<p>This master's thesis deals with the control design method called Non-linear Dynamic Inversion (NDI) and how it can be applied to Unmanned Aerial Vehicles (UAVs). In this thesis, simulations are conducted using a model for the unmanned aerial vehicle SHARC (Swedish Highly Advanced Research Configuration), which Saab AB is developing. </p><p>The idea with NDI is to cancel the non-linear dynamics and then the system can be controlled as a linear system. This design method needs much information about the system, or the output will not be as desired. Since it is impossible to know the exact mathematical model of a system, some kind of robust control theory is needed. In this thesis integral action is used. </p><p>A problem with NDI is that the mathematical model of a system is often very complex, which means that the controller also will be complex. Therefore, a controller that uses pure NDI is only discussed, and the simulations are instead based on approximations that use a cascaded NDI. Two such methods are investigated. One that uses much information from aerodata tables, and one that uses the derivatives of some measured outputs. Both methods generate satisfying results. The outputs from the second method are more oscillatory but the method is found to be more robust. If the signals are noisy, indications are that method one will be better.</p>

O trabalho trata da formulação e implementação numérica de um modelo de dano para o concreto. O objetivo é o de reproduzir laços de histerese observados experimentalmente em diagramas de tensão-deformação uniaxiais quando, a partir de um certo nível de solicitação, e dano correspondente, passam-se a realizar ciclos de descarregamento e recarregamento. O modelo propõe que os laços de histerese resultam da dissipação conjunta decorrente do dano evolutivo e do fenômeno de fricção interna nas faces das fissuras. O fenômeno de fricção interna está relacionado à suposição de que as fissuras não se abrem simplesmente por separação formando faces com superfícies lisas, mas tendem a combinar modos de separação e deslizamento segundo superfícies com uma certa rugosidade. O deslizamento entre as faces das fissuras dá origem a uma deformação por escorregamento, assumida como responsável por um comportamento plástico com encruamento cinemático não-linear. A abordagem termodinâmica considerada no sentido de contemplar a danificação e a fricção interna leva à dedução de que a tensão total num ponto do meio com dano resulta dividida em uma parcela dita tensão elasto-danificada e em uma outra denominada tensão de escorregamento, também dependente do nível de danificação da estrutura. Dois tipos básicos de aplicações são realizadas: as primeiras visam avaliar a influência das variáveis de estado e dos parâmetros do material na forma dos diagramas de tensão-deformação. As outras aplicações são dedicadas à utilização do modelo na análise do comportamento dinâmico de uma viga em concreto armado submetida à vibração forçada. Destaca-se o efeito dos laços de histerese sobre o amortecimento estrutural. Mostra-se que a utilização deste modelo de dano leva à geração de um amortecimento estrutural que com os modelos de dano usuais somente poderia ser obtido a partir da utilização do modelo de amortecimento de Rayleigh, por exemplo, na equação do movimento. / This work deals with numerical formulation and implementation of a damage model to concrete. The aim is to reproduce hysteresis loops observed experimentally in uniaxial stress-strain diagram when cycles of unloading and reloading are produced. The model proposes that the hysteresis loops are due to the coupling between the dissipation of damage process and the internal friction phenomenon between the faces of crack lips. The phenomenon of internal friction is related to the supposition that the crack doesn\'t open itself by separation mode creating plane lips, but there is a tendency to combine opening and sliding modes across irregular surfaces. The sliding between the crack surfaces generates a sliding strain, which is assumed as responsible by a plastic behaviour with non-linear kinematics hardening. The thermodynamic approach considering damage and internal friction leads to the deduction that the total stress at a point is composted by one component named elasto-damage stress and another named sliding stress, both dependent of the structure damage level. Two basic kinds of simulation are performed: the first one aims to evaluate the influence of the state variables and the material parameters over the stress-strain diagrams. The following simulations are dedicated to the use of the model in dynamic analysis of a reinforced concrete beam subjected to vibration. It is emphasized the effect of the structural damping from hysteresis loops. It is showed that the present damage model generates structural damping that could only be obtained with conventional damage model by using the Rayleigh damping approach in the equation of motion.

Cette thèse porte sur l'interaction entre un plasma de bord de tokamak et une perturbation magnétique résonante (RMP), utilisée principalement pour le contrôle de phénomènes de relaxations quasi-périodiques, présents dans un régime de confinement amélioré. Il permet notamment d'atteindre des conditions favorables aux réactions de fusion nucléaire. Il a été observé que la présence de perturbations magnétiques modifie la topologie magnétique au bord ce qui engendre une diminution de l'amplitude des relaxations, voire leur suppression. De précédents travaux ont étudié l'effet de perturbations magnétiques sur un plasma relaxant via des simulations numériques. Le modèle utilisé était dans un cas électrostatique, c'est à dire que la topologie magnétique n'évoluait pas dans le temps. Dans cette thèse, l'étude est faite dans un modèle de plasma de bord prenant en compte les fluctuations magnétiques via le code numérique EMEDGE3D. Ce code a été modifié pour pouvoir imposer une perturbation magnétique résonante. Des vérifications par des modèles réduits ont été menées sur la pénétration d'une perturbation magnétique ainsi que sur l'effet d'une vitesse cisaillée sur la pénétration. Ensuite, un RMP a été imposé dans un plasma non turbulent avec et sans vitesse cisaillée. Un phénomène d'écrantage, empêchant la pénétration d'une perturbation, a été identifié analytiquement et observé dans les simulations. Cette étude a été réitérée dans un plasma turbulent, et aussi en présence d'une barrière (vitesse cisaillée). Le plasma turbulent engendre une amplification du RMP, tandis que la barrière est affectée par la présence de cellules de convection fixes générées par la perturbation. / In this thesis, the interaction between tokamak edge plasma and resonant magnetic perturbation (RMPs) is studied. It is mainly used to mitigate quasi-periodic relaxations in enhanced confinement regime. This regime allows to obtain good conditions for nuclear fusion. Introduction of a RMP in a tokamak plasma has been observed to modified the magnetic topology at the edge and decrease the relaxation amplitude up to complete suppression. Previous works studied the RMP effect on a plasma with relaxations, via numerical simulations. The model used for that consider the electrostatic approximation, where the magnetic topology does not evolve in time. In this thesis, the study is done with an edge plasma model taking into account magnetic fluctuations via the numerical code EMEDGE3D. This code has been modified to include the resonant magnetic perturbation. Comparison with reduced models has been carried out on the RMP penetration and the effect of sheared velocity on it. Then a RMP has been induced in a stable plasma, with or without imposed sheared rotation. A condition on the sheared velocity has been identified to avoid the screening effect, that would prevent the RMP penetration, analytically and in numerical simulations. This works has been repeated in a turbulent plasma in presence or not of transport barrier (sheared velocity). The turbulent plasma generate an effective RMP amplification, while the transport barrier is affected by locked convective cells due to the RMP.

In this study, a method is developed for modelling and classifying behaviour of maritime vessels in a piracy situation. Prior knowledge is used to construct a probabilistic graphical model of maritime vessel behaviour. This model is a novel variant of a dynamic Bayesian network (DBN), that extends the switching linear dynamic system (SLDS) to accommodate contextual information. A generative model and a classifier model are developed. The purpose of the generative model is to generate simulated data by modelling the behaviour of fishing vessels, transport vessels and pirate vessels in a maritime piracy situation. The vessels move, interact and perform various activities on a predefined map. A novel methodology for evaluating and optimising the generative model is proposed. This methodology can easily be adapted to other applications. The model is evaluated by comparing simulation results with 2011 pirate attack reports. The classifier model classifies maritime vessels into predefined categories according to their behaviour. The classification is performed by inferring the class of a vessel as a fishing, transport or pirate vessel class. The classification method is evaluated by classifying the data generated by the generative model and comparing it to the true classes of the simulated vessels. / Thesis (PhD)--University of Pretoria, 2014. / tm2015 / Electrical, Electronic and Computer Engineering / PhD / Unrestricted

Cette thèse se propose d’étudier la discrimination des comportements par les quantités de réponses itérées. Ce champ de recherche s’inscrit dans celui, plus large, de la discrimination des comportements par les quantités. Nous avons tout d’abord cherché à explorer la façon dont les comportements peuvent être discriminés par des quantités différentes d’itérations de réponses et de quelle façon l’installation ou non de cette discrimination peut être mise en correspondance avec une évolution de l’entropie du débit des réponses itérées. Ensuite, nous avons exploré la dynamique des réponses itérées en fonction de la quantité des itérations requises. Après cela, nous avons cherché à savoir s’il était possible de mettre en correspondance l’existence de dynamiques de réponses propres à des quantités avec l’installation ou non d’une discrimination des comportements par des couples différents composés de ces mêmes quantités. Enfin, en vue d’explorer en quoi l’installation d’’une dynamique des réponses permettait l’installation d’une discrimination des comportements, nous avons perturbé, en modifiant la topographie des réponses requises, l’installation de cette dynamique de réponses, et en avons mesuré les effets sur l’installation de la discrimination de comportements par ces quantités de réponses itérées. Les résultats montrent que l’évolution de l’entropie des débits de réponses itérées est orientée en fonction de l’installation ou non d’une discrimination des comportements par des quantités différentes d’itérations de réponses requises : l’entropie diminue lorsque la discrimination par des quantités de réponses itérées est installée, et augmente dans le cas contraire. Ce résultat n’est vrai que pour les participants ayant atteint au moins une fois le critère de décision que nous nous étions fixé pour juger de l’installation de la discrimination. L’analyse de la dynamique des réponses itérées en fonction de chaque quantité d’itérations requise montre qu’il existe presque systématiquement une dynamique propre à chaque quantité d’itérations de réponses requise, pour tous les participants, cette dynamique de réponses étant également propre à la topographie de la réponse requise. La perturbation de l’installation de la dynamique des réponses a, en outre, un effet sur l’installation de la discrimination des comportements par les quantités d’itérations de réponses. Au regard de ces résultats, nous pouvons conclure qu’il existe une correspondance entre l’installation ou non de la discrimination par des quantités de réponses itérées et l’évolution de l’entropie des débits de ces réponses itérées. Néanmoins, tandis que nous pouvons conclure à la pertinence d’analyser la dynamique des réponses itérées pour en étudier la complexité, la mise en correspondance entre celle-ci et la façon dont chaque quantité d’itérations de réponses prend ou non le contrôle de comportements spécifiques n’est pas possible à ce stade, bien que des pistes vers de futures études aient été proposées, pour explorer plus avant cette relation / This dissertation focuses on discrimination of behaviors by iterated responses, which falls in the scope of field of discrimination by quantities. First, we investigate how discrimination by several couples of iterated responses quantities could be related with the evolution of instantaneous rates of iterated responses entropy. Then, iterated responses dynamic was analyzed for several iterated responses quantities, and response topographies. The third experiment investigates the correspondence between specific dynamics of responses exhibited in fixed-ratio schedules and discrimination by couples of quantities of iterated responses. At last, effects of the disruption of the installation of the dynamic of responses on discrimination by these quantities of iterated responses have been measured. Results show that discrimination by quantities of iterated responses is possible, and that specific dynamics of responses match specific quantities of iterated responses. Although correspondence between such dynamics and discrimination have not been demonstrated, effects of disruption of dynamic of responses installation have been observed suggesting that a link between dynamic of responses and discrimination exists.

Nas últimas três décadas os phase locked loops (PLLs) totalmente digitais têm recebido muita atenção devido, principalmente, às vantagens que eles oferecem em comparação aos PLLs analógicos. Essas vantagens incluem melhor desempenho, maior velocidade e confiabilidade, tamanho reduzido e menor custo. Os PLLs também são amplamente utilizados em sistemas de comunicações e em outras aplicações digitais. A presente dissertação é uma contribuição no campo dos PLLs digitais adaptativos e otimizados para a sua implementação em hardware. É feito uma análise de suas características dinâmicas e proposta uma nova estrutura de PLL digital capaz de melhorar a resposta da malha em termos de tempo de aquisição e largura de banda. A Malha Síncrona Digital \"Tanlock\" com Estimação de Frequência e Ganho Adaptativo para Convergência Rápida, como é chamada, foi desenvolvida a partir da malha digital \"Tanlock\", utilizando-se teoremas de ponto fixo e mapas contrativos para determinar as condições de ganho que garantam convergência rápida e melhor utilização da largura de banda. Resultados das simulações são comparados com os obtidos teoricamente para avaliar o desempenho da malha proposta. / In the last three decades, fully-digital Phase-Locked-Loops (PLLs) systems have received a lot of attention due to its advantages in comparison with analog PLLs. These advantages include improved transient response, reliability and also reduced size and cost. The PLLs are widely used in communications systems and many other digital applications. This dissertation is a contribution to the field of digital adaptive PLLs optimized to hardware implementation. Here, a new PLL structure is presented; the Frequency Sensing Adaptive TDTL is an improvement to the classic Time-Delay Tanlock structure, alowing fast convergence to the synchronous states, using fixed-point theorems and contractive maps to determine the gain conditions which ensure the rapid convergence and also providing wider bandwidth. The results of simulations are compared with those obtained theoretically in order to assess the loop performance.

Numerical assessment of soil structure interaction problems require heavy computational efforts because of the dynamic and iterative (nonlinear) nature of the problems. Furthermore, modeling soil-structure interaction may require

Nachrichtentechnische Systeme werden stets durch unvermeidbare zufällige Störungen beeinflußt. Neben anderen Komponenten sind davon besonders Oszillatoren betroffen. Die durch die Störungen verursachten zufälligen Schwankungen in der Oszillatorausgabe können als Amplituden- und Phasenabweichungen modelliert werden. Dabei zeigt sich, daß vor allem zufällige Phasenfluktuationen von Bedeutung sind. Zufällige Phasenfluktuationen können unter Verwendung stochastischer Prozesse zweiter Ordnung mit kurzem oder langem Gedächtnis modelliert werden. Inhalt der Dissertation ist die Herleitung eines Verfahrens zur Simulation zufälliger Phasenfluktuationen von Oszillatoren mit kurzem Gedächtnis unter Berücksichtigung von Datenblattangaben.

Automated analysis of temporal data is a task of utmost importance for intelligent machines. For example, ubiquitous computing systems need to understand the intention of humans from the stream of sensory information, and health-care monitoring systems can assist patients and doctors by providing automatically annotated daily health reports. We present a set of extensions of switching linear dynamic systems (SLDSs) which provide the ability to capture the higher-order temporal structures within data and to produce more accurate results for the tasks such as labeling and estimation of global variations within data. The presented models are formulated within a dynamic Bayesian network formulation along with the inference and learning methods thereof. First, segmental SLDSs (S-SLDSs) produce superior labeling results by capturing the descriptive duration patterns within each LDS segment. The encoded duration models describe data more descriptively and allow us to avoid the severe problem of over-segmented labels, which leads to superior accuracy. Second, parametric SLDSs (P-SLDSs) allows us to encode the temporal data with global variations. In particular, we have identified two types of global systematic variations : temporal and spatial variations. The P-SLDS model assumes that there is an underlying canonical model which is globally transformed in time and space by the two associated global parameters respectively. Third, we present hierarchical SLDSs (H-SLDSs), a generalization of standard SLDSs with hierarchic Markov chains. H-SLDSs are able to encode temporal data which exhibits hierarchic structure where the underlying low-level temporal patterns repeatedly appear among different higher-level contexts. The developed SLDS extensions have been applied to two real-world problems. The first problem is to automatically decode the dance messages of honey bee dances where the goal is to correctly segment the dance sequences into different regimes and parse the messages about the location of food sources embedded in the data. The second problem is to analyze wearable exercise data where we aim to provide an automatically generated exercise record at multiple temporal and semantic resolutions. It is demonstrated that the H-SLDS model with multiple layers can be learned from data, and can be successfully applied to interpret the exercise data at multiple granularities.

Nas últimas três décadas os phase locked loops (PLLs) totalmente digitais têm recebido muita atenção devido, principalmente, às vantagens que eles oferecem em comparação aos PLLs analógicos. Essas vantagens incluem melhor desempenho, maior velocidade e confiabilidade, tamanho reduzido e menor custo. Os PLLs também são amplamente utilizados em sistemas de comunicações e em outras aplicações digitais. A presente dissertação é uma contribuição no campo dos PLLs digitais adaptativos e otimizados para a sua implementação em hardware. É feito uma análise de suas características dinâmicas e proposta uma nova estrutura de PLL digital capaz de melhorar a resposta da malha em termos de tempo de aquisição e largura de banda. A Malha Síncrona Digital \"Tanlock\" com Estimação de Frequência e Ganho Adaptativo para Convergência Rápida, como é chamada, foi desenvolvida a partir da malha digital \"Tanlock\", utilizando-se teoremas de ponto fixo e mapas contrativos para determinar as condições de ganho que garantam convergência rápida e melhor utilização da largura de banda. Resultados das simulações são comparados com os obtidos teoricamente para avaliar o desempenho da malha proposta. / In the last three decades, fully-digital Phase-Locked-Loops (PLLs) systems have received a lot of attention due to its advantages in comparison with analog PLLs. These advantages include improved transient response, reliability and also reduced size and cost. The PLLs are widely used in communications systems and many other digital applications. This dissertation is a contribution to the field of digital adaptive PLLs optimized to hardware implementation. Here, a new PLL structure is presented; the Frequency Sensing Adaptive TDTL is an improvement to the classic Time-Delay Tanlock structure, alowing fast convergence to the synchronous states, using fixed-point theorems and contractive maps to determine the gain conditions which ensure the rapid convergence and also providing wider bandwidth. The results of simulations are compared with those obtained theoretically in order to assess the loop performance.