Comparison of Rotor-Airframe Interaction Utilizing Overset and Unstructured Grid Techniques

ACCURATE simulation of the viscous flow around a helicopter rotor is still a challenging problem. To predict the blade loading accurately, the numerical method must have a capability to capture the vortical wake generated from the rotor blade tip precisely, which significantly affects the aerodynamic performance of the rotor. Therefore, both the numerical scheme with low numerical dissipation and sufficient dense mesh far away from the blade surface are required for rotor computations to simulate the tip vortex with minimal distortion. Furthermore, for a hovering case, a large numerical integration time is needed for the capture of the fully developed vortical wake [1]. In addition, the existence of incompressible flow near the root will also make the convergence slow. Compared with fixed-wing computations, hovering rotor computations not only require more accurate numerical scheme but also will take much more run time to converge. During the past two decades many numerical methods [2–7] have been developed to simulate the viscous flow around a hovering rotor. The methods [2,3] with the single-block structured grid suffer from the grid-quality problem, because the mesh must be stretched and skewed to cover the complete computational domain and to match the periodic boundary condition. To avoid this problem, the chimera technique [4–7] suggested by Benek et al. [8] was introduced into hovering rotor computations. To capture the tip vortex without unphysical distortion and predict the blade loading accurately, some high-order schemes and vortex-adapted chimera methods [6,7] were presented. Although significant advances in the accurate simulation of the tip vortex have been made with these advanced methods, the large computation cost and the low robustness, as well as the complexity make them far away from engineering applications. As mentioned above, hovering rotor computations, especially the computations with high-order methods, require more run time to converge than their fixed-wing counterparts. A few attempts [1,9,10] have been made to use multigrid methods to accelerate the convergence of hovering rotor calculations. Up to now, there is few works concerning the multigrid method for the viscous flow computation about a hovering rotor based on overset grids. The main objective of the present study is to develop an efficient multigrid algorithm coupled with chimera grids for Navier–Stokes computations about a hovering rotor. It should be mentioned that the experience of this work can be extended to accelerate the computation about a rotor in forward flight using moving overset grids,which is themainmotivation of this research. Furthermore, this experience can also be the basis of the future multigrid computation using high-order methods and chimera grids for rotor flows.

In support of NASA’s next generation heavy lift launch vehicle (HLLV), a simulation protocol has been developed to generate databases of the aerodynamic force and moment coefficients for HLLV ascent. The simulation protocol has been established and validated with a series of computational analyses that ensure best practices are achieved. Results of the sensitivity analyses using a full-scale Ares V flight vehicle are next applied in a validation study with three scaled-down Ares V wind tunnel test articles. Three independent computational fluid dynamic (CFD) flow solvers were included in the study. These included OVERFLOW, a viscous Reynolds Averaged Navier-Stokes (RANS) solver for structured overset grids, USM3D, a viscous RANS solver for unstructured tetrahedral grids, and Cart3D, an inviscid Euler solver using unstructured Cartesian grids and adjoint-based adaptive mesh refinement. First, a series of tests was independently performed for each applicable CFD code, including a grid convergence study and sensitivity studies of turbulence models and convective flux discretization methods. Once the proper grid resolution, physical models, and numerical parameters were determined for each of the codes, the process was continued with a code-to-code comparison. Each CFD code was applied to the Ares V flight vehicle at several points in the ascent trajectory, with all three codes obtaining consistent force and moment predictions. Finally, an extensive validation of the CFD approach was performed, in which the three codes were used to generate aero-databases of force and moment coefficients for three distinct Ares V wind tunnel test articles. These computations were performed concurrent to the experimental databases generated in the 14-inch wind tunnel at Marshall Space Flight Center (MSFC). Comparisons of the CFD results with the experimental data are reported and the viscous flow results compare well.

[Abstract] A method of unstructured dynamic overset grids is developed for the simulation of rotor forward flight flow and unsteady rotor-fuselage interaction aerodynamics. Two overset grids are formed by dividing the initial integral grid into two parts, namely, rotational grid and static grid, and the rotational grid rotates with the rotor while the static grid keep static during the calculating process. Also the hole fringes are formed and donor elements for hole fringe points are founded at the same time when the overset grids are formed. An appropriate overlapped area can be ensured during the unsteady calculation process, and the hole fringe needn’t to be formed repeatedly in this method. A fast searching method is used to update the donor elements, which can reduce the computational effort. The linear spring analogy is used in the rotational grid to calculate the flapping and pitching movement. The unsteady Euler equations are solved using duel time stepping to simulate a two-bladed rotor forward flight flow and Georgia Tech configuration rotor-fuselage interaction flowfield and the calculation results agree well with Experimental data.

Grid adaptation using general elements was applied to a hypersonic high angle of attack re-entry vehicle to determine if there could be improvements made in the prediction of heat transfer starting from a baseline unstructured grid. Grid adaptation was able to greatly reduce the errors in heat transfer prediction that are common when computing on unstructured grids. The reduction in error is due to a shrinking of average cell size near the bow shock. The smaller cells cause a less drastic jump in the solution of the Riemann problem at the shock. Additionally, a hybrid numerical scheme was used to add numerical dissipation near the bow shock region to help further reduce the jump in the solution of the Riemann problem at the shock. Average errors in heat transfer near the stagnation region were reduced by approximately 50% due to the smaller average cell size and hybrid numerical scheme. Overall the approach of combining grid adaptation and a hybrid numerical scheme appears to provide a robust and reliable method of computing heat transfer rates for hypersonic re-entry vehicles using unstructured grids.

An overset grid thin-layer Navier-Stokes code has been extended to include dynamic motion of helicopter rotor blades through relative grid motion. The unsteady flowfield and airloads on an AH-IG rotor in forward flight were computed to verify the methodology and to demonstrate the method's potential usefulness towards comprehensive helicopter codes. In addition, the method uses the blade's first harmonics measured in the flight test to prescribe the blade motion. The solution was impulsively started and became periodic in less than three rotor revolutions. Detailed unsteady numerical flow visualization techniques were applied to the entire unsteady data set of five rotor revolutions and exhibited flowfield features such as blade vortex interaction and wake roll-up. The unsteady blade loads and surface pressures compare well against those from flight measurements. Details of the method, a discussion of the resulting predicted flowfield, and requirements for future work are presented. Overall, given the proper blade dynamics, this method can compute the unsteady flowfield of a general helicopter rotor in forward flight.

To accurately predict the ice shape on the fuselage under rotor-wake effects, discrete modeling of individual tip vortices is essential to determine the local aerodynamic interactions between the vortex and the fuselage. To this end, the actuator surface model is developed and applied to the present study. Then, water-droplet-trajectory prediction, thermal analysis, and ice growth module are seamlessly integrated. This study aims to perform detailed analysis of the rotor downwash effects on ice that accretes on a helicopter fuselage and to determine the variation with respect to the forward flight speed. As a result of comprehensive numerical computations, the following conclusions are reached. First, the rotor inflow transforms the droplet trajectories, the location of ice accretion and the amount of ice are changed. Actually, 16.5% less ice is accumulated on the overall fuselage in the rotor–fuselage interaction case compared to the isolated fuselage case. Second, a clear-cut distinction of the icing locations on the fuselage is observed with various advance ratios. In hovering conditions, massive ice is accumulated on the tail-boom and fuselage nose region, where the tip vortices collide with fuselage. Due to the large collision area with high speed of droplets, 15.2 and 9.1% more ice is accumulated in hovering case than that of forward flight case with the advance ratio 0.075 and 0.15, respectively. However, as the forward speed grows from 0.15 to 0.2, the total mass of ice exponentially increases. In case of the advance ratio 0.2, 36% more ice is accumulated than that of the hovering case.

The determination of the air loads acting on rotor blades in forward flight presents an interesting and challenging problem in applied aerodynamics. Of particular importance for design purposes are the oscillatory components of this loading occurring at harmonics of the rotor speed. Unlike a wing, the trailing and shed vortex system of the blade generates a spiral wake that returns close to the blade. Because of its close proximity to the blade, the wake cannot be considered as rigid. Also, since the resulting loads are highly time-dependent, unsteady aerodynamic effects become important. An analytical treatment for the limiting case of hovering flight results in a simple closed-form solution that demonstrates that the oscillatory lift under normal operating conditions can be less than 50% of its quasi-static value. It is shown that the air loads in forward flight depend primarily on the equivalent lift deficiency and the down wash generated by the steady-state blade vortex system. It is also shown that, at low speeds, the rotor wake can be drawn up into the rotor leading edge and that this is probably the primary cause of transition roughness and noise encountered under certain flight conditions.

: An international computational aerodynamics study under the auspices of The Technical Cooperation Program (TTCP) Weapons Technology Panel 2 (APN-TP-2) involving participants from defense research laboratories of the United States, United Kingdom, Canada, and Australia was recently completed. The purpose of this study was to examine computational predictive technologies for finned missile shapes by comparing Navier-Stokes predictions to experimental data. Experimental data consisting of surface pressures on the body and fins, flow field pitot pressures, and force measurements were available for comparison to the computational results. The computational results for this study established an extensive database for evaluation and comparison. The fall database consists of results from six Navier-Stokes codes obtained by seven multi-block patched and unstructured grids for five distinct test cases. The statistical analysis techniques developed to help provide an evaluation of the predictive techniques are described. Quantitative results of the analysis of the differences between computational and experimental results are presented graphically and quantitatively in terms of medians, standard deviation, and a figure of merit to assist in the overall evaluation of the study results. The good performance achieved using the Spalart-Allmaras turbulence model and multi-block patched and unstructured grid techniques are noted in the findings.

Aeroelastic Reynolds-averaged Navier‐Stokes and Euler computations are presented for an articulated model rotor in hover and forward flight. Comparative rigid-blade simulations are carried out to assess the effects of blade dynamics and elasticity on the numerical results. The INROT flow solver operates on deformable structured overset grids and can be tightly coupled with a finite element model of the rotor blade structure (DYNROT) based on Timoshenko beam theory. The order of time accuracy of fluid and structure modules is maintained in the overall analysis by an appropriate staggered coupling scheme. At the investigated thrust setting, global hover performance values computed by the coupled fully turbulent Navier‐Stokes analysis agree fairly well with available experimental data. In forward flight, the aeroelastic results are in much better agreement with the measurements than those obtained from rigid-blade simulations with prescribed articulation. Apart from superior rotor power predicition, the local pitching moment coefficients computed by the viscous analysis are found to correlate better with wind-tunnel data than the corresponding Euler output.

This allows unregistered users to read the abstract as a teaser for the complete chapter. The formulas for calculating the variation of aerodynamic coefficient of iced blade airfoil were introduced according to rotor icing test data from NASA and the formulas for aerodynamic force and moment of rotor icing were deduced. Further, the model of flight dynamics for helicopter icing was built. Icing effect on trim characteristics and stability of certain single-rotor helicopter in hover and forward speed was studied. The results show that whether in hover or forward flight, when the helicopter iced, the amount of the collective control, the longitudinal control, and the lateral control of the rotor, the tail rotor collective control as well as the roll angle increases, while the pitch angle decreases; the longitudinal long-period and short-period mode goes more stable, and the same as the lateral helical mode, while the roll mode and Dutch roll mode go more unstable.

In this paper, the unsteady aerodynamics of a ducted fan micro air vehicle is investigated using an unstructured overset grid technique and momentum source method. The in-house programmed compressible Navier–Stoke solver is preconditioned for low Mach number flow regime, and a dual time-stepping strategy is employed to guarantee the computing accuracy and efficiency. Momentum source items are added in the Navier–Stoke solver to replace the contra-rotating propellers in numerical simulation which simplify the inherently unsteady flow into a quasi-steady one. The developed method was verified and validated as a reliable tool for predicting the unsteady aerodynamic performance in low Reynolds flow regime. The effects of reduced frequency, flight velocity and propeller speed on the aerodynamic performance of the ducted fan micro air vehicle are evaluated in this paper. Results show that the hysteresis effect of aerodynamic coefficient increases as induced frequency, freestream velocity and propeller speed increases.

A multidisciplinary gradient-based sensitivity-analysis methodology is evaluated for optimization of rotorcraft configurations. The tightly coupled discipline models include physics-based fluid dynamics and rotorcraft comprehensive analysis. A discretely consistent adjoint method accounts for sensitivities of the unsteady flow and unstructured, dynamic, overset grids; whereas sensitivities of structural responses to aerodynamic loads are computed using a complex-variable method. The methodology is applied to optimize the shape of UH-60A Blackhawk helicopter blades for hover and forward-flight conditions. The objective of the multipoint design is to simultaneously increase the rotorcraft figure of merit in hover flight and reduce the rotor power in forward flight. Trimmed loose-coupling solutions for the baseline configuration are used to initiate the tight-coupling multidisciplinary analysis. The target thrust, rolling moment, and pitching moment are enforced as optimization constraints. The optimized configuration improves the optimization metrics at both design points. The improved performance and all constraints are maintained over many revolutions beyond the optimization interval, satisfying the required flight conditions. The computational cost of the optimization cycle is assessed in a high-performance computing environment and found affordable for design of rotorcraft in general level-flight conditions.

The design of new helicopter rotor blades is a challenging task. The individual blade sections undergo very different flow conditions during the various flight regimes of the helicopter. In forward flight, the advancing side operates in a transonic regime where potentially shock waves can occur, while on the retreating side little flow velocities at high angle of attacks are seen up to reverse flow. In hover, the oncoming tip vortex of the previous blade drastically influences the inflow on the rotor. This paper joins the classical blade shape design with numerical optimization techniques. Opposing to the direct solution process of directly modifying the blade shape, the rotor blade is conventionally considered as a set of airfoils, a planform and a twist distribution. First, rotor airfoils are found through numerical optimization and then placed on the reference rotor. The planform and twist of this rotor are then also numerical optimized. The obtained trade-off blade for hover and forward flight drastically improved the performance over the reference HART-II blade (11% in forward flight, 5% in hover flight). In order to arrive at industrial relevant blades, further work including more disciplines becomes necessary.

This work presents a framework for the optimisation of various aspects of rotor blades in forward flight. The proposed method employs CFD in conjunction with artificial neural networks (ANNs) as metamodels, and genetic algorithms (GAs) as optimisation methods. The approach is demonstrated using a relatively well-known case, which is the optimisation of linear twist of rotors in hover, other cases include transonic aerofoils and wing planform. The method was then used to optimise the anhedral and sweep of the UH60-A rotor blade in forward flight. For each case, a parameterisation method is defined, a specific objective function created using the initial CFD data and the ANNs was subsequently used to evaluate this objective function for optimisation using the GAs. The obtained results suggest optima in agreement with engineering intuition but provide precise information about the shape of the final lifting surface and its performance. The results were checked using different optimisation methods and metamodels and were not sensitive to the employed techniques with substantial overlap between the outputs of the selected methods. The main CPU cost was associated with the population of the CFD database necessary for the metamodel.

Analysis of non-conservative interpolation techniques in overset grid finite-volume methods