Distributed memory systems have gained immense popularity due to their favorable price/performance ratios. This study seeks to reduce the complexities, involved in developing parallel applications for distributed memory systems. The Loci system is a coordination framework which was developed to eliminate most of the accidental complexities involved in numerical simulation software development. A distributed memory version of Loci is developed and has been tested and validated using a finite-rate chemically reacting flow solver developed in the sequential Loci framework. The application developed in the original sequential version of Loci was parallelized with minimal changes in its source code. A comparison with the results from the original sequential version guarantees a correct implementation. The performance measurements indicate that an efficient implementation has been achieved.

No description available.

No description available.

A coupled density based solver in the framework of foam-extend is used to perform simulations of transonic flows. The solver is based on an explicit and time-accurate algorithm and is coupled to a compressible Unsteady Reynolds-Averaged Navier-Stokes (URANS) and a Large Eddy Simulation (LES) module. The solver is first attested on canonical compressible flow scenarios such as a 1-D shock tube and the transonic flow through a 2-D channel. Following this, a 2-D URANS simulation of the flow within the passages of a High Pressure Turbine Nozzle Guide Vane (HPT-NGV) is performed and compared against experimental data. Finally, preliminary results of a 3-D LES on a simplified geometry of the HPT-NGV are presented. In the future, this numerical setup will be used to study indirect combustion noise in aircraft engines.

Trajectory modeling is one of the most important aspects of any mission design. The trajectory should be able to propagate the S/C to the final destination while optimizing the flight duration, the total change in velocity and also the total launch mass. The Spacecraft Trajectory Optimizer (STO) tool described in this report first solves the Gauss Lambert problem and generates initial departure and arrival conditions which can also be expressed as porkchop plots. These initial conditions are then used as input to optimize the flight steps which are based on a patched conic approximation with the elliptical transfer with respect to the Sun and the hyperbolic transfers at the departure and arrival planet's sphere of influence. The tool is completely based on MATLAB 2007 or later and uses ODE45 for trajectory propagation and FMINCON with Active-set algorithm for optimization. The results obtained in house were compared with four Mars Sample return orbits calculated at ESOC and there is a very good correlation between the required change in velocities and transfer duration for e.g. Orbit case: O22S, ESOC values: total Delta V = 3.946 - 4.119 [km/s], TOF = 329 - 342 [days] &amp; STO values: Delta V = 3:986 [km/s] &amp; TOF = 335 [days]. The in house data was also used as an input in the System Tool Kit (a professional trajectory calculation software) for modeling an interplanetary trajectory to Mars and the S/C arrived at Mars without any optimization. Therefore, even though the STO does not have all the capabilities of a professional software it can be used for preliminary mission analysis as it offers quite accurate results for interplanetary transfers. / <p>Validerat; 20131127 (global_studentproject_submitter)</p>

The availability of high performance computing clusters has allowed scientists and engineers to study more challenging problems. However, new algorithms need to be developed to take advantage of the new computer architecture (in particular, distributed memory clusters). Since the solution of linear systems still demands most of the computational effort in many problems (such as the approximation of partial differential equation models) iterative methods and, in particular, efficient preconditioners need to be developed. In this study, we consider application of incomplete LU (ILU) preconditioners for finite element models to partial differential equations. Since finite elements lead to large, sparse systems, reordering the node numbers can have a substantial influence on the effectiveness of these preconditioners. We study two implementations of the ILU preconditioner: a stucturebased method and a threshold-based method. The main emphasis of the thesis is to test a variety of breadth-first ordering strategies on the convergence properties of the preconditioned systems. These include conventional Cuthill-McKee (CM) and Reverse Cuthill-McKee (RCM) orderings as well as strategies related to the physical distance between nodes and post-processing methods based on relative sizes of associated matrix entries. Although the success of these methods were problem dependent, a number of tendencies emerged from which we could make recommendations. Finally, we perform a preliminary study of the multi-processor case and observe the importance of partitioning quality and the parallel ILU reordering strategy. / Master of Science

Owing to advancements in component re-use technology, component-based software development (CBSD) has come a long way in developing complex commercial software systems while reducing software development time and cost. However, assembling distributed resource-constrained and safety-critical systems using current assembly techniques is a challenge. Within complex systems when there are numerous ways to assemble the components unless the software architecture clearly defines how the components should be composed, determining the correct assembly that satisfies the system assembly constraints is difficult. Component technologies like CORBA and .NET do a very good job of integrating components, but they do not automate component assembly; it is the system developer's responsibility to ensure thatthe components are assembled correctly. In this thesis, we first define a component-based system assembly (CBSA) technique called "Constrained Component Assembly Technique" (CCAT), which is useful when the system has complex assembly constraints and the system architecture specifies component composition as assembly constraints. The technique poses the question: Does there exist a way of assembling the components that satisfies all the connection, performance, reliability, and safety constraints of the system, while optimizing the objective constraint? To implement CCAT, we present a powerful framework called "CoBaSA". The CoBaSA framework includes an expressive language for declaratively describing component functional and extra-functional properties, component interfaces, system-level and component-level connection, performance, reliability, safety, and optimization constraints. To perform CBSA, we first write a program (in the CoBaSA language) describing the CBSA specifications and constraints, and then an interpreter translates the CBSA program into a satisfiability and optimization problem. Solving the generated satisfiability and optimization problem is equivalent to answering the question posed by CCAT. If a satisfiable solution is found, we deduce that the system can be assembled without violating any constraints. Since CCAT and CoBaSA provide a mechanism for assembling systems that have complex assembly constraints, they can be utilized in several industries like the avionics industry. We demonstrate the merits of CoBaSA by assembling an actual avionic system that could be used on-board a Boeing aircraft. The empirical evaluation shows that our approach is promising and can scale to handle complex industrial problems.

The development of a shape optimization solver using the existing Computational Fluid Dynamics (CFD) codes is taken up as topic of research in this thesis. A shape optimizer was initially developed based on Genetic Algorithm (GA) coupled with a CFD solver in an earlier work. The existing CFD solver is based on Kinetic Flux Vector Splitting and uses least squares discretization. This solver requires a cloud of points and their connectivity set, hence this CFD solver is a meshless solver. The advantage of a meshless solver is utilised in avoiding re-gridding (only connectivity regeneration is required) after each shape change by the shape optimizer. The CFD solver is within the optimization loop, hence evaluation of CFD solver after each shape change is mandatory. Although the earlier shape optimizer developed was found to be robust, but it was taking enoromous amount of time to converge to the optimum solution (details in Appendix). Hence a new evolving method, Ant Colony Optimization (ACO), is implemented to replace GA. A shape optimizer is developed coupling ACO and the meshless CFD solver. To the best of the knowledge of the present author, this is the first time when ACO is implemented for aerodynamic shape optimization problems. Hence, an exhaustive validation has become mandatory. Various test cases such as regeneration problems of (1) subsonic - supersonic nozzle with a shock in quasi - one dimensional flow (2) subsonic - supersonic nozzle in a 2-dimensional flow field (3) NACA 0012 airfoil in 2-dimensional flow and (4) NACA 4412 airfoil in 2-dimensional flow have been successfully demonstrated. A comparative study between GA and ACO at algorithm level is performed using the travelling salesman problem (TSP). A comparative study between the two shape optimizers developed, i.e., GA-CFD and ACO-CFD is carried out using regeneration test case of NACA 4412 airfoil in 2-dimensional flow. GA-CFD performs better in the initial phase of optimization and ACO-CFD performs better in the later stage. We have combined both the approaches to develop a hybrid GA-ACO-CFD solver such that the advantages of both GA-CFD and ACO-CFD are retained with the hybrid method. This hybrid approach has 2 stages, namely, (Stage 1) initial optimum search by GA-CFD (coarse search), the best members from the optimized solution from GA-CFD are segregated to form the input for the fine search by ACO-CFD and (Stage 2) final optimum search by ACO-CFD (fine search). It is observed that this hybrid method performs better than either GA-CFD or ACO- CFD, i.e., hybrid method attains better optimum in less number of CFD calls. This hybrid method is applied to the following test cases: (1) regeneration of subsonic-supersonic nozzle with shock in quasi 1-D flow and (2) regeneration of NACA 4412 airfoil in 2-dimensional flow. Two applications on shape optimization, namely, (1) shape optimization of a body in strongly rotating viscous flow and (2) shape optimization of a body in supersonic flow such that it enhances separation of binary species, have been successfully demonstrated using the hybrid GA-ACO-CFD method. A KFVS based binary diffusion solver was developed and validated for this purpose. This hybrid method is now in a state where industrial shape optimization applications can be handled confidently.

Computing 3D high lift flows has been a challenge to the CFD community because of three important reasons: complex physics, complex geometries and large computational requirements. In the recent years, considerable progress has been made in understanding the suitability of various CFD solvers in computing 3D high lift flows, through the systematic studies carried out under High Lift Prediction workshops. The primary focus of these workshops is to assess the ability of the CFD solvers to predict CLmax and αmax associated with the high lift flows, apart from the predictability of lift and drag of such flows in the linear region. Now there is a reasonable consensus in the community about the ability of the CFD solvers to predict these quantities and fresh efforts to further understand the ability of the CFD solvers to predict more complex physics associated with these flows have already begun. The goal of this thesis is to assess the capability of the computational methods in predicting such complex flow phenomena associated with the 3D High-Lift systems. For evaluation NASA three element Trapezoidal wing configuration which poses a challenging task in numerical modeling was selected. Unstructured data based 3D RANS solver HiFUN (HiFUN stands for High Resolution Flow Solver for UNstructured Meshes) is used in investigating the high lift flow. The computations were run fully turbulent, using the one equation Spalart-Allmaras turbulence model. A summary of the results obtained using the flow solver HiFUN for the 3D High lift NASA Trapezoidal wing are presented. Hybrid unstructured grids have been used for the computations. Grid converged solution obtained for the clean wing and the wing with support brackets, are compared with experimental data. The ability of the solver to predict critical design parameters associated with the high lift flow, such as αmax and CLmax is demonstrated. The utility of the CFD tools, in predicting change in aerodynamic parameters in response to perturbational changes in the configuration is brought out. The solutions obtained for the high lift configuration from two variants of the Spalart-Allmaras turbulence model are compared. To check the unsteadiness in the flow, particularly near stall, unsteady simulations were performed on static grid. Lastly, hysteresis on lower leg of lift curve is discussed, the results obtained for quasi-steady and dynamic unsteady simulations are presented. Inferences from the study on useful design practices pertaining to the 3D high lift flow simulations are summarized.

L’acceptation et l’utilisation des services Web en industrie se développent de par leursupport au développement d’application distribuées comme compositions d’entitéslogicielles plus simples appelées services. En complément à la vérification, le testpermet de vérifier la correction d’une implémentation binaire (code source nondisponible) par rapport à une spécification. Dans cette thèse, nous proposons uneapproche boîte-noire du test de conformité de compositions de services centralisées(orchestrations). Par rapport à l’état de l’art, nous développons une approchesymbolique de façon à éviter des problèmes d’explosion d’espace d’état dus à la largeutilisation de données XML dans les services Web. Cette approche est basée sur desmodèles symboliques (STS), l’exécution symbolique de ces modèles et l’utilisationd’un solveur SMT. De plus, nous proposons une approche de bout en bout, quiva de la spécification à l’aide d’un langage normalisé d’orchestration (ABPEL) etde la possible description d’objectifs de tests à la concrétisation et l’exécution enligne de cas de tests symboliques. Un point important est notre transformation demodèle entre ABPEL et les STS qui prend en compte les spécifications sémantiquesd’ABPEL. L’automatisation de notre approche est supportée par un ensemble d’outilsque nous avons développés. / Web services are gaining industry-wide acceptance and usage by fostering the developmentof distributed applications out of the composition of simpler entities calledservices. In complement to verification, testing allows one to check for the correctnessof a binary (no source code) service implementation with reference to a specification.In this thesis, we propose black box conformance testing approach for centralizedservice compositions (orchestrations). With reference to the state of the art, wedevelop a symbolic approach in order to avoid state space explosion issues due to theXML data being largely used in Web services. This approach is based on symbolicmodels (STS), symbolic execution, and the use of a satisfiability modulo theory(SMT) solver. Further, we propose a comprehensive end-to-end approach that goesfrom specification using a standard orchestration language (ABPEL), and the possibledescription of test purposes, to the online realization and execution of symbolic testcases against an implementation. A crucial point is a model transformation fromABPEL to STS that we have defined and that takes into account the peculiarities ofABPEL semantics. The automation of our approach is supported by a tool-chainthat we have developed.