turbulence plays an important role in the flow physics through the high pressure turbine (HPT), yet its evolution through the stage is still poorly understood. Furthermore, achieving high values, on the order of 10–20%, can be both challenging and costly in an experimental facility. Numerous experimental efforts have been undertaken to mimic the turbulence levels exiting a combustor.

Flow exiting the combustor is highly turbulent and contains significant spatial gradients of pressure and temperature. The high pressure turbine nozzle vanes operating in this environment redistribute these spatial gradients and impact the inflow characteristics of the turbine rotor blades. The present study investigates the redistribution of total temperature through a turbine nozzle vane. Numerical investigation was performed using three-dimensional RANS analysis. Simulations were conducted using the Wilcox k–ω turbulence model and Shear Stress Transport (SST) with and without γ–Reθ transition model. Experimental measurements were obtained in an annular nozzle cascade facility. Two sets of inlet conditions were considered. The first was a nominally uniform total temperature. The second had a span-wise variation of total temperature. Both sets of inlet conditions had nominally the same inlet total pressure and inlet Mach number. Span-wise redistribution was evaluated using the circum-ferentially averaged total temperature profile at a plane downstream of the nozzle. Physical arguments about the influence of nozzle secondary flows on this redistribution are presented.

Machine learning was applied to large-eddy simulation (LES) data to develop nonlinear turbulence stress and heat flux closures with increased prediction accuracy for trailing-edge cooling slot cases. The LES data were generated for a thick and a thin trailing-edge slot and shown to agree well with experimental data, thus providing suitable training data for model development. A gene expression programming (GEP) based algorithm was used to symbolically regress novel nonlinear explicit algebraic stress models and heat-flux closures based on either the gradient diffusion or the generalized gradient diffusion approaches. Steady Reynolds-averaged Navier–Stokes (RANS) calculations were then conducted with the new explicit algebraic stress models. The best overall agreement with LES data was found when selecting the near wall region, where high levels of anisotropy exist, as training region, and using the mean squared error of the anisotropy tensor as cost function. For the thin lip geometry, the adiabatic wall effectiveness was predicted in good agreement with the LES and experimental data when combining the GEP-trained model with the standard eddy-diffusivity model. Crucially, the same model combination also produced significant improvement in the predictive accuracy of adiabatic wall effectiveness for different blowing ratios (BRs), despite not having seen those in the training process. For the thick lip case, the match with reference values deteriorated due to the presence of large-scale, relative to slot height, vortex shedding. A GEP-trained scalar flux model, in conjunction with a trained RANS model, was found to significantly improve the prediction of the adiabatic wall effectiveness.