Browsing by Author "Meade, Andrew J., Jr."
Now showing 1 - 20 of 41
Results Per Page
Sort Options
Item A Computational Analysis of Novel Non-circular Nozzles(2014-04-24) Baskici, Gokhan; Akin, John Edward.; Tezduyar, Tayfun E.; Meade, Andrew J., Jr.This thesis presents a computational analysis for determining the flow properties of novel non-circular nozzles. In last few decades, non-circular nozzles have been investigated intensively due to their remarkably superior capabilities in enhancing mass entrainment over circular nozzles. In this thesis, to increase the amount of ambient fluid entrained in the jet flow, three different geometrical modifications are applied to non-circular nozzles. These modifications include changing contraction profiles, the twist angles of transition, and nozzle lengths. The flow properties of air emanating from geometrically modified non-circular nozzles are studied by using computational fluid dynamics (CFD) in the Star CCM+ fluid solver. This computational analysis shows that suitably modified non-circular nozzles are very effective passive flow conditioning devices and can modify the flow field. Particularly, nozzle with a sharp exit and large twist angle entrains the ambient fluid at a higher rate than the circular and other modified non-circular nozzles.Item A greedy algorithm for learning pilot ratings from helicopter shipboard dynamic interface tests(2007) Srivastava, Ankur; Meade, Andrew J., Jr.In a real world pattern recognition application a user cannot assess the performance of a classifier on an unlabeled data set. Classifiers cannot give their best performance because they require user-controlled parameters. As a Solution, a Sequential Function Approximation (SFA) method has been' developed for classification that determines the values of the control parameters during learning. In this dissertation, experiments were carried out on real world data sets where SFA, using only the training subset, had comparable performance to a number of other popular classification schemes whose user-defined parameters were optimized utilizing the entire data set. By the statistical significance of the results it was concluded at 95% confidence that the performance of SFA will be equivalent or significantly better than those of the other popular classification tools. After establishing SFA as a proper classification tool in this dissertation, it is applied to a US Navy flight test problem. The current problem at hand is to predict pilot ratings from HH-60H Sea-Hawk helicopters based on 369 at sea take-off and landing DI tests. Least significant inputs with respect to classification were pointed out with the potential of accelerating through the DI test matrix. And finally an effort was made to give the DI test pilots an estimate of how many tests were necessary to be conducted before generating enough data for the SFA classification tool to satisfactorily learn.Item A Multiscale Model of the Enhanced Heat Transfer in a CNT-Nanofluid System(2011) Lee, Jonathan Winnie; Barrera, Enrique V.; Meade, Andrew J., Jr.Over the last decade, much research has been done to understand the role of nanoparticles in heat transfer fluids. While experimental results have shown "anomalous" thermal enhancements and non-linear behavior with respect to CNT loading percentage, little has been done to replicate this behavior from an analytical or computational standpoint. This study is aimed towards using molecular dynamics to augment our understanding of the physics at play in CNT-nanofluid systems. This research begins with a heat transfer study of individual CNTs in a vacuum environment. Temperature gradients are imposed or induced via various methods. Tersoff and AIREBO potentials are used for the carbon-carbon interactions in the CNTs. Various chirality CNTs are explored, along with several different lengths and temperatures. The simulations have shown clear dependencies upon CNT length, CNT chirality, and temperature. Subsequent studies simulate individual CNTs solvated in a simple fluidic box domain. A heat flux is applied to the domain, and various tools are employed to study the resulting heat transfer. The results from these simulations are contrasted against the earlier control simulations of the CNT-only domain. The degree by which the solvation dampens the effect of physical parameters is discussed. Effective thermal conductivity values are computed, however the piecewise nature of the temperature gradient makes Fourier's law insufficient in interpretting the heat transfer. Nevertheless, the computed effective thermal conductivities are applied to classical models and better agreement with experimental results is evident. Phonon spectra of solvated and unsolvated CNTs are compared. However, a unique method utilizing the Irving-Kirkwood relations reveals the spatially-localized heat flux mapping that fully illuminates the heat transfer pathways in the solid-fluid composite material. This method confirms why conventional models fail at predicting effective thermal conductivity. Specifically, it reveals the volume of influence that the CNT has on its surrounding fluid.Item A numerical study of a laminar, compressible boundary layer about an airfoil(1992) Strong, Stuart Lawson; Meade, Andrew J., Jr.The primary goal of this study is to develop a finite element method for the calculation of an attached, two-dimensional, laminar, compressible boundary layer about an air-foil in a subsonic free stream. An introduction to the subsequent viscous-inviscid interaction model is also given. The two-dimensional partial differential equations are reduced to integral equations that are independent of density and resemble the weak form of the two dimensional incompressible boundary layer equations. A Galerkin finite element method is applied to this Dorodnitsyn formulation and discretized across the layer using linear interpolation functions. The finite element discretization yields a system of first order ordinary differential equations, solved by an implicit, non-iterative finite difference marching scheme in the stream-wise direction. Results presented include the coefficient of friction and displacement thickness about a circular cylinder in an incompressible freestream, the compressible boundary layer about a NACA 0012 airfoil, and a validation of the linear thermal equation.Item A Penalty Method Approach to Experimental Data Coupling in the Sequentially Optimized Meshfree Approximation(2014-04-23) Wood, Jeffre; Meade, Andrew J., Jr.; Akin, John Edward.; Tezduyar, Tayfun E.This thesis presents an overview of recent changes that have been made to the Sequentially Optimized Meshfree Approximation (SOMA) and provides a proof of concept for a penalty method algorithm for coupling experimental data with a computational fluid dynamics (CFD) solver. The penalty method provides a means to apply experimental data to improve SOMA's convergence rate and time to convergence. Using the driven cavity problem as validation, approximations produced using this adaptation are shown. The results successfully reproduces the flow from which the data was taken creating a means of using limited experimental data to generate a complete picture of a simulation domain. Flexibility in data type, location, and quantity are demonstrated as well as the effects of experimental error on the results and a means of negating it.Item A study of neural networks in thermal systems(1994) Penaranda, Guillermo; Meade, Andrew J., Jr.Neural networks have been found to be useful as a technique for the modeling of non-linear functions or processes that involve several variables. The primary goal of this thesis is to explore the feasibility of applying feedforward backpropagation neural networks in the optimization of multistage thermal systems. Basically, the idea consists of using neural networks as a function approximation technique for each stage of a multistage process. After the successful approximation, existing optimization methods are used to obtain the parameters that optimize the system. In addition, it is shown how feedforward backpropagation neural networks can be used in solving calculus of variation problems, by separating the process into discrete stages, thus forming a multistage process problem. Finally, parallel work was done in developing a faster deterministic training algorithm, as an alternative to the time consuming backpropagation training algorithm.Item Adaptive Techniques Applied to the Sequentially Optimized Meshfree Approximation(2014-04-22) Mittelman, Rachel; Akin, John Edward.; Stanciulescu, Ilinca; Meade, Andrew J., Jr.This thesis advances the meshless Sequentially Optimized Meshfree Approximation (SOMA) from a fixed grid to an adaptive one by applying residual-based adaptive techniques. In its fixed grid form, SOMA constructs an approximation of an equation solution using optimized radial basis functions (RBFs), but deletes the RBF parameters once each basis function is appropriately added. The first proposed method saves this information, constructs an approximation of the solution, and intelligently adds points to the problem domain. The second proposed method is a flexible interpolation scheme which does not require this basis saving technique, although the two techniques can be combined. When applied to various equations, these adaptive algorithms demonstrate the convergence required to achieve a satisfactory level of precision, saving time and computational effort for the same mathematical result as a denser grid. Applications of this algorithm include function approximation as well as differential equations which demonstrate its capability and robustness.Item An object-oriented framework for solving model problems using the sequential function approximation algorithm(2001) Fernandez, Alvaro Agustin; Meade, Andrew J., Jr.This dissertation describes and tests an Object-Oriented framework, written in Fortran 90, for the Sequential Function Approximation (SFA) algorithm. The SFA algorithm is a meshless method which places its basis functions in the domain sequentially, using optimization techniques. The framework described herein allows the user to define the domain, boundary conditions, and governing equations of 1-D and 2-D problems with minimal user coding, and to solve them using the SFA method. This work advances the state of knowledge in the fields of meshless methods in general and of the SFA method in particular. Unsteady transport problems are solved for the first time with the SFA method: diffusive, convective-diffusive, and purely convective problems are solved using a semi-discrete approach and stabilized with the Streamline-Upwind Petrov-Galerkin (SUPG) technique. Additionally, some light is shed on the role of consistency. SFA is placed within the broader context of meshless methods, and made consistent by transforming it into a sequentially solved Partition of Unity (POU) method. Consistency is experimentally found to improve the convergence behavior of all model problems solved. The improvement is most notable in problems with convection phenomena, although some improvement is seen even in purely diffusive problems. Other hypotheses regarding the SFA method are investigated as well.Item Applying regularization to the fusion of empirical and numerical data(2001) Sonneborn, Hans Christoph; Meade, Andrew J., Jr.A method is presented to integrate computational and experimental data sets, allowing development of an accurate and comprehensive model of a system response surface. The method is derived from Generalized Tikhonov Regularization for ill-posed problems. Through several numerical examples, the application of the new method to perform data fusion is demonstrated. The results show that a priori computational models may be improved by integrating experimental or computational data from other sources. The results also demonstrate the ability of the method to use an a priori model to smoothly interpolate sparse, noisy data. The method is compared to an earlier iterative approach for determining the regularization parameter. The limitations of the methodology in certain problem formulations are examined and suggestions for future work are described.Item Calibration of Flush Air Data Sensing Systems Using Surrogate Modeling Techniques(2011) Srivastava, Ankur; Meade, Andrew J., Jr.In this work the problem of calibrating Flush Air Data Sensing (FADS) has been addressed. The inverse problem of extracting freestream wind speed and angle of attack from pressure measurements has been solved. The aim of this work was to develop machine learning and statistical tools to optimize design and calibration of FADS systems. Experimental and Computational Fluid Dynamics (EFD and CFD) solve the forward problem of determining the pressure distribution given the wind velocity profile and bluff body geometry. In this work three ways are presented in which machine learning techniques can improve calibration of FADS systems. First, a scattered data approximation scheme, called Sequential Function Approximation (SFA) that successfully solved the current inverse problem was developed. The proposed scheme is a greedy and self-adaptive technique that constructs reliable and robust estimates without any user-interaction. Wind speed and direction prediction algorithms were developed for two FADS problems. One where pressure sensors are installed on a surface vessel and the other where sensors are installed on the Runway Assisted Landing Site (RALS) control tower. Second, a Tikhonov regularization based data-model fusion technique with SFA was developed to fuse low fidelity CFD solutions with noisy and sparse wind tunnel data. The purpose of this data model fusion approach was to obtain high fidelity, smooth and noiseless flow field solutions by using only a few discrete experimental measurements and a low fidelity numerical solution. This physics based regularization technique gave better flow field solutions compared to smoothness based solutions when wind tunnel data is sparse and incomplete. Third, a sequential design strategy was developed with SFA using Active Learning techniques from the machine learning theory and Optimal Design of Experiments from statistics for regression and classification problems. Uncertainty Sampling was used with SFA to demonstrate the effectiveness of active learning versus passive learning on a cavity flow classification problem. A sequential G-optimal design procedure was also developed with SFA for regression problems. The effectiveness of this approach was demonstrated on a simulated problem and the above mentioned FADS problem.Item Computational Aerodynamics Modeling of Flapping Wings With Video-Tracked Locust-Wing Motion(2013-07-24) Puntel, Anthony; Tezduyar, Tayfun E.; Akin, John Edward.; Meade, Andrew J., Jr.; Takizawa, KenjiThe thesis focuses on special space--time computational techniquesintroduced recently for computational aerodynamics modeling of flapping wings of an actual locust. These techniques complement the Deforming-Spatial-Domain/Stabilized Space--Time (DSD/SST) formulation, which is the core computational technique. The DSD/SST formulation was developed for flows with moving interfaces, and the version used in the computations is "DST/SST-VMST," which is the space--time version of the residual-based variational multiscale (VMS) method. The special space--time techniques are based on using NURBS basis functions for the temporal representation of the motion of the locust wings. The motion data is extracted from the high-speed video recordings of a locust in a wind tunnel. In addition, temporal NURBS basis functions are used in representation of the motion of the volume meshes computed and also in remeshing. These ingredients provide an accurate and e fficient way of dealing with the wind tunnel data and the mesh. The thesis includes a detailed study on how the spatial and temporal resolutions influence the quality of the numerical solution.Item Construction of airfoil performance tables by the fusion of experimental and numerical data(2004) Navarrete, Jose; Meade, Andrew J., Jr.A method that combines experimental airfoil coefficient data with numerical data has been developed to construct airfoil performance tables given limited data sets. This work addresses the problem faced by engineers and aerodynamicists that currently rely on incomplete performance tables when researching airfoil characteristics. The method developed utilizes the Sequential Function Approximation (SFA) neural network tool and employs a simple regularization scheme to fuse multi-dimensional experimental and computational fluid dynamics (CFD) data efficiently. The method is considered an adaptive and robust tool requiring relatively little computational demand and minimal user dependence. An existing performance table for the NACA 0012 airfoil was used as a test case to verify the feasibility of the SFA-fused network. A second test case assesses the method's viability for a more realistic and challenging problem using highly sparse and scattered data sets for the SC1095 airfoil. Results from both studies realize the method's capability to make consistent approximations and smooth interpolations given only limited experimental data. Comparisons are made with other scattered data approximation techniques. The testing conditions, requirements, and limitations of this approach are discussed and future applications and recommendations are made.Item Developing Innovative Designs with Manufacturing Capability Using the Level Set Method(2012-09-05) Baradaran Nakhjavani, Omid; Meade, Andrew J., Jr.; Akin, John Edward.; Padgett, Jamie E.; Dick, Andrew J.This thesis discusses how to use topology and shape optimization, specifically the level set method, for innovative design. The level set method is a numerical algorithm that simulates the expansion of dynamic implicit surfaces. In this research, the equations for manufacturability are generated and solved through use of the level set method joined with the COMSOL multi-physics package. Specific constraints are added to make the optimization practical for engineering design. The resulting method was applied to design the best underlying support structure, conforming to both curvature and manufacturability constraints, for the longerons used with the International Space Station solar panels.Item Exploration of Tikhonov regularization for the fusion of experimental data and computational fluid dynamics(1999) Wang, Wei; Meade, Andrew J., Jr.A method is developed to fuse Computational Fluid Dynamics (CFD) simulations and experimental data through the use of Tikhonov regularization. Inviscid-Viscous Interaction and Thin-Layer Navier-Stokes Equation models are used to provide CFD solutions for the flow past NACA 0012 and RAE 2822 airfoils, respectively. The velocity profile within the boundary layer and the pressure coefficient on the surface of the airfoil are merged with the corresponding experimental data. A finite element approach is applied to accomplish the numerical solution of the Tikhonov regularization method. By using over- or under-relaxation technique, relatively few iterations are needed to achieve the convergence of the fusion method. The results demonstrate that a-priori CFD solutions of low fidelity can be improved by the experimental data with less computational cost compared with more sophisticated CFD models. Alternatively, the sparse and scattered experimental data are efficiently processed by utilizing CFD models as regularization. The limitations of the Tikhonov regularization method have been examined. The result shows that the fusion method has significant advantages over a nonlinear least-square polynomial approach for interpolating and extrapolating experimental data.Item Fluid--Structure Interaction Modeling of Modified-Porosity Parachutes and Parachute Clusters(2013-09-16) Boben, Joseph; Tezduyar, Tayfun E.; Akin, John Edward.; Meade, Andrew J., Jr.; Takizawa, KenjiTo increase aerodynamic performance, the geometric porosity of a ringsail spacecraft parachute canopy is sometimes increased, beyond the "rings" and "sails" with hundreds of "ring gaps" and "sail slits." This creates extra computational challenges for fluid--structure interaction (FSI) modeling of clusters of such parachutes, beyond those created by the lightness of the canopy structure, geometric complexities of hundreds of gaps and slits, and the contact between the parachutes of the cluster. In FSI computation of parachutes with such "modified geometric porosity," the flow through the "windows" created by the removal of the panels and the wider gaps created by the removal of the sails cannot be accurately modeled with the Homogenized Modeling of Geometric Porosity (HMGP), which was introduced to deal with the hundreds of gaps and slits. The flow needs to be actually resolved. All these computational challenges need to be addressed simultaneously in FSI modeling of clusters of spacecraft parachutes with modified geometric porosity. The core numerical technology is the Stabilized Space--Time FSI (SSTFSI) technique, and the contact between the parachutes is handled with the Surface-Edge-Node Contact Tracking (SENCT) technique. In the computations reported here, in addition to the SSTFSI and SENCT techniques and HMGP, we use the special techniques we have developed for removing the numerical spinning component of the parachute motion and for restoring the mesh integrity without a remesh. We present results for 2- and 3-parachute clusters with two different payload models. We also present the FSI computations we carried out for a single, subscale modified-porosity parachute.Item Fluid-Structure Interaction Modeling of Parachutes with Disreefing and Modified Geometric Porosity and Separation Aerodynamics of a Cover Jettisoned to the Spacecraft Wake(2012-04-24) Fritze, Matt; Tezduyar, Tayfun E.; Akin, John Edward.; Meade, Andrew J., Jr.; Takizawa, KenjiFluid--structure interaction (FSI) modeling of spacecraft parachutes involves a number of computational challenges. The canopy complexity created by the hundreds of gaps and slits and design-related modification of that geometric porosity by removal of some of the sails and panels are among the formidable challenges. Disreefing from one stage to another when the parachute is used in multiple stages is another formidable challenge. This thesis addresses the computational challenges involved in disreefing of spacecraft parachutes and fully-open and reefed stages of the parachutes with modified geometric porosity. The special techniques developed to address these challenges are described and the FSI computations are be reported. The thesis also addresses the modeling and computation challenges involved in very early stages, where the sudden separation of a cover jettisoned to the spacecraft wake needs to be modeled. Higher-order temporal representations used in modeling the separation motion are described, and the computed separation and wake-induced forces acting on the cover are reported.Item Fluid-Structure Interaction Modeling of the Reefed Stages of the Orion Spacecraft Main Parachutes(2014-04-25) Boswell, Cody W; Tezduyar, Tayfun E.; Akin, John Edward.; Meade, Andrew J., Jr.; Takizawa, KenjiSpacecraft parachutes are typically used in multiple stages, starting with a "reefed" stage where a cable along the parachute skirt constrains the diameter to be less than the diameter in the subsequent stage. After a certain period of time during the descent, the cable is cut and the parachute "disreefs" (i.e. expands) to the next stage. Computing the parachute shape at the reefed stage and fluid–-structure interaction (FSI) modeling during the disreefing involve computational challenges beyond those we have in FSI modeling of fully-open spacecraft parachutes. These additional challenges are created by the increased geometric complexities and by the rapid changes in the parachute geometry. The computational challenges are further increased because of the added geometric porosity of the latest design, where the "windows" created by the removal of panels and the wider gaps created by the removal of sails compound the geometric and flow complexity. Orion spacecraft main parachutes will have three stages, with computation of the Stage 1 shape and FSI modeling of disreefing from Stage 1 to Stage 2 being the most challenging. We present the special modeling techniques we devised to address the computational challenges and the results from the computations carried out. We also present the methods we devised to calculate for a parachute gore the radius of curvature in the circumferential direction. The curvature values are intended for quick and simple engineering analysis in estimating the structural stresses.Item Fluid–Structure Interaction Modeling of the Orion Spacecraft Drogue Parachutes(2014-04-25) Kolesar, Ryan; Tezduyar, Tayfun E.; Akin, John Edward.; Meade, Andrew J., Jr.; Takizawa, KenjiAt higher altitudes, prior to the deployment of the main parachutes, the Orion spacecraft descent to Earth will rely on deceleration by drogue parachutes. These parachutes have a ribbon construction, and in fluid–structure interaction (FSI) modeling this creates geometric and flow complexities comparable to those encountered in FSI modeling of the main parachutes, which have a ringsail construction. The drogue parachutes to be used with the Orion spacecraft have 24 gores, with 52 ribbons in each gore, resulting in hundreds of gaps that the flow goes through. We address this computational challenge, as was done for the main parachutes, with the Homogenized Modeling of Geometric Porosity (HMGP). Like the main parachutes, the drogue parachutes will be used in multiple stages, starting with a "reefed" stage where a cable along the parachute skirt constrains the diameter to be less than the diameter in the subsequent stage. After a certain period of time during the descent, the cable is cut and the parachute "disreefs" (i.e. expands) to the next stage. Computing the parachute shape at the reefed stage and FSI modeling during the disreefing involve computational challenges beyond those in FSI modeling of fully-open drogue parachutes. Orion spacecraft drogue parachutes will have three stages, with FSI modeling of disreefing from Stage 1 to Stage 2 being somewhat more challenging than disreefing from Stage 2 to Stage 3. We present the special modeling techniques we devised to address the computational challenges and the results from the computations carried out. We also present the methods we devised to calculate for a parachute gore the radius of curvature in the circumferential direction. The curvature values are intended for quick and simple engineering analysis in estimating the structural stresses. The flight envelope of the Orion drogue parachutes includes regions where the Mach number is high enough to require a compressible-flow solver. We present some preliminary computations for such cases.Item Low Dean number flows in helical ducts of rectangular cross section(1996) Thomson, David Lee; Bayazitoglu, Yildiz; Meade, Andrew J., Jr.The flow in a helical duct is characterized by increased fluid mixing, accomplished by the inducement of a secondary flow in the plane normal to the helix centerline. Two independent phenomena interact to produce this secondary flow. First, the curvature of the duct (i.e. its torroidal nature) causes Dean's type recirculation. Second, the torsion due to the non-planarity of the helix causes additional mixing. The secondary flow alters the axial velocity profile and increases the pressure drop compared to a straight duct. Imposing a rectangular cross section on such a duct complicates the analysis compared to a circular or elliptical cross section. A series solution based on curvature is introduced. The components of the series are determined using appropriate eigenfunction expansions. However, the resulting low order solution is limited to low Dean number flows. The analytical solution is useful for flows where curvature (torroidal ducts) or curvature and torsion (helical ducts) are important.Item Nonlinear Aeroelastic Analysis of UAVs: Deterministic and Stochastic Approaches(2012-09-05) Sukut, Thomas; Spanos, Pol D.; Meade, Andrew J., Jr.; Dick, Andrew J.Aeroelastic aspects of unmanned aerial vehicles (UAVs) is analyzed by treatment of a typical section containing geometrical nonlinearities. Equations of motion are derived and numerical integration of these equations subject to quasi-steady aerodynamic forcing is performed. Model properties are tailored to a high-altitude long-endurance unmanned aircraft. Harmonic balance approximation is employed based on the steady-state oscillatory response of the aerodynamic forcing. Comparisons are made between time integration results and harmonic balance approximation. Close agreement between forcing and displacement oscillatory frequencies is found. Amplitude agreement is off by a considerable margin. Additionally, stochastic forcing effects are examined. Turbulent flow velocities generated from the von Karman spectrum are applied to the same nonlinear structural model. Similar qualitative behavior is found between quasi-steady and stochastic forcing models illustrating the importance of considering the non-steady nature of atmospheric turbulence when operating near critical flutter velocity.
- «
- 1 (current)
- 2
- 3
- »