This paper presents techniques that actively exploit chaotic behavior to accomplish otherwise-impossible control tasks. The state space is mapped by numerical integration at different system parameter values and trajectory segments from several of these maps are automatically combined into a path between the desired system states. A fine-grained search and high computational accuracy are required to locate appropriate trajectory segments, piece them together and cause the system to follow this composite path. The sensitivity of a chaotic system's state-space topology to the parameters of its equations and of its trajectories to the initial conditions make this approach rewarding in spite of its computational demands.

Columns subjected to time varying axial load may exhibit dynamic instability due to parametric resonance. This type of instability is inherent in structures; it is not due to material or geometrical imperfections, and can occur even in perfectly constructed structures. This characteristic makes parametric resonance a very difficult to predict and therefore dangerous phenomenon. In this thesis the stability of a structural column under bounded noise axial load is studied by use of Lyapunov exponents. Bounded noise is especially useful as a loading because it may be used to represent both wide and narrow band processes, making the stability equations developed general enough to handle a wide variety of real world probabilistic loadings. The equation of motion of the first mode of vibration for this system is a second-order nonlinear stochastic ordinary differential equation. The nonlinearity makes the system exhibit bifurcating behaviour where stability shifts from the trivial solution to a non-zero mean stationary solution. The stability of the trivial and non-trivial solutions is important in obtaining a complete picture of the dynamical behaviour of the system. The effect that damping, the amplitude of noise, and the level of nonlinearity have on the stability of a structural column is studied using both analytical and numerical approaches. The largest Lyapunov exponent of the trivial solution is determined analytically by using time averaged versions of the original equation of motion. The validity of the analytical time averaged equation of motion is also verified with Monte Carlo simulations. Due to the mathematical complexity the largest Lyapunov exponent of the non-trivial stationary solutions is obtained using Monte Carlo simulation only.

Columns subjected to time varying axial load may exhibit dynamic instability due to parametric resonance. This type of instability is inherent in structures; it is not due to material or geometrical imperfections, and can occur even in perfectly constructed structures. This characteristic makes parametric resonance a very difficult to predict and therefore dangerous phenomenon. In this thesis the stability of a structural column under bounded noise axial load is studied by use of Lyapunov exponents. Bounded noise is especially useful as a loading because it may be used to represent both wide and narrow band processes, making the stability equations developed general enough to handle a wide variety of real world probabilistic loadings. The equation of motion of the first mode of vibration for this system is a second-order nonlinear stochastic ordinary differential equation. The nonlinearity makes the system exhibit bifurcating behaviour where stability shifts from the trivial solution to a non-zero mean stationary solution. The stability of the trivial and non-trivial solutions is important in obtaining a complete picture of the dynamical behaviour of the system. The effect that damping, the amplitude of noise, and the level of nonlinearity have on the stability of a structural column is studied using both analytical and numerical approaches. The largest Lyapunov exponent of the trivial solution is determined analytically by using time averaged versions of the original equation of motion. The validity of the analytical time averaged equation of motion is also verified with Monte Carlo simulations. Due to the mathematical complexity the largest Lyapunov exponent of the non-trivial stationary solutions is obtained using Monte Carlo simulation only.

In the first part of this dissertation we explore the consequences of high frequency operation of Josephson junction arrays. At high frequencies these systems are no longer well modeled by Kirchhoffs laws, and new dynamical equations are derived directly from Maxwells equations. From these equations we derive a reduced set of averaged equations which greatly simplify the analysis of high frequency arrays. The averaged equations allow us to examine experimental strategies for obtaining higher power outputs from arrays. These strategies rely on resonant architectures that place the junctions near antinodes of a desired standing wave mode of the fluctuating current. Simple, heuristic rules are derived for the proper placement of junctions. The second part of the dissertation is devoted to stochastic resonance. A new theory is proposed to explain both two-state and excitable stochastic resonance. Previous theories explaining the two types of stochastic resonance yield similar results while using different analytic strategies. A constrained asymmetric rate model is derived that in one limit produces the proper result for the two-state system, while in another limit models the excitable system. The result that the constrained asymmetric rate model gives in the excitable limit is off by a factor of two, and this discrepancy is examined. Furthermore, we study the consequences of adding a colored noise source to the classic two-state model of stochastic resonance. We will find that when both white and colored noise sources are present, stochastic resonance will occur as a function of colored noise strength only if the correlation time of the colored noise source is small enough. Two theories are proposed to explain this phenomenon and both are examined in detail.

This dissertation presents three inter-related studies. Chapter 2 presents a study of scaling of crack propagation in rubber sheets. Two different scaling laws for supersonic and subsonic cracks were discovered. Experiments and numerical simulations have been conducted to investigate subsonic and supersonic cracks. The experiments are performed at 85 °C to suppress strain-induced crystallites that complicate experiments at lower temperature. Calibration experiments were performed to obtain the parameters needed to compare with a theory including viscous dissipation. Both experiments and numerical simulations support supersonic cracks, and a transition from subsonic to supersonic is discovered in the plot of experimental crack speed curves versus extension ratio for different sized samples. Both experiments and simulations show two different scaling regimes: the speed of subsonic cracks scales with the elastic energy density while the speed of supersonic cracks scales with the extension ratio. Crack openings have qualitatively different shapes in the two scaling regimes. Chapter 3 describes a theory of oscillating cracks. Oscillating cracks are not seen very widely, but observed in rubber and gels. A theory has been proposed for the onset of oscillation in gels, but the oscillation of cracks in rubber has not been explained. This study provides a theory able to describe both rubber and gels and recover the experimental phase diagram for oscillating cracks in rubber. The main new idea is that the oscillations of cracks follow from basic features of fracture mechanics and are independent of details of the crack equation of motion. From the fact that oscillations exist, one can deduce some conditions on forms that equations of motion can take. A discrete model of hydraulic fracture is mentioned in Chapter 4. Hydraulic fracturing is a stimulation treatment wherein fluids are injected into reservoirs under high pressure to generate fractures in reservoirs. In this study, a lattice-based pseduo-3D model is developed to simulate hydraulic fracturing. This mode has been validated via a comparison with the KGD model. A series of pilot simulations was systematically tested for complex geometries under more realistic operation conditions, including flexible boundary conditions, randomness in elastic properties of shales and perforations. The simulation results confirm that perforation is likely to increase the complexity of fracture networks; the results also suggest that the interference between neighboring fractures is key to fracture network formation. / text

No description available.

The framework for the Hall magnetohydrodynamic (MHD) model for plasma physics is built up from kinetic theory and used to analytically solve problems of interest in the field. The Hall MHD model describes fast magnetic reconnection processes in space and laboratory plasmas. Specifically, the magnetic reconnection process at an X-type neutral point, where current sheets form and store enormous amounts of magnetic energy which is later released as magnetic storms when the sheets break up, is investigated. The phenomena of magnetic flux pile-up driving the merging of antiparallel magnetic fields at an ion stagnation-point flow in a thin current sheet, called the Parker problem, also receives rigorous mathematical analysis.

We created an analytical reduced-order model (macromodel) for an electrically actuated circular plate with an in-plane residual stress for applications in capacitive micromachined ultrasonic transducers (CMUTs). After establishing the equations governing the plate, we discretized the system by using a Galerkin approach. The distributed-parameter equations were then reduced to a finite system of ordinary-differential equations in time. We solved these equations for the equilibrium states due to a general electric potential and determined the natural frequencies of the axisymmetric modes for the stable deflected position. As expected, the fundamental natural frequency generally decreases as the electric forcing increases, reaching a value of zero at pull-in. However, strain-hardening effects can cause the frequencies to increase with voltage. The macromodel was validated by using data from experiments and simulations performed on silicon-based microelectromechanical systems (MEMS). For example, the pull-in voltages differed by about 1% from values produced by full 3-D MEMS simulations. The macromodel was then used to investigate the response of an electrostatically actuated clamped circular plate to a primary resonance excitation of its first axisymmetric mode. The method of multiple scales was used to derive a semi-analytical expression for the equilibrium amplitude of vibration. The plate was found to always transition from a hardening-type to a softening-type behavior as the DC voltage increases towards pull-in. Because the response of CMUTs is highly influenced by the boundary conditions, an updated reduced-order model was created to account for more realistic boundary conditions. The electrode was still considered to be infinitesimally thin, but the electrode was allowed to have general inner and outer radii. The updated reduced-order model was used to show how sensitive the pull-in voltage is with respect to the boundary conditions. The boundary parameters were extracted by matching the pull-in voltages from the macromodel to those from finite element method (FEM) simulations for CMUTs with varying outer and inner radii. The static behavior of the updated macromodel was validated because the pull-in voltages for the macromodel and FEM simulations were very close to each other and the extracted boundary parameters were physically realistic. A macromodel for CMUTs was then created that includes both the boundary effects and an electrode of finite thickness. Matching conditions ensured the continuity of displacements, slopes, forces, and moments from the composite to the non-composite regime of the CMUT. We attempted to validate this model with results from FEM simulations. In general, the center deflections from the macromodel fell below those from the FEM simulation, especially for relatively high residual stresses, but the first natural frequencies that accompany the deflections were very close to those from the FEM simulations. Furthermore, the forced vibration characteristics also compared well with the macromodel predictions for an experimental case in which the primary resonance curve bends to the right because the CMUT is a hardening-type system. The reduced-order model accounts for geometric nonlinear hardening, residual stresses, and boundary conditions related to the CMUT post, allows for general design variables, and is robust up to the pull-in instability. However, even more general boundary conditions need to be incorporated into the model for it to be a more effective design tool for capacitive micromachined ultrasonic transducers. / Ph. D.

No description available.

Linear vibration absorbers are a valuable tool used to suppress vibrations due to harmonic excitation in structural systems. Limited evaluation of the performance of nonlinear vibration absorbers for nonlinear structures exists in the current literature. The state of the art is extended in this work to vibration absorbers in their three major physical implementations: the mechanical vibration absorber, the inductive-resistive shunted piezoelectric vibration absorber, and the electronic vibration absorber (also denoted a positive position feedback controller). A single, consistent, physically similar model capable of examining the response of all three devices is developed. The performance of vibration absorbers attached to single-degree-of-freedom structures is next examined for performance, robustness, and stability. Perturbation techniques and numerical analysis combine to yield insight into the tuning of nonlinear vibration absorbers for both linear and nonlinear structures. The results both clarify and validate the existing literature on mechanical vibration absorbers. Several new results, including an analytical expression for the suppression region's location and bandwidth and requirements for its robust performance, are derived. Nonlinear multiple-degree-of-freedom structures are next evaluated. The theory of Nonlinear Normal Modes is extended to include consideration of modal damping, excitation, and small linear coupling, allowing estimation of vibration absorber performance. The dynamics of the N+1-degree-of-freedom system reduce to those of a two-degree-of-freedom system on a four-dimensional nonlinear modal manifold, thereby simplifying the analysis. Quantitative agreement is shown to require a higher order model which is recommended for future investigation. Finally, experimental investigation on both single and multi-degree-of-freedom systems is performed since few experiments on this topic are reported in the literature. The experimental results qualitatively verify the analytical models derived in this work. The dissertation concludes with a discussion of future work which remains to allow nonlinear vibration absorbers, in all three physical implementations, to enter the engineer's toolbox. / Ph. D.