Computational Fluid Dynamics Predictions Research Articles

Low-Order Response Surface Model for Simulating Wake Interference in Aerial Refueling

Aerial refueling is a critical element of Navy operations, made difficult by the inherent aerodynamic interference between the tanker and receiver aircraft. Current flight simulation capabilities are unable to accurately reproduce the complex wake interference effects on the receiver aircraft, solely accounting for the inviscid vortex. This research develops an aerodynamic interference module that may be incorporated into the simulation environment to account for incremental changes in the forces and moments on the receiver aircraft as it approaches the tanker during typical refueling scenarios. The interference module is based on low-order response surface models derived from a database of high-fidelity computational fluid dynamics predictions. The low-order model encompasses all of the complex wake effects on the receiver aerodynamic coefficients and may be quickly interrogated for use in real-time flight simulation. Predicted aerodynamic coefficient increments are validated against flight data for an E-2D receiver trailing behind a KC-135 tanker.

The role of turbulence models in the modeling of membrane bioreactors

In membrane bioreactors (MBRs), aeration mitigates fouling by creating turbulence and increasing shear stress on the membrane surface. To study the hydrodynamics of MBRs, several researchers have been using computational fluid dynamics (CFD). However, the literature lacks papers on the sensitivity of the CFD predictions for MBRs to the modeling choices related to the different phenomena involved in MBRs, including turbulence. This work evaluates the bubbly flow in an immersed MBR with the most common turbulence models (k-ε with two different wall functions and SST k-ω) at two aeration rates (5 and 15 m3 h−1). Predictions of velocities near the membranes and of local and spatially averaged shear stresses on the membrane modules were compared between the models and to literature data. The predictions were highly sensitive to the turbulence modeling approach, and their sensitivity was position dependent. The differences between the models reached 21.6% and 86.1% considering spatially averaged and local values of shear stress, respectively. The SST k-ω model exhibited the best prediction at the low aeration rate, indicating the relevance of the wall region for this condition. These findings show that more attention must be given to the choice of turbulence model in MBR simulations.

CFD prediction of wave-induced forces on ships running in irregular stern quartering seas

Surf-riding/broaching is a dangerous event for high-speed ships running in stern quartering seas. Theoretical methods for the prediction of wave-induced forces on ships in surf-riding and broaching situations are still of the qualitative level. Therefore, an unsteady Reynolds-averaged Navier–Stokes computational fluid dynamics (CFD) solver with overset grids was applied to predict the wave-induced forces on ships in irregular stern quartering seas. To validate the CFD method, a dedicated model experiment was conducted for slender ships, and the wave-induced forces were measured to be for a direct comparison with CFD predictions using the same irregular waves encountered. The comparison results indicate good agreement between the CFD results and the experimental fluid dynamics (EFD) results except for the wave-induced roll moment.

Transient prediction capabilities of a novel physics-based ignition delay model in multi-pulsed direct injection diesel engines

This work is an extension of a novel physics-based ignition delay modeling methodology previously developed by the authors to predict physical and chemical ignition delays of multiple injections during steady operations in diesel engines. The modeling methodology is refined in this work to consider the influence of additional operating parameters such as volumetric efficiency, exhaust temperature and pressure on the ignition delay of multiple injections. Computational fluid dynamics predictions on two different engines indicated that the main spray encounters local temperatures about 60 K above average temperatures for about 1 mg of pilot. Hence, the modeling methodology was further refined to include this effect by considering the air mass trapped in pilot spray, computed based on the spray penetration and cone angle and tuned using results of the computational fluid dynamics studies. Comparisons of the ignition delay predictions with the stock boost temperature sensor and a specially incorporated, transient-capable fine wire thermocouple indicated that the measurements with stock sensor could be satisfactorily used for transients. Cycle-by-cycle changes in ignition delay could be predicted accurately when transients were imposed in boost pressure, rail pressure and main injection quantity in a turbocharged intercooled diesel engine controlled with an open engine control unit. Further validations were done even under a transient cycle when the engine was controlled by its stock engine control unit. The same tuning constants could be used for the prediction of the ignition delay under transients on another naturally aspirated engine. This indicates the suitability of the model for application in different engines. Finally, the model was incorporated within an open engine controller, and cycle-by-cycle prediction of ignition delays of the pilot and main injections were done in real time. It was possible to compute the ignition delays in less than 2 ms within engine control unit using the already available sensor inputs within an error band of ±60 µs.

Airfoil parameterization evaluation based on a modified PARASEC method for a H-Darrious rotor

The airfoil research is one of the most essential issue which has great influence on the vertical axis wind turbine (VAWT) performance. Thus, a comprehensive research was conducted in this study with the objective of evaluating the effect of the twelve airfoils' controlling parameters on a Darrieus rotor's performance. The airfoil parameterization method, orthogonal algorithm and computational fluid dynamic (CFD) prediction were intergtated to fulfill this objective. A modification was conducted for the airfoil parameterization method to avoid trailing edge crossing and make controlling parameter's geometrical meaning more explicitly. The analysis results sorted the effect of design parameters by impact weight or contribution ratio. A remarkable finding was that the design parameters on the lower surface have greater influence than those on the upper surface. Orthogonal analysis gained a best airfoil whose power coefficient (CP) is 13.26% larger than that of NACA 0015. Flow field analysis found that a tiny positive vortex was enlarged intensively with the increasing of the azimuth angle, which degrades the rotor's performance.

Prediction of the extent of thrombus formation in the parent artery after endovascular occlusion of a distal anterior cerebral artery aneurysm using computational fluid dynamics

Endovascular coiling of a cerebral aneurysm and coil occlusion of the parent artery have been occasionally performed to treat cerebral aneurysms; however, it is difficult to predict the accurate extent of thrombus formation in the parent artery proximal to the coiled aneurysm and the coil-occluded parent artery preoperatively, and unexpected occlusion of the arterial branches can occur by thrombus extension into or in the parent artery. The authors describe a case of a distal anterior cerebral artery (ACA) aneurysm treated by endovascular parent artery occlusion (PAO) with preoperative computational fluid dynamics (CFD) prediction of the extent of thrombus formation. A 73-year-old woman presented with subarachnoid hemorrhage and an aneurysm that was located on the right pericallosal artery distal to the paracentral artery bifurcation. Endovascular coiling of the aneurysm and the pericallosal artery was planned. In advance of the treatment, CFD was performed to predict the extent of thrombus formation with specific wall shear stress and shear rate thresholds. The hemodynamic results indicated that coiling of the aneurysm resulted in thrombus formation in the pericallosal artery up to just distal to the paracentral artery ostium; therefore, the treatment was implemented according to the CFD prediction. Postoperative digital subtraction angiography revealed that the extent of thrombus formation was consistent with the preoperative CFD prediction. This technique may prevent unexpected occlusion of arterial branches.

Experimental and numerical investigations of flow-induced vibration of tube arrays subjected to cross flow

Flow-induced vibration of tube array structures is a major concern in nuclear engineering. In this paper, the dynamical behavior of flexible tubes in cross flow is studied. A computational fluid dynamics (CFD) model is developed to investigate the interaction between flow perturbations and tube oscillations. The Large Eddy Simulation (LES) is validated and adopted in the simulations. The motion of tubes in different tube configurations (i.e., single tube, three in line tubes, normal square tube array, and normal triangle tube array with different pitch to diameter ratios) under different flow conditions are analyzed and compared by time trace, power spectral density (PSD), phase plane plot and Poincaré map. Experiments have been conducted using a flexible tube in a closed-loop piping system to compare with the CFD model predictions. It is found that tube undergoes vortex-induced vibration and then fluid elastic instability, limit-cycle oscillation occurs when the inlet flow velocity exceeds a critical point. The central tube in normal square tube array is the most stable among the three configurations. The critical pitch to diameter ratio of tube array under reduced velocity Ur = 6 is about 5.5. The experimental observations are in good agreement with the CFD predictions.

Turbulence Modeling Effects on the CFD Predictions of Flow over a Detailed Full-Scale Sedan Vehicle

In today’s road vehicle design processes, Computational Fluid Dynamics (CFD) has emerged as one of the major investigative tools for aerodynamics analyses. The age-old CFD methodology based on the Reynolds Averaged Navier–Stokes (RANS) approach is still considered as the most popular turbulence modeling approach in automotive industries due to its acceptable accuracy and affordable computational cost for predicting flows involving complex geometries. This popular use of RANS still persists in spite of the well-known fact that, for automotive flows, RANS turbulence models often fail to characterize the associated flow-field properly. It is even true that more often, the RANS approach fails to predict correct integral aerodynamic quantities like lift, drag, or moment coefficients, and as such, they are used to assess the relative magnitude and direction of a trend. Moreover, even for such purposes, notable disagreements generally exist between results predicted by different RANS models. Thanks to fast advances in computer technology, increasing popularity has been seen in the use of the hybrid Detached Eddy Simulation (DES), which blends the RANS approach with Large Eddy Simulation (LES). The DES methodology demonstrated a high potential of being more accurate and informative than the RANS approaches. Whilst evaluations of RANS and DES models on various applications are abundant in the literature, such evaluations on full-car models are relatively fewer. In this study, four RANS models that are widely used in engineering applications, i.e., the realizable k − ε two-layer, Abe–Kondoh–Nagano (AKN) k − ε low-Reynolds, SST k − ω , and V2F are evaluated on a full-scale passenger vehicle with two different front-end configurations. In addition, both cases are run with two DES models to assess the differences between the flow predictions obtained using RANS and DES.

Pump Grooved Seals: A Computational Fluid Dynamics Approach to Improve Bulk-Flow Model Predictions

In multiple stage centrifugal pumps, balance pistons, often comprising a grooved annular seal, equilibrate the full pressure rise across the pump. Grooves in the stator break the evolution of fluid swirl and increase mechanical energy dissipation; hence, a grooved seal offers a lesser leakage and lower cross-coupled stiffness than a similar size uniform clearance seal. To date, bulk-flow modelbulk-flow models (BFMs) expediently predict leakage and rotor dynamic force coefficients of grooved seals; however, they lack accuracy for any other geometry besides rectangular. Note that scalloped and triangular (serrated) groove seals are not uncommon. In these cases, computational fluid dynamics (CFD) models seals of complex shape to produce leakage and force coefficients. Alas, CFD is not yet ready for routine engineer practice. Hence, an intermediate procedure presently takes an accurate two-dimensional (2D) CFD model of a smaller flow region, namely a single groove and adjacent land, to produce stator and rotor surface wall friction factors, expressed as functions of the Reynolds numbers, for integration into an existing BFM and ready prediction of seal leakage and force coefficients. The selected groove-land section is well within the seal length and far away from the effects of the inlet condition. The analysis takes three water lubricated seals with distinct groove shapes: rectangular, scalloped, and triangular. Each seal, with length/diameter L/D = 0.4, has 44 grooves of shallow depth dg ∼ clearance Cr and operates at a rotor speed equal to 5,588 rpm (78 m/s surface speed) and with a pressure drop of 14.9 MPa. The method validity is asserted when 2D (single groove-land) and three-dimensional (3D) (whole seal) predictions for pressure and velocity fields are compared against each other. The CFD predictions, 2D and 3D, show that the triangular groove seal has the largest leakage, 41% greater than the rectangular groove seal does, albeit producing the smallest cross-coupled stiffnesses and whirl frequency ratio (WFR). On the other hand, the triangular groove seal has the largest direct stiffness and damping coefficients. The scalloped groove seal shows similar rotordynamic force coefficients as the rectangular groove seal but leaks 13% more. For the three seal groove types, the modified BFM predicts leakage that is less than 6% away from that delivered by CFD, whereas the seal stiffnesses (both direct and cross-coupled) differ by 13%, the direct damping coefficients by 18%, and the added mass coefficients are within 30%. The procedure introduced extends the applicability of a BFM to predict the dynamic performance of grooved seals with distinctive shapes.

In Vivo Intrathecal Tracer Dispersion in Cynomolgus Monkey Validates Wide Biodistribution Along Neuraxis.

It is commonly believed that in intrathecal (IT) drug delivery, agent distribution is confined to a narrow region close to the injection site, thereby undermining the efficacy of the method. To test the claim, multimodal in vivo imaging was used to experimentally observe the effects of IT infusion in cynomolgus monkey, looking at cerebrospinal fluid flow, anatomy, and dispersion of a radiolabeled tracer. At high infusion rates, the tracer reached the cervical region after only 2 h, demonstrating rapid and wide distribution. The same in vivo nonhuman primate imaging data also provided evidence in support of a computational fluid dynamic model for the prediction of drug distribution following IT injection. Tracer dispersion was predicted in two specimens matching the distribution acquired with positron emission tomography (PET). For the third specimen, tracer dispersion simulations were conducted as a blind study: predictions were made before in vivo biodistribution data was known. In all cases, the computational fluid dynamics (CFD) predictions of drug dispersion after IT administration showed close spatio-temporal agreement with tracer biodistribution in vivo. Validation by in vivo nonhuman primate data confirms our ability to predict the biodistribution of intrathecally administered agents in subject-specific models of the central nervous system from first principles. The experiments reinstate IT delivery as a viable administration method when targeting molecules to the whole spine or the brain. The proposed computational methodology enables rational design of novel therapies for neurological diseases that require reliable, efficient, and safe delivery of therapeutic agents to specific target sites in the central nervous system.

Numerical simulation studies of the effect of atmospheric stratification on the dispersion of LNG vapor released from the top of a storage tank

In this paper, Computational Fluid Dynamics (CFD) approach was used to study patterns of Liquefied Natural Gas (LNG) vapor dispersion under different atmospheric stabilities. First of all, the Burro 8 field test was used to validate prediction results generated by Fluidyn-PANACHE, a CFD modeling software. The test data confirmed that PANACHE is able to describe the dense gas behavior of LNG vapor dispersion. The CFD predictions regarding lower flammable limit (LFL) distances and downwind maximum concentration were in approximate agreement with the test data. After this, the dispersion of LNG vapor, which results from its accidental release from the top of large-scale storage tank under different atmospheric stabilities, were simulated. The results showed that the influence of atmospheric stability on the vertical and lateral distributions of vapor cloud, the downwind near-ground gas concentrations and the LFL distance is significant. The vertical spread of cloud under unstable condition is slightly larger than that under neutral and stable conditions. The dilution of the LNG cloud is suppressed under stable condition due to weaker turbulent mixing. Also, stable atmospheric condition can increase the near-ground gas concentrations, the lateral distribution of the cloud and the LFL distance. The longitudinal distribution of the cloud near the ground under neutral condition is larger than that under unstable and stable condition.

CFD Simulations of Molten Salt Reactor Experiment Core

At present, no clear guidelines exist for modeling non-water-cooled small modular reactors (SMRs) despite the rising need for high-fidelity simulation tools to support regulators and the industry. Most SMR concepts currently under the Canadian prelicensing review adopted non-water-cooled–reactor technologies [molten salt reactor (MSR), gas-cooled reactor, and liquid metal–cooled reactor] that are new for Canada. There is a need for a modeling tool set that is broadly applicable for the assessment of advanced technologies used in SMRs. Computational fluid dynamics (CFD) can be used in performance evaluation and safety analysis of non-water-cooled SMRs for modeling three-dimensional (3-D) fluid flow and heat transfer in geometries of arbitrary complexity without resorting to geometry-specific empirical correlations. This study investigates the capabilities of existing models within a commercial CFD code to simulate the flow and heat transfer characteristics in a MSR configuration. The Oak Ridge National Laboratory (ORNL) Molten Salt Reactor Experiment (MSRE) configuration was simulated in this study using a stand-alone CFD approach, and CFD predictions were assessed with ORNL data. Intricate geometry details within the MSRE core were included in the computational model to study the associated geometric effects. The results obtained in this study showcased the ability of CFD to predict 3-D effects within the computational domain especially at the lower plenums. The predicted trends for the temperature rise in the fuel and moderator within the core were in good agreement with the ORNL data. The results presented in this paper constitute the first step in developing Canadian Nuclear Laboratories’ capability for CFD modeling of non-water SMRs.

Improving accuracy and efficiency of CFD predictions of propeller open water performance

This research proposes mesh and domain optimization strategies for a popular Computational Fluid Dynamics (CFD) technique to estimate the open water propulsive characteristics of fixed pitch propellers accurately and time-efficiently based on examining the effect of various mesh and computation domain parameters. It used a Reynolds-Averaged Navier-Stokes (RANS) solver to predict the propulsive performance of a fixed pitch propeller with varied meshing, simulation domain and setup parameters. The optimized mesh and domain size parameters were selected using Design of Experiments (DoE) methods enabling simulations in a limited memory and in a timely manner without compromising the accuracy of results. The predicted thrust and torque for the propeller were compared to the corresponding measurements for determining the prediction accuracy. The authors found that the optimized meshing and setup arrangements reduced the propeller opens simulation time by at least a factor of six as compared to the generally popular CFD parameter setup. In addition, the accuracy of propulsive characteristics was improved by up to 50% as compared to published simulation results. The methodologies presented in this paper can be similarly applied to other simulations such as calm water ship resistance, ship propulsion etc. to systematically derive the optimized meshing arrangement for simulations with minimal simulation time and maximum accuracy. This investigation was carried out using a commercial CFD package; however, the findings can be applied to any RANS solver.

CFD prediction of convective heat transfer and pressure drop of pigs in group using virtual wind tunnels: Influence of grid resolution and turbulence modelling

Ventilation for the animal house is important since it is highly linked with the convective heat removal from animals. In this context, the animals as obstacles affect the ventilation airflow and consequently the air distribution within Animal Occupied Zone (AOZ). To study the convective heat transfer (CHT) of animals and the pressure drop (PD) resulting from the animals, computational fluid dynamics (CFD) can be an efficient tool, particularly in comparison with experimental methods. However, simulation accuracy is influenced by grid treatment and the choice of turbulence models. To reveal the grid effects on CHT and PD, fifteen 3D meshes with different facial grids sizes (0.04 m, 0.02 m, and 0.008 m) and prismatic layer thicknesses (PLT) (0 m, 0.0015 m, 0.009 m, 0.02 m, 0.04 m) were tested based on staggered tube bundle models representing simplified bluff bodies using virtual wind tunnels. Then, four commonly used turbulence models, namely, standard k–ɛ, realisable k–ɛ, standard k–ω, and SST k–ω, together with two wall treatments were evaluated and compared with the empirical calculation. The evaluation of turbulence models and wall treatment was then carried out based on grouped pig models using virtual wind tunnels. The results showed that the PLT could be an important influence on the stability of the simulation with varied facial grid sizes. The k–ω models performed better than the k–ɛ models on the predictions of both the CHT and the PD based on the tube bundle model. Wall function provided acceptable results on CHT, however, significant flaws on the PD on both tube bundles and pig models.

Influence of Purge Flow Swirl at Exit to the High-Pressure Compressor on OGV/Pre-Diffuser and Combustion System Aerodynamics

As aero gas turbine designs strive for ever greater efficiencies, the trend is for engine overall pressure ratios to rise. Although this provides greater thermal efficiency, it means that cycle temperatures also increase. One potential solution to managing the increasing temperatures is to employ a cooled cooling air system. In such a system, a purge flow into the main gas path downstream of the compressor will be required to prevent hot gas being ingested into the rotor drive cone cavity. However, the main gas path in compressors is aerodynamically sensitive and it is important to understand, and mitigate, the impact such a flow may have on the compressor outlet guide vanes, pre-diffuser, and the downstream combustion system aerodynamics. Initial computational fluid dynamics (CFD) predictions demonstrated the potential of the purge flow to negatively affect the outlet guide vanes and alter the inlet conditions to the combustion system. The purge flow modified the incidence onto the outlet guide vane, at the hub, such that the secondary flows increased in magnitude. An experimental assessment carried out using an existing fully annular, isothermal test facility confirmed the CFD results and importantly demonstrated that the degradation in the combustor inlet flow resulted in an increased combustion system loss. At the proposed purge flow rate, equal to ∼1% of the mainstream flow, these effects were small with the system loss increasing by ∼4%. However, at higher purge flow rates (up to 3%), these effects became notable and the outlet guide vane and pre-diffuser flow degraded significantly with a resultant increase in the combustion system loss of ∼13%. To mitigate these effects, CFD was used to examine the effect of varying the purge flow swirl fraction in order to better align the flow at the hub of the outlet guide vane. With a swirl fraction of 0.65 (x rotor speed), the secondary flows were reduced below that of the datum case (with no purge flow). Experimental data showed good agreement with the predicted flow topology and performance trends but the measured data showed smaller absolute changes. Differences in system loss were measured with savings of around 10% at the turbine feed ports for a mass flow ratio of 1% and a swirl fraction of 0.65.

Optimization of Piecewise Conical Nozzles: Theory and Application

An optimization study based on computational fluid dynamics (CFD) in combination with Stratford's analytical separation criterion was developed for the design of piecewise conical contraction zones and nozzles. The risk of flow separation was formally covered by a newly introduced dimensionless separation number. The use of this separation number can be interpreted as an adaption of Stratford's separation criterion to piecewise conical nozzles. In the nozzle design optimization process, the risk of flow separation was reduced by minimizing the separation number. It was found that the flow-optimized piecewise conical nozzle did not correspond to a direct geometric approximation of an ideal polynomial profile. In fact, it was beneficial to reduce the flow deflection in the outlet region for a piecewise conical nozzle to increase the nozzle performance. In order to validate the novel design method, extensive tests for different nozzle designs were conducted by means of wind tunnel tests. The measured velocity profiles and wall pressure distributions agreed well with the CFD predictions.

CFD investigation of NOx reduction with a flue-gas internal recirculation burner in a mid-sized boiler

The demand for reduction of nitrogen oxide (NOx) emissions from industrial facilities continuously increases, and considerable efforts have been exerted to achieve this goal. In this work, we propose a novel flue-gas internal recirculation (FIR) burner emphasizing the function of FIR to accomplish single-digit NOx emissions from a mid-/large-sized combustion system. In the new design, a FIR passage is installed inside the conventional non-FIR burner to draw back the flue gas from the combustion chamber and release it into the chamber as a mixture of air and flue gas. The effectiveness of FIR burner is evaluated by employing extensive computational fluid dynamics (CFD) simulations with an enhanced reaction rate model. The existing eddy dissipation model for reaction rate, including the turbulence-chemical interaction, is improved by introducing a position-dependent scaling factor, which is validated by comparison with temperature profiles in experiments. CFD predictions show that a small amount of flue gas returned to the burner still significantly alters the flow structure and temperature distribution. Accordingly, NOx emission is dramatically diminished (82.83 and 9.7 ppm in the non-FIR and FIR systems, respectively) using the FIR burner. These observations confirm that the new FIR burner effectively accomplishes ultra-low NOx emissions in field-scale combustion systems. In addition, the fundamentals of NOx reduction by the FIR burner are thoroughly examined in the present study. The findings will provide essential knowledge in designing other ultra-low NOx burners.

Use of Computational Fluid Dynamics (CFD) Dispersion Parameters in the Development of a New DPI Actuated with Low Air Volumes.

To determine the predictive power of computational fluid dynamics (CFD)-based dispersion parameters in the development of a new inline DPI that is actuated with low volumes of air. Four new versions of a dose aerosolization and containment (DAC)-unit DPI were created with varying inlet and outlet orifice sizes and analyzed with results from five previous designs. A concurrent in vitro and CFD analysis was conducted to predict the emitted dose (ED; as a % of loaded dose) and aerosol mass median aerodynamic diameter (MMAD) produced by each device when actuated with 10ml air bursts. CFD simulations of device operation were used to predict flow field and particle-based dispersion parameters. Comparisons of experimental and CFD results indicated that multiple flow field and particle-based dispersion parameters could be used to predict ED (minimum RMS Error = 4.9%) and MMAD (minimum RMS Error = 0.04μm) to a high degree of accuracy. Based on experiments, the best overall device produced mean (standard deviation; SD) ED = 82.9(4.3)% and mean MMAD (SD) = 1.73(0.07)μm, which were in close agreement with the CFD predictions. A unique relationship was identified in the DAC-unit DPI in which reducing turbulence also reduced the MMAD.

Enhanced design methodology of a low power stall regulated wind turbine. BEMT and MRF-RANS combination and comparison with existing designs

Wind energy importance has increased over the past decades. Energy generation by small turbines installed near urban locations has experienced noticeable growth. This work is focused on the development of a design methodology for a low power blade well suited for all the wind operation conditions.First, a complete Design of Experiments will be presented using the low computational cost tool Blade Element Momentum Theory (BEMT) in order to discard those designs which are clearly not suited to the requirements of the system. Later, the remaining were analyzed using a Computational Fluid Dynamics (CFD) methodology in order to account for three dimensional effects. The value of the left slope of the non-dimensional power curve has been found to be a key parameter for the design.This methodology has been validated with experimental results available from NREL Phase VI wind turbine, allowing to conclude that BEMT is capable to provide with pre-design accurate results which, nevertheless, should corrected by CFD.The results of the proposed design are analyzed and compared to the CFD predictions of a commercial existing blade designed to comply with similar working. For the proposed design, predictions indicate better behavior in terms of maximum power and controllability.

CFD and experimental investigation into a non‐intrusive method for measuring cooling air mass flow rate through a synchronous generator

This study presents a detailed methodology for non-intrusive measurement of cooling air mass flow rate that enables an overall machine evaluation. This approach enables the simultaneous measurement of air mass flow with shaft torque at differing operating points while minimising the change in air flow introduced by the measurement system. The impact of geometric parameters in the designed system is investigated using a detailed 180° computational fluid dynamics (CFD) model. Special attention was paid to minimising their influence on pressure drop, the mass flow rate through the machine, and measurement uncertainty. Based on the results of this investigation, the system was designed and manufactured, and the experimentally measured data was used to validate the CFD predictions. For the as optimal identified configuration, the flow rate is predicted to decrease by 2.2% relative to unrestricted operation. The achieved measurement uncertainty is ± 2.6% at synchronous speed.