Optimization of a Minienvironment Design Using Computational Fluid Dynamics

Until recently, airfoil design has largely been a process of trial and error that has relied heavily upon the intuition and experience of the designer. Using an experimental database for airfoils, a designer would normally employ an educated guess to make modifications to the design. He would then build a new model and test it in the wind tunnel. After several of these design iterations, an airfoil with the desired characteristics would be obtained. Now, with the development of advanced Computational Fluid Dynamics (CFD) codes, much of the trial and error process in aerodynamic design can be eliminated. A CFD code models air flow around a given geometry by numerically solving the governing equations of fluid dynamics on the computer. Rather than testing a new model in the wind tunnel at each design iteration, it is much faster and cheaper to test new airfoil designs with CFD calculations. However each new design requires many (several hundred) computationally intensive CFD calculations before an acceptable design evolves using an optimization procedure. This process is called direct design optimization. In order to alleviate the problems associated with the computationally intensive nature of the direct design methods, attempts have been made to calculate the shape of an airfoil that would have a specified flow distribution over its surface. This is essentially the inverse of a direct CFD calculation in which the shape of the airfoil is given and the flow is calculated; hence it is referred to as inverse design. Summaries of previous work in inverse design are given by Jameson [1], Cabuk and Modi [2], Lewis, Peters and Agarwal, [3] and Agarwal and Lewis [4], and Lewis [5]. In this paper, the inverse design has been formulated as a variational problem in which a control system minimizes the differences between the actual and the desired velocity (or pressure distributions) over the airfoil surface by making successive changes to the shape of the airfoil. A flow chart of the design method is shown in Fig. 1. First, an initial airfoil and the desired velocity (or pressure distribution) are input. The Mach number and angle of attack are also specified. Then a flow solver (Full Potential or Euler) computes the actual velocity and pressure distribution around the airfoil. Next, a cost functional, which represents the deviation of the actual velocity (or pressure distribution) from the desired velocity (or pressure distribution) is computed. If the cost functional has reached a minimum, then the process terminates. Otherwise, a variation is made to the shape of the airfoil such that the cost functional is reduced. The process is repeated so that the cost functional is minimized thereby minimizing the deviation between the actual and the desired velocity (or pressure distributions). The inverse design enhances the efficiency of the aerodynamic design process by several orders of magnitude. Several test cases are shown to demonstrate the method and its robust characteristics (accuracy, stability and efficiency) for aerodynamic design.

A two-dimensional hypersonic scramjet inlet has been investigated in a combined experimental and analytical program aimed at addressing the fundamental issues related to the design of scramjet inlets. The experimental portion of the program was conducted in the Calspan 48-in. shock tunnel at Mach 10 and 13. The computational analysis was conducted using a two-dimensional parabolized Navier-Stokes computational fluid dynamics (CFD) code. This article addresses the issues concerned with the flow over the external forebody of the inlet which consists of a blunted wedge followed by an isentropic compression. The pressure and heat transfer distributions over the forebody are investigated for ranges of Reynolds number, Mach number, wall-to-freestream temperature ratio, and nose bluntness. Comparison of the test results and CFD predictions show that good agreement for the heat transfer distributions is achieved. However, the predicted pressure distribution on the forward blunted wedge was consistently underpredicted by 18-33% relative to the experimental measurements. Several phenomena were investigated in an attempt to explain the discrepancy between predicted and measured pressure distributions, including classical viscous leading-edge interactions, blunt leading-edge interactions, slip flow effects, flow condensation, flow angularity, and facility Mach number uncertainty. Although the discrepancy in the forebody pressure ratios could be caused by a combination of the factors listed above, a deficiency in the modeling of the viscous interaction region by the CFD codes and facility flow angularity are shown to be the strongest contributors.

CAPTools - An interactive toolkit for mapping CFD codes onto parallel architectures

In a reactor severe accident, molten jet breakup and solidification are important behaviors after large pours of molten material fall into the coolant in-vessel or ex-vessel. However, heat and mass transfer processes inside melt during jet breakup have not been studied sufficiently. Existing research on jet fragmentation is relatively macroscopic, and the micro interface condensation details are not well studied. In this paper, a two-dimensional multiphase computational fluid dynamics (CFD) code with the Volume of Fluid (VOF) method and solidification model is applied to simulate molten jet breakup with surface solidification. The VOF model is used to capture the interface, study the details, and add the influence of solidification. Solidification and instability can be seen at the interface. In order to simulate melt solidification, an energy equation is modeled using an enthalpy-based formulation, and viscosity variation during phase change is taken into account. The comparative results between the CFD code and jet breakup experiments show that melt jet front position histories, breakup length, and breakup time are in good agreement with the experiments. The simulation results show that crust formation of the jet surface suppresses surface instability and jet breakup behavior. As the interfacial temperature decreases, the droplet cumulative mass fraction decreases, and the solidified metal proportion increases. The simulation results by the CFD code with the solidification model are valuable and important for understanding the molten jet breakup mechanism.

This paper describes the work done to achieve high parallel performance for an unstructured, unsteady turbomachinery computational fluid dynamics (CFD) code. The aim of the work described here is to be able to scale problems to the thousands of processors that current and future machine architectures will provide. The CFD code is in design use in industry and is also used as a research tool at a number of universities. High parallel scalability has been achieved for a range of turbomachinery test cases, from steady-state hexahedral mesh cases to fully unsteady unstructured mesh cases. This has been achieved by a combination of code modification and consideration of the parallel partitioning strategy and resulting load balancing. A sliding plane option is necessary to run fully unsteady multistage turbomachinery test cases and this has been implemented within the CFD code. Sample CFD calculations of a full turbine including parts of the internal air system are presented.

Thermal-hydraulic analysis of the TRADE spallation target

A computational fluid dynamics (CFD) code, developed at Argonne national laboratory to simulate turbulent mixing, combustion reaction, radiation heat transfer, and pollutant kinetics of combustion flow, has been used to study various melting furnaces. The code employs an integral approach to incorporate a lumped combustion reaction model in the flow calculation and a separate hybrid technique to perform pollutant kinetics calculations. The CFD code has been validated with experimental data collected from industrial furnaces and then used for a parametric study of various furnace geometries and operation conditions. The furnace configuration greatly effected the flow property distribution as well as the combustion efficiency. The air injection velocity effected the flow penetration and the species mixing. The injection angle also significantly effected the species mixing. And finally, the equivalence ratio effected the temperature and pollutant concentrations. The study demonstrates that CFD can be a useful tool for analyzing the flow field of the combustion space in industrial furnaces.

Computational fluid dynamic (CFD) simulations in support of the SHARP-B2 (Slender Hypervelocity Aerothermodynamic Research Probes) program are presented. Pre-flight simulations of the aerothermal flight environment have been made to aid in the design of the flight hardware and instrumentation. Simulations of a critical flight qualification ground test performed in the 20 MW Panel Test Facility (PTF) arc jet, which is equipped with a semi-elliptical nozzle, have also been performed. The arc jet simulations were performed with the same CFD code that is used in the flight calculations. For both the flight and the arc jet simulations, the CFD results are used to provide surface heat transfer coefficients that serve as boundary conditions for an in-depth conduction thermal response code. By using the same CFD tool to compute the aerothermal environments for both the ground test and the flight test environment, useful comparisons are made regarding the traceability of the ground test environment to flight. Introduction The NASA Ames Research Center has been developing new Ultra High Temperature Ceramics (UHTC's) with the goal that these materials will enable sharp leading edges to be used for future space vehicles. Because of their unique thermo-structural properties, these materials are capable of operation at temperatures near 5100 °F without ablation. The materials have been developed and tested in ground-based arc jet facilities, and a flight test program designated SHARP (Slender Hypervelocity Aerothermodynamic Research Probes) has been initiated. The first flight demonstration, SHARP-B1, incorporated a UHTC nosetip mounted on a U.S. Air Force Mkl2A entry vehicle in collaboration with Sandia National Laboratory (SNL), and was successfully flown in May, 1997. The second flight test, SHARP-B2, which incorporated four instrumented UHTC strakes mounted on the side of the entry vehicle, was flown in September 2000. The goal of these flight tests is to assess the performance of the materials under entry conditions. The use of Computational Fluid Dynamic (CFD) simulations for the design of Thermal Protection Systems (TPS) has evolved to an extent that they have been integrated into the preliminary design process of Reusable Launch Vehicle (RLV) programs such as the X-33. Real-gas CFD simulations with finite-rate chemistry and proper boundary conditions for coupling with material thermal response codes are now routinely performed on high-end laboratory workstations. Much of the ground based testing of advanced TPS components is done in arc-heated test facilities such as those located in the Arc Jet Complex at the NASA Ames Research Center. These facilities are capable of simulating the high enthalpy flow environment experienced by entry vehicles for a broad range of test conditions and configurations. Semi-elliptical nozzles are available for the testing of flat panels in a high enthalpy boundary layer environment. CFD is • Aerospace Engineer, Reacting Flow Environments Branch, Member AIAA. t Senior Research Scientist, ELORET, Member AIAA. Copyright © 2001 by the American Institute of Aeronautics and Astronautics, Inc. No copyright is asserted in the United States under Title 17, U.S. Code. The U.S. Government has a royalty-free license to exercise all rights under the copyright claimed herein for Governmental Purposes. All other rights are reserved by the copyright owner. (c)2001 American Institute of Aeronautics & Astronautics or Published with Permission of Author(s) and/or Author(s)' Sponsoring Organization.

I. Abstract Computational Fluid Dynamics (CFD) based simulations, along with parametric and non-parametric Reduced-Order Models (ROM) for gust responses are presented. A CFD code is enhanced to simulate responses of an airfoil to arbitrary shaped gust inputs. Time-domain Auto-Regressive-Moving-Average (ARMA) models are identified based on CFD responses to random gust excitations, using system-identification methods. Responses to discrete gusts of various shapes, amplitudes and gradient lengths are computed via the ROMs and compared to responses simulated directly by the CFD code. The ROMs predict the lift and pitching moment histories accurately throughout the subsonic and transonic regimes. They offer significant savings in computational resources compared to the full CFD simulation, since only one CFD run is required for ROM identification, from which responses to arbitrary shaped gusts can be rapidly estimated. The combination of ROMs and full CFD simulations offers a computationally efficient tool set of various-fidelity time-domain models for gust responses. The ROMs can be used for rapid tuned-gust analyses, and the critical cases can be simulated with a full CFD run, providing pressure distribution for airframe structural design.

Comparison of unshrouded impeller analysis and experiment

Accurate prediction of bubble behavior in large diameter pipes is crucial for evaluating the performance of safety systems, steam generators, and heat exchangers in nuclear systems. Bubble behavior in large diameter pipes under two-phase flow significantly differs from that in small pipes. With the increasing use of computational fluid dynamics (CFD) codes, predicting interfacial area concentration (IAC) is critical for understanding multi-dimensional bubble behavior. This study developed a two-group local bubble size model for bubbly, slug, and churn flows under adiabatic conditions. The model includes correlations for void fraction and bubble sizes of two groups, which were implemented into CFD codes and validated against experimental data from large diameter pipes with low-pressure air–water flow. Results show the model's prediction accuracy surpasses existing correlations. The developed correlations are applicable across a range of flow conditions covering pipe diameters in the range 0.05–0.152 m, pressures from atmospheric to 300 kPa, superficial liquid velocities from 0.25 m/s to 2.85 m/s, and superficial gas velocities from 0.04 m/s to 5.48 m/s. The model is expected to enhance the prediction capabilities of CFD codes for the adiabatic two-group two-phase flows in the large diameter pipes.

As modern military aircraft become lighter, faster and more maneuverable, the consideration of aeroelastic effects during the design process can provide significant benefits. Computational aeroelasticity provides an attractive alternative to wind tunnel testing of flexible models in terms of accurately predicting and simulating the various linear and non-linear phenomena in a cost effective way. Computational Structural Dynamic (CSD) and Computational Fluid Dynamic (CFD) codes have reached a level of development where they can accurately analyse the structural and fluid behaviour. Aeroelastic simulation of individual components of an aircraft is now commonly done but problems arise when simulating a whole aircraft configuration. This is because the CSD solver calculates the elastic response' of the aircraft on a structural grid which usually does not coincide with the CFD surface grid and hence a scheme is required to transfer displacement and force values between the CSD and CFD grids. The various aerodynamic surface patches are driven by different structural components which may require different transformation methods. For example a fuselage, if modelled as a 1-dimensional beam, would require a different transformation technique than the wings which are modelled as 2-dimensional plates. To address this, a modified version of the Constant Volume Tetrahedron (CVT) transformation scheme is proposed for 1-dimensional structural grids. A tagging procedure is used where the fluid grid nodes are identified as being driven by 1 or 2-dimensional structural components and then the appropriate version of the transformation scheme is applied. The other difficulty is that the component interfaces in the fluid grid need to match up properly for the simulation to be successful. To overcome this a weighting method has been developed which forces the grid points at the component interfaces of the fluid grid to match up correctly by averaging the transformation within a predefined hierarchy. In the current work, this methodology has been demonstrated on a generic F16 aircraft configuration. The robustness of the transformation technique is evaluated by using a number of structural models to drive the fluid surface motions.

The hydraulic design optimization and performance analysis of a mixed-flow pump for marine waterjet propulsion has been carried out using mean streamline analysis and three-dimensional computational fluid dynamics (CFD) code. In the present study, conceptual design optimization has been formulated with a nonlinear objective function to minimize fluid dynamic losses and then a commercial CFD code was incorporated to allow for detailed flow dynamic phenomena in the pump system. The newly designed mixedflow model pump has been tested in the laboratory. Predicted performance curves by the CFD code agree fairly well with experimental data for a newly designed mixed-flow pump over the normal operating conditions. The design and prediction methods presented herein can be used efficiently as a unified hydraulic design process of mixed-flow pumps for marine waterjet propulsion.

The Use of Computational Fluid Dynamic Code for Modelling Spray from a Mistblower

Transitional flow has a significant impact on vehicles operating at supersonic and hypersonic speeds. An economic way to simulate this problem is to use computational fluid dynamics (CFD) codes. However, not all CFD codes can solve transitional flows. This paper examines the ability of the Spalart–Allmaras one-equation BCM (SA-BCM) transitional model to solve hypersonic transitional flow, implemented in the open-source CFD code Eilmer. Its performance is validated via existing wind tunnel data. Eight different hypersonic flow conditions are applied. A flat plate model is built for the numerical tests. The results indicate that the existing SA-BCM model is sensitive to the freestream turbulence intensity and the grid size. It is not accurate in all the test cases, though the transitional length can be matched by tuning the freestream intensity. This is likely due to the intermittency term of the SA-BCM model not being appropriately calibrated for high-velocity flow, though if the model can be recalibrated it may be able to solve the general high-velocity flows. Although the current SA-BCM model is only accurate under certain flow conditions after one calibration process, it remains attractive to CFD applications. As a one-equation model, the SA-BCM model runs much faster than multiple-equation flow models.