Many of the phenomena studied in the natural and social sciences are governed by processes which are discrete and nonlinear in nature, while the most highly developed and commonly used mathematical models are linear and continuous. There are significant differences between the discrete and the continuous, the nonlinear and the linear cases, and the development of mathematical models which exhibit the discrete, nonlinear properties occurring in nature and society is critical to future scientific progress. This thesis presents the basic theory of discrete dynamical systems and stability analysis and explores several applications of this theory to nonlinear systems which model interactions involving economic agents and biological populations. In particular we will explore the stability properties of equilibria associated with inter-species and intergenerational population dynamics in biology and market price and agent composition dynamics in economics.

This thesis is mainly concerned with the properties of some discrete nonlinear Schrödinger equations. These naturally arise in many different physical contexts as the limiting form of general dynamical lattice equations that incorporate nonlinearity and coupling. Interest is focused on theoretical models of coupled optical waveguides constructed from materials with a nonlinear index of refraction. In arrays of waveguides the overlap of the evanescent electric field of the modes in neighbouring waveguides provides a coupling and the nonlinearity of the material provides a mechanism to halt the discrete diffraction that otherwise would spread localized energy across the array. In particular, waveguide structures where also a nonlinear coupling is taken into account are studied. It is noted that the equation for the evolution of the complex amplitudes of the electric field along an array of waveguides also can be used to describe the dynamics of Bose-Einstein condensates trapped in a periodic optical potential. Possible excitations in arrays in both one and two dimensions are considered, with emphasis on the effects of the nonlinear coupling. Localized excitations are considered from the viewpoint of the theory of discrete breathers, or intrinsic localized modes, i.e., solutions of the dynamical equations that are periodic in time and have a spatial localization. The general theory of such solutions, that appear under very general circumstances in nonlinear lattice equations, is reviewed. In an array of waveguides this means that light can propagate along the array confined essentially to one or a few waveguides. In general a distinction is made between excitations that are centred on a waveguide, or site in the lattice, and excitations that are centred inbetween waveguides. Usually only the former give stable propagation. When the localized beam can be displaced to neighbouring waveguides the array can operate as an optical switch. With the inclusion of nonlinear coupling between the sites, as in the model derived in this thesis, the stability of the site-centred and bond-centred solutions can be exchanged. It is shown how this leads to the existence of highly localized mobile solutions that can propagate transversely in the one-dimensional array of waveguides. The inversion of stability of stationary solutions occurs also in the two-dimensional array, but in this setting it fails to give good mobility of localized excitations. The reason for this is also explained. In a two-dimensional lattice a discrete breather can have the form of a vortex. This means that the phase of the complex amplitude will vary on a contour around the excitation, such that the phase is increased by 2πS, where S is the topological charge, on the completion of one turn. Some ring-like vortex excitations are considered and in particular a stable vortex with S=2 is found. It is also noted that the effect of charge flipping, i.e., when the topological charge periodically changes between -S and S, is connected to the existence of quasiperiodic solutions. The nonlinear coupling of the waveguide model will also give rise to some more exotic and novel properties of localized solutions, e.g., discrete breathers with a nontrivial phase. When the linear coupling and the nonlinear coupling have opposite signs, there can be a decoupling in the lattice that allows for compact solutions. These localized excitations will have no decaying tail. Of interest is also the flexibility in controlling the transport of power across the array when it is excited with a nonlinear plane wave. It is shown how a change of the amplitude of a plane wave can affect the magnitude and direction of power flow in the array. Also the continuum limit of the one-dimensional discrete waveguide model is considered with an equation incorporating both nonlocal and nonlinear dispersion. In general continuum equations the balance between nonlinearity and dispersion can lead to the formation of localized travelling waves, or solitons. With nonlinear dispersion it is seen that these solitons can be nonanalytic and have discontinuous spatial derivatives. The emergence of short-wavelength instabilities due to the simultaneous presence of nonlocal and nonlinear dispersion is also explained.

This thesis deals with reachability and freeness problems for systems with linear dynamics, including hybrid systems and matrix semigroups. Hybrid systems are a type of dynamical system that exhibit both continuous and discrete dynamic behaviour. Thus they are particularly useful in modelling practical real world systems which can both flow (continuous behaviour) and jump (discrete behaviour). Decision questions for matrix semigroups have attracted a great deal of attention in both the Mathematics and Theoretical Computer Science communities. They can also be used to model applications with only discrete components. For a computational model, the reachability problem asks whether we can reach a target point starting from an initial point, which is a natural question both in theoretical study and for real-world applications. By studying this problem and its variations, we shall prove in a formal mathematical sense that many problems are intractable or even unsolvable. Thus we know when such a problem appears in other areas like Biology, Physics or Chemistry, either the problem itself needs to be simplified, or it should by studied by approximation. In this thesis we concentrate on a specific hybrid system model, called an HPCD, and its variations. The objective of studying this model is twofold: to obtain the most expressive system for which reachability is algorithmically solvable and to explore the simplest system for which it is impossible to solve. For the solvable sub-cases, we shall also study whether reachability is in some sense easy or hard by determining which complexity classes the problem belongs to, such as P, NP(-hard) and PSPACE(-hard). Some undecidable results for matrix semigroups are also shown, which both strengthen our knowledge of the structure of matrix semigroups, and lead to some undecidability results for other models.

This dissertation presents theoretical and experimental investigations into the dynamical behavior of Micro electromechanical systems (MEMS) resonators and their exploitation for filtering, sensing, and logic applications. The dissertation is divided into two parts: MEMS coupled structures and MEMS dynamic logic devices. First, a theoretical and experimental investigation is presented on both electrostatically and mechanically coupled resonator. Static and dynamic analysis is presented for weakly electrostatically coupled silicon microbeams and also for strongly mechanically coupled polyimide microbeams. The static analysis focuses on revealing pull-in characteristics, while the dynamic analysis focuses on the frequency response of the system and its exploitation for potential applications in filtering and amplification. Next, the phenomenon of mode localization is explored theoretically and experimentally on both electrostatically and mechanically weakly coupled resonators. Eigenvalue analysis is conducted and the dynamic response of the coupled system under different external perturbations is investigated. It is observed, that the exploitation of mode localization depends on the choice of the resonator to be under direct excitation, its stiffness to be perturbed, and which resonator is used to record the output results. These understandings will potentially help improve the performance of MEMS mode-localized sensors. Finally, three techniques to realize cascadable MEMS logic devices are presented. MEMS logic device vibrates at two steady states; a high on-resonance state (1) and a low off-resonance state (0). First, a MEMS logic device is presented capable of performing the AND/NAND logic gate and a tri-state logic gate using mixed-frequency excitation. This work is based on the concept of activation (1) and deactivation (0) of combination resonances due to the mixing of two or more input signals. Second, exploitation of subharmonic resonance under an AC only excitation to perform AND logic operation is presented. Finally, another MEMS logic device is presented working on the principal of activation (1) and deactivation (0) of second resonant mode of a clamped-clamped microbeam. This device is capable of performing OR, XOR and NOT gate. Experimental demonstration of the cascadability is shown for this case cascading OR and NOT gate to perform a logically complete NOR logic gate.

No description available.

No description available.

[EN] The dynamics of linear operators, namely linear dynamics, is mainly concerned with the behaviour of iterates of linear transformations. Hypercyclicity is the study of linear operators that possess a dense orbit. Although the first examples of hypercyclic operators are due to G. D. Birkhoff (in 1929), G. R. MacLane (in 1952) and S. Rolewicz (in 1969), we can date the birth of the linear dynamics in 1982 with the unpublished PhD thesis of C. Kitai. Since then, many mathematicians have contributed to the development of this flourishing new area of the analysis. Linear dynamics connects functional analysis and dynamics. As for the classical dynamical systems, one can study the dynamics of linear operators from a topological point of view. In this context, we state that an operator has the specification property (SP). Precisely, the aim of this PhD thesis is to study the specification property on linear dynamical systems. A continuous map on a compact metric space satisfies the specification property if one can approximate pieces of orbits by a single periodic orbits with a certain uniformity. This Doctoral dissertation is a compendium of articles on the specification property. It is structured in four parts preceded by a chapter which introduces the notation, definitions and the basic results that will be needed throughout the thesis. The shift operators on sequence spaces constitute one of the most important test ground for discrete linear dynamical systems. Due to its simple structure, every time you introduce a new property in linear dynamics it is common to check it on weighted shifts operators. It is for this reason that the first part of this research work is devoted to study the specification property for unilateral and bilateral backward shift operators on weighted l^p-spaces and the relationship with other dynamical properties. In Chapter 3 we extend the results on the SP to shift operators on separable sequence F-spaces. An F-space is a vector space that is endowed with an F-norm and that is complete under the induced metric. The notion of an F-norm has the advantage that one can largely argue as if one was working in a Banach space. One need to be aware of the fact that the positive homogeneity of a norm is no longer available. The spaces l^p with 0 < p < 1 are F-spaces. Chaotic dynamical systems have received a great deal of attention in recent years. An operator is chaotic if it has a dense set of periodic points. The specification property is an interesting and rather strong notion of chaos (in the topological sense). We also consider a qualitative strengthening of hypercyclicity namely frequent hypercyclicity. It was introduced by Bayart and Grivaux, motivated by Birkhoff's ergodic theorem. An operator is frequently hypercyclic if there is some element whose orbit meets every non-empty open set very often. In Chapter 4 the specification property is deeply studied for linear and continuous operators on separable F-spaces. In addition, we are interested in finding out its relation with other dynamical properties such as mixing, Devaney chaos and frequent hypercyclicity. The results that we have achieved have been accepted to be publish in Journal of Mathematical Analysis and Applications. Finally, in the last chapter of this dissertation, we examine the specification property for strongly continuous semigroups on Banach spaces, that is, for C_0-semigroups. They can viewed as the continuous-time analogue of the discrete-time case of iterates of a single operator; in other words, the parameter in the continuous case plays the role of the iterations in the discrete case. Now the translation semigroups substitute the shift operators as test classes. Once again, we study the relationship between the specification property and mixing, chaos and frequent hypercyclicity properties of a C_0-semigroup. / [ES] La dinámica de operadores lineales, o simplemente dinámica lineal, estudia las órbitas generadas por las iteraciones de una transformación lineal. La hiperciclicidad es el estudio de los operadores lineales que poseen una órbita densa. Si bien G. D. Birkhoff (en 1929), G. R. MacLane (en 1952) y S. Rolewicz (en 1969) obtuvieron ejemplos de operadores lineales hipercíclicos, podemos fijar el nacimiento de la dinámica lineal en 1982 con la tesis de C. Kitai. Desde entonces muchos matemáticos han contribuido al desarrollo de esta floreciente área del análisis. La dinámica lineal conecta el análisis funcional y la dinámica. Al igual que en sistemas dinámicos clásicos, podemos estudiar la dinámica de operadores lineales desde un punto de vista topológico. En este contexto, hablamos de que un operador tiene la propiedad de especificación (SP). Precisamente, al estudio de la propiedad de especificación en sistemas dinámicos lineales está dedicada la presente tesis doctoral. Una aplicación continua en un espacio métrico satisface la propiedad de especificación si para cualquier familia de puntos podemos aproximar, con una cierta uniformidad, partes de sus órbitas por una sola órbita de un punto periódico. La tesis es un compendio de artículos sobre la propiedad de especificación. Se estructura en cuatro partes precedidas de un capítulo dedicado a introducir la notación, definir los conceptos y enunciar los resultados de ámbito general que van a ser utilizados en el resto de la memoria. Los operadores "shift" (desplazamiento) constituyen una de las clases más importantes, como campo de pruebas, en sistemas dinámicos lineales discretos. Debido a su estructura simple, siempre que se introduce un nuevo concepto en dinámica lineal es habitual comprobarlo sobre shifts ponderados. Por este motivo, en la primera parte de esta memoria, se estudia la propiedad de especificación para operadores desplazamiento unilaterales y bilaterales en espacios l^p ponderados y la relación con otras propiedades dinámicas. En el capítulo 3 se generalizan los resultados sobre la propiedad SP a operadores desplazamiento en F-espacios separables de sucesiones. Un F-espacio es un espacio vectorial, dotado de una F-norma, que es completo con la métrica inducida. La noción de F-norma tiene la ventaja de que permite trabajar como en un espacio de Banach llevando cuidado con la homogeneidad de la norma que ahora no se cumple Los sistemas dinámicos caóticos han recibido gran atención en los últimos años. Un operador lineal es caótico si admite un conjunto denso de puntos periódicos. La propiedad de especificación es una noción de caos (en el sentido topológico) más potente que la debida a Devaney. Otra variante más fuerte que la hiperciclicidad es la hiperciclicidad frequente. Este concepto fue introducido por Bayart y Grivaux motivados por el teorema ergódico de Birkhoff. Un operador es frecuentemente hipercíclico si algún elemento tiene una órbita que corta muy a menudo a cada conjunto abierto no vacío. En el capítulo 4 de esta tesis se estudia con profundidad la propiedad de especificación para operadores lineales y continuos definidos en F-espacios separables. Los resultados que presentamos han sido aceptados para su publicación en J. Math. Anal. Appl. Finalmente, en la cuarta parte de este trabajo, se extiende la propiedad de especificación a semigrupos de operadores fuertemente continuos en espacios de Banach, esto es, C_0-semigrupos. Estos operadores pueden verse como la versión continua del caso discreto correspondiente a las iteraciones de un único operador. Ahora, la labor de los operadores desplazamiento en espacios de sucesiones como clases de prueba la desempeñan los semigrupos de traslación. Al igual que en capítulos anteriores, se estudia la relación de la propiedad SP para C_0-semigrupos con otras propiedades dinámicas. / [CAT] La dinàmica d'operadors lineals, o simplement dinàmica lineal, estudie les òrbites generades per les iteracions d'una transformació lineal. La hiperciclicitat es el estudi dels operadors lineal que posseeixen una òrbita densa. Si bé G. D. Birkhoff (en 1929), G. R. MacLane (en 1952) y S. Rolewicz (en 1969) van obtenir exemples d'operadors lineals hipercíclics, podem fixar el naixement de la dinàmica lineal en 1982 amb la tesi de C. Kitai [68]. Des de llavors molts matemàtics han contribuït al desenvolupament d'esta florent area de l'anàlisi. La dinàmica lineal connecta el anàlisi funcional y la dinàmica. Igual que en sistemes dinàmics clàssics, podem estudiar la dinàmica d'operadors lineals des d'un punt de vista topològic. En eixe context, parlem que un operador té la propietat d'especificació (SP). Precisament, al estudi de la propietat d'especificació en sistemes dinàmics lineals està dedicada la present tesi doctoral. Una aplicació continua en un espai mètric compleix la propietat d'especificació si per a qualsevol família de punts podem aproximar, amb certa uniformitat, parts de les seues òrbites per una sola òrbita d'un punt periòdic. La tesi es un compendi de articles sobre la propietat d'especificació. S'estructura en quatre parts precedides d'un capítol dedicat a introduir la notació, definir els conceptes i enunciar els resultats d'àmbit general que seran utilitzats en la resta de la memòria. Els operadors "shifts" (desplaçaments) constitueixen una de les classes més importants, com a camp de proves, en sistemes dinàmics lineals discrets. Degut a la seua estructura simple, sempre que es introdueix un nou concepte en dinàmica lineal es habitual comprovar-ho sobre shifts ponderats. Per esta raó, en la primera part d'esta memòria, s'estudia la propietat d'especificació per a operadors desplaçament unilaterals i bilaterals en espais l^p ponderats i la relació amb altres propietats dinàmiques. En el capítol 3 es generalitzen els resultats sobre la propietat SP a operadors desplaçament en F-espais separables de successions. Un F-espai es un espai vectorial, dotat d'una F-norma, que és complet amb la mètrica induida. La noció de F-norma té l'avantatge que permet treballar com en un espai de Banach anant en compte amb l'homogeneitat de la norma que ara no es compleix. Els espais l^p amb 0 < p < 1 són exemples de F-espais. Els sistemes dinàmics caòtics han rebut gran atenció en els últims anys. Un operador lineal és caòtic si admet un conjunt dens de punts periòdics. La propietat d'especificació és una noció de caos (en el sentit topològic) més potent que la deguda a Devaney. Una altra variant més forta que la hiperciclicitat és la hiperciclicitat freqüent. Aquest concepte va ser introduït per Bayart i Grivaux motivats per el teorema ergòdic de Birkhoff. Un operador és freqüentment hipercíclic si algun element té una òrbita que talle molt sovint a cada conjunt obert no vuit. En el capítol 4 d'esta tesi se estudie amb profunditat la propietat d'especificació per a operadors lineals i continus definits en F-espais separables. També s'incideix en la connexió de dita propietat amb altres propietats dinàmiques. Els resultats que presentem han estat acceptats per a la seva publicació en J. Math. Anal. Appl. Finalment, en la quarta part d'aquest treball, s'estén la propietat d'especificació a semigrups d'operadors fortament continus en espais de Banach, això és, C_0-semigrups. Aquests operadors poden veure's com la versió continua del cas discret corresponen a les iteracions d'un únic operador; en altres paraules, el paper de les iteracions en el cas discret ho assumeix el paràmetre en el cas continu. Ara, la labor del operadors desplaçament en espais de successions com classes de prova l'exerceixen els semigrups de translació. Igual que en capítols anteriors, s'estudia la relació de la propietat SP per a C0-semigrups amb altres propie / Bartoll Arnau, S. (2016). The specification property in linear dynamics [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/61633 / TESIS

<p>We study bifurcations of periodic travelling waves in granular dimer chains from the anti-continuum limit, when the mass ratio between the light and heavy beads tends to zero. We show that every limiting periodic wave is uniquely continued with respect to the mass ratio parameter and the periodic waves with the wavelength larger than a certain critical value are spectrally stable. Numerical computations are developed to study how this solution family is continued to the limit of equal mass ratio between the beads, where periodic travelling waves of granular monomer chains exist.</p> / Master of Science (MSc)

<p>We consider existence and stability of breather solutions to discrete nonlinear Schrodinger (dNLS) and discrete Klein-Gordon (dKG) equations near the anti-continuum limit, the limit of the zero coupling constant. For sufficiently small coupling, discrete breathers can be uniquely extended from the anti-continuum limit where they consist of periodic oscillations on excited sites separated by "holes" (sites at rest).</p> <p>In the anti-continuum limit, the dNLS equation linearized about its discrete breather has a spectrum consisting of the zero eigenvalue of finite multiplicity and purely imaginary eigenvalues of infinite multiplicities. Splitting of the zero eigenvalue into stable and unstable eigenvalues near the anti-continuum limit was examined in the literature earlier. The eigenvalues of infinite multiplicity split into bands of continuous spectrum, which, as observed in numerical experiments, may in turn produce internal modes, additional eigenvalues on the imaginary axis. Using resolvent analysis and perturbation methods, we prove that no internal modes bifurcate from the continuous spectrum of the dNLS equation with small coupling.</p> <p>Linear stability of small-amplitude discrete breathers in the weakly-coupled KG lattice was considered in a number of papers. Most of these papers, however, do not consider stability of discrete breathers which have "holes" in the anti-continuum limit. We use perturbation methods for Floquet multipliers and analysis of tail-to-tail interactions between excited sites to develop a general criterion on linear stability of multi-site breathers in the KG lattice near the anti-continuum limit. Our criterion is not restricted to small-amplitude oscillations and it allows discrete breathers to have "holes" in the anti-continuum limit.</p> / Doctor of Philosophy (PhD)

Dynamical maps for magnetic components are fundamental to studies of beam dynamics in accelerators. However, it is usually not possible to write down maps in closed form for anything other than simplified models of standard accelerator magnets. In the work presented here, the magnetic field is expressed in analytical form obtained from fitting Fourier series to a 3D numerical solution of Maxwell’s equations. Dynamical maps are computed for a particle moving through this field by applying a second order (with the paraxial approximation) explicit symplectic integrator. These techniques are used to study the beam dynamics in the first non-scaling FFAG ever built, EMMA, especially challenging regarding the validity of the paraxial approximation for the large excursion of particle trajectories. The EMMA lattice has four degrees of freedom (strength and transverse position of each of the two quadrupoles in each periodic cell). Dynamical maps, computed for a set of lattice configurations, may be efficiently used to predict the dynamics in any lattice configuration. We interpolate the coefficients of the generating function for the given configuration, ensuring the symplecticity of the solution. An optimisation routine uses this tool to look for a lattice defined by four constraints on the time of flight at different beam energies. This provides a way to determine the tuning of the lattice required to produce a desired variation of time of flight with energy, which is one of the key characteristics for beam acceleration in EMMA. These tools are then benchmarked against data from the recent EMMA commissioning.