The solution of a single optimization problem often requires computationally-demanding evaluations; this is especially true in optimal design of engineering components and systems described by partial differential equations. We present a technique for the rapid and reliable optimization of systems characterized by linear-functional outputs of partial differential equations with affine parameter dependence. The critical ingredients of the method are: (i) reduced-basis techniques for dimension reduction in computational requirements; (ii) an "off-line/on-line" computational decomposition for the rapid calculation of outputs of interest and respective sensitivities in the limit of many queries; (iii) a posteriori error bounds for rigorous uncertainty and feasibility control; (iv) Interior Point Methods (IPMs) for efficient solution of the optimization problem; and (v) a trust-region Sequential Quadratic Programming (SQP) interpretation of IPMs for treatment of possibly non-convex costs and constraints. / Singapore-MIT Alliance (SMA)

In this paper we describe an approach to generating low-order models of spiral inductors that accurately capture the dependence on both frequency and geometry (width and spacing) parameters. The approach is based on adapting a multiparameter Krylov-subspace based moment matching method to reducing an integral equation for the three dimensional electromagnetic behavior of the spiral inductor. The approach is demonstrated on a typical on-chip rectangular inductor. / Singapore-MIT Alliance (SMA)

In this thesis the characteristics of pitch and roll damping coefficients of missile models are investigated by using Computational Fluid Dynamics (CFD) techniques. Experimental data of NACA0012 airfoil, Basic Finner (BF) and Modified Basic Finner (MBF) models are used for validation and verification studies. Numerical computations are performed from subsonic to supersonic flow regimes. Grid refinement and turbulence model selection studies are conducted before starting the dynamic motion simulations. Numerical method of dynamic motion simulation is validated with a 2D NACA0012 airfoil. After the validation of numerical method, forced-oscillation motion is given to the BF and MBF models. In order to get deeper understandings about the characteristics of dynamic pitching and rolling motions, parametric studies are performed. The amplitude and frequency of forced-oscillation motions are investigated one by one. The effects of angle of attacks are also investigated for both pitching and rolling motions. The results of CFD simulations are compared with experimental data obtained from different wind tunnel and free flight tests. It is seen from these comparisons that experimental and numerical results are in good agreement throughout the whole flow regime. In conclusion, the numerical method presented in this study is validated and can be used for the prediction of pitch and roll damping coefficient of any missile configurations.

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease affecting motor neurons in the spinal cord, brain stem and motor cortex, leading to paralysis, respiratory failure and death. In about 5% of ALS cases, the disease is associated with mutations in the CuZn-superoxide dismutase (hSOD1) gene. As a rule, ALS caused by hSOD1 mutations is inherited dominantly and the mutant hSOD1s cause ALS by the gain of a noxious property. The present study focused on two hSOD1 mutations with widely differing characters. In Scandinavia, ALS caused by the D90A mutation is inherited in a recessive pattern. Elsewhere, families with dominant inheritance have been found. The properties of D90A mutant hSOD1 are very similar to those of the wild-type protein. The G127insTGGG (G127X) mutation causes a 21 amino acid C-terminal truncation which probably results in an unstable protein. The aim of this thesis was to generate transgenic mice expressing D90A and G127X mutant hSOD1s and to compare these mice with each other and with mice expressing other mutant hSOD1s, in search of a common noxious property. The findings were also compared with the results from studies of human CNS tissue. The cause of the different inheritance patterns associated with D90A mutant hSOD1 was investigated by analyzing erythrocytes from heterozygous individuals from dominant and recessive pedigrees. There was no evidence that a putative protective factor in recessive pedigrees acts by down-regulating the synthesis of D90A mutant hSOD1. In cerebrospinal fluid, there was no difference in hSOD1 content between homozygous D90A patients, ALS patients without hSOD1 mutations and controls. hSOD1 cleaved at the N-terminal end was found in both controls and D90A patients, but the proportion was significantly larger in the latter group. This indicates a difference in degradation routes between mutant and wild-type hSOD1. Both D90A and G127X transgenic mice develop an ALS-like phenotype. Similar to humans, the levels of D90A protein were high. The levels of G127X hSOD1 were very low in the tissues but enriched in the CNS. Similarly, in an ALS patient heterozygous for G127X hSOD1, the levels of the mutant protein were overall very low, but highest in affected CNS areas. Despite the very different levels of mutant hSOD1, both D90A and G127X transgenic mice developed similar levels of detergent-resistant aggregates in the spinal cord when terminally ill. Surprisingly, mice overexpressing wild-type hSOD1 also developed detergent-resistant aggregates, although less and later. Most of the hSOD1 in the CNS of transgenic mice was inactive due to deficient copper charging or because of reduced affinity for the metal. The stabilizing intrasubunit disulfide bond of hSOD1 was partially or completely absent in the different hSOD1s. Both these alterations could increase the propensity of mutant hSOD1s to misfold and form aggregates. The results presented here suggest that the motor neuron degeneration caused by mutant hSOD1s may be attributable to long-term exposure to misfolded, aggregation-prone, disulfide-reduced hSOD1s and that the capacity to degrade such hSOD1s is lower in susceptible CNS areas compared with other tissues. The data also suggest that wild-type hSOD1 has the potential to participate in the pathogenesis of sporadic ALS.

The high amount of scientific and communications data produced by low earth orbiting satellites necessitates economical methods of communication with these satellites. A volumetric phased array for demonstrating horizon-to-horizon electronic tracking of the NASA satellite EO-1 was developed and demonstrated. As a part of this research, methods of optimizing the elemental antenna as well as the antenna on-board the satellite were investigated. Using these optimized antennas removes the variations in received signal strength that are due to the angularly dependent propagation loss exhibited by the communications link. An exhaustive study using genetic algorithms characterized two antenna architectures, and included optimizations for radiation pattern, bandwidth, impedance, and polarization. Eleven antennas were constructed and their measured characteristics were compared to those of the simulated antennas. Additional studies were conducted regarding the optimization of aperiodic arrays. A pattern-space representation of volumetric arrays was developed and used with a novel tracking algorithm for these arrays. This algorithm allows high-resolution direction finding using a small number of antennas while mitigating aliasing ambiguities. Finally, a method of efficiently applying multiple beam synthesis using the Fast Fourier Transform to aperiodic arrays was developed. This algorithm enables the operation of phased arrays combining the benefits of aperiodic element position with the efficiency of FFT multiple beam synthesis. Results of this research are presented along with the characteristics of the volumetric array used to track EO-1. Experimental data and the interpretations of that data are presented, and possible areas of future research are discussed.

Data centers are computing infrastructure facilities used by industries with large data processing needs and the rapid increase in power density of high performance computing equipment has caused many thermal issues in these facilities. Systems-level thermal management requires modeling and analysis of complex fluid flow and heat transfer processes across several decades of length scales. Conventional computational fluid dynamics and heat transfer techniques for such systems are severely limited as a design tool because their large model sizes render parameter sensitivity studies and optimization impractically slow. The traditional proper orthogonal decomposition (POD) methodology has been reformulated to construct physics-based models of turbulent flows and forced convection. Orthogonal complement POD subspaces were developed to parametrize inhomogeneous boundary conditions and greatly extend the use of the existing POD methodology beyond prototypical flows with fixed parameters. A flux matching procedure was devised to overcome the limitations of Galerkin projection methods for the Reynolds-averaged Navier-Stokes equations and greatly improve the computational efficiency of the approximate solutions. An implicit coupling procedure was developed to link the temperature and velocity fields and further extend the low-dimensional modeling methodology to conjugate forced convection heat transfer. The overall reduced-order modeling framework was able to reduce numerical models containing 105 degrees of freedom (DOF) down to less than 20 DOF, while still retaining greater that 90% accuracy over the domain. Rigorous a posteriori error bounds were formulated by using the POD subspace to partition the error contributions and dual residual methods were used to show that the flux matching procedure is a computationally superior approach for low-dimensional modeling of steady turbulent convection. To efficiently model large-scale systems, individual reduced-order models were coupled using flow network modeling as the component interconnection procedure. The development of handshaking procedures between low-dimensional component models lays the foundation to quickly analyze and optimize the modular systems encountered in electronics thermal management. This modularized approach can also serve as skeletal structure to allow the efficient integration of highly-specialized models across disciplines and significantly advance simulation-based design.

In this study Jarrow Turnbull (1995) Model, which is a reduced form approach for credit risk models, is employed to estimate the default intensity of US corporate bonds conditionally based on a fixed recovery rate. The estimations are performed with respect to the ratings of the bonds and the results were consistent with the ratings. US Treasury Bills are also used to since zero coupon default free prices, modeled by Svensson (1994) are necessary for pricing the default risky coupon bonds.

No description available.

In this thesis the path integral formalism is applied to the calculation of the dynamics of dissipative quantum systems. The time evolution of a system of bilinearly coupled bosonic modes is treated using the real-time path integral technique in coherent-state representation. This method is applied to a damped harmonic oscillator within the Caldeira-Leggett model. In order to get the stationary trajectories the corresponding Lagrangian function is diagonalized and then the path integrals are evaluated by means of the stationary-phase method. The time evolution of the reduced density matrix in the basis of coherent states is given in simple analytic form for weak system-bath coupling, i.e. the so-called rotating-wave terms can be evaluated exactly but the non-rotating-wave terms only in a perturbative manner. The validity range of the rotating-wave approximation is discussed from the viewpoint of spectral equations. In addition, it is shown that systems without initial system-bath correlations can exhibit initial jumps in the population dynamics even for rather weak dissipation. Only with initial correlations the classical trajectories for the system coordinate can be recovered. The path integral formalism in a combined phase-space and coherent-state representation is applied to the problem of curve-crossing dynamics. The system of interest is described by two coupled one-dimensional harmonic potential energy surfaces interacting with a heat bath. The mapping approach is used to rewrite the Lagrangian function of the electronic part of the system. Using the Feynman-Vernon influence-functional method the bath is eliminated whereas the non-Gaussian part of the path integral is treated using the perturbation theory in the small coordinate shift between potential energy surfaces. The vibrational and the population dynamics is considered in a lowest order of the perturbation. The dynamics of a Gaussian wave packet is analyzed along a one-dimensional reaction coordinate. Also the damping rate of coherence in the electronic part of the relevant system is evaluated within the ordinary and variational perturbation theory. The analytic expressions for the rate functions are obtained in the low and high temperature regimes.

In der vorliegenden Arbeit wird eine Prozedur zur Ordnungsreduktion von Finite Elemente Modellen mikromechanischer Struktur mit elektrostatischem Wirkprinzip entwickelt und analysiert. Hintergrund der Ordnungsreduktion ist eine Koordinatentransformation von lokalen Finite Elemente Koordinaten in globale Koordinaten. Die globalen Koordinaten des reduzierten Modells werden durch einige wenige Formfunktionen beschrieben. Damit wird das Makromodell nicht mehr durch lokale Knotenverschiebungen beschrieben, sondern durch globale Formfunktionen, welche die gesamte Deformation der Struktur beeinflussen. Es wird gezeigt, dass Eigenvektoren der linearisierten mechanischen Struktur einfache und effiziente Formfunktionen darstellen. Weiterhin kann diese Methode für bestimmte Nichtlinearitäten und für verschiedene in Mikrosystemen auftretende Lasten angewendet werden. Das Ergebnis sind Makromodelle, die über Klemmen in Systemsimulatoren eingebunden werden können, die Genauigkeiten einer Finite Elemente Analyse erreichen und für Systemsimulationen typische Laufzeitverhalten besitzen.