<p>Important factors in manufacturing are quality and cost. Measuring machines play an important role for these fields. In order to meet higher demands on cost and accuracy, measuring machines can be constructed with weaker materials and increased mechanical flexibilities, and therefore there is a need to include the flexibilities in measuring machine models to obtain good performance. </p><p>The core theme in this thesis is modeling and idenfication of the physical parameters of drive mechanisms of a Brown&Sharpe Inc. Global A coordinate measuring machine. The approximation made is that the drive mechanisms can be described by a mass connected by springs, dampers and gear changes. It has been found that a one-spring model gives a reasonably good description of the studied CMM drive mechanism. The physical parameters of this model are identified using off-line algorithms. The algorithms are based on prediction error methods. For the off-line identification the MATLAB System Identification Toolbox and the bond graph representation is used. </p><p>The chosen model is then used for control. Traditional control and a Model-reference Adaptive System is derived and studied with the aim to increase the damping of CMM drive mechanisms. It is found that the adaptive system has very good disturbance rejection and can correct for drastic model errors. Another impact is that the damping of the studied drive mechanism can be increased with at least a factor of nine.</p>

Important factors in manufacturing are quality and cost. Measuring machines play an important role for these fields. In order to meet higher demands on cost and accuracy, measuring machines can be constructed with weaker materials and increased mechanical flexibilities, and therefore there is a need to include the flexibilities in measuring machine models to obtain good performance. The core theme in this thesis is modeling and idenfication of the physical parameters of drive mechanisms of a Brown&amp;Sharpe Inc. Global A coordinate measuring machine. The approximation made is that the drive mechanisms can be described by a mass connected by springs, dampers and gear changes. It has been found that a one-spring model gives a reasonably good description of the studied CMM drive mechanism. The physical parameters of this model are identified using off-line algorithms. The algorithms are based on prediction error methods. For the off-line identification the MATLAB System Identification Toolbox and the bond graph representation is used. The chosen model is then used for control. Traditional control and a Model-reference Adaptive System is derived and studied with the aim to increase the damping of CMM drive mechanisms. It is found that the adaptive system has very good disturbance rejection and can correct for drastic model errors. Another impact is that the damping of the studied drive mechanism can be increased with at least a factor of nine.

[pt] Este trabalho tem por objetivo a modelagem do comportamento dinâmico do motor de CC shunt, através da metodologia dos grafos de ligação. Primeiro será feito um levantamento de todos os parâmetros mecânicos, elétricos e magnéticos da máquina. O segundo passo será a obtenção de um grafo de ligação do motor e de suas respectivas equações de estado representativas. Através destas equações e dos parâmetros obtidos em laboratório será montado um modelo de simulação computacional pelo Matlab para verificação do comportamento da máquina desde a partida até o regime permanente, comprovando a veracidade do modelo. / [en] This work has the goal of modeling the dynamic behavior of a CC shunt motor, through the methodology of bond graphs. First all the mechanical, eletrical and magnetic parameters of the machine are measured. The second step will be the development of the motor bond graph and of its respective state equations. Using these equations and the measured parameters a Matlab simulation is set up to verify the transient behavior of yhe machine, checking the truthfulness of the model.

O problema de geração de estruturas de coalizão (CSG) envolve o particionamento do conjunto de agentes em todos os subconjuntos(ou, coalizões) possíveis. O que torna esse problema desafiador é o número de coalizões possíveis crescer exponencialmente a medida que novos agentes são inseridos, o número de coalizões é (2n − 1) onde n é o número de agentes. Entretanto, em muitas aplicações do mundo real, existem limitações inerentes nas coalizões possíveis: por exemplo, determinados agentes podem ser proibidos de estar na mesma coalizão, ou a estrutura de coalizão pode ser obrigada a conter coalizões do mesmo tamanho. Quando consideramos CSG restrito por grafos, onde a viabilidade de uma coalizão é restrita por um grafo de sinergia dos agentes, a complexidade computacional pode ser a mesma ou menor, dependendo do que se considera uma coalizão válida. Os grafos de sinergia são representações dos agentes como sendo os vértices e as suas relações são as arestas. Este trabalho é um estudo sobre a utilização de restrições envolvendo grafos como uma heurística sobre as coalizões para o problema enumeração de coalizão, de forma a considerar uma coalizão factível ou não de acordo com a densidade do subgrafo induzido pelos agentes. Os trabalhos atuais, que utilizam os grafos de restrição como heurística para reduzir a complexidade computacional, consideram uma coalizão válida somente se o subgrafo formado pelos agentes da coalizão é conexo. Verificou-se experimentalmente para grafos com a propriedade power law, comum em uma variedade de grafos reais, que restringir uma coalizão válida como sendo um subgrafo conexo pode não ser uma redução significativa. Entretanto a utilização de um subgrafo com restrições mais fortes, em particular uma clique garante uma redução exponencial do número de coalizões consideradas. Não existem teoremas que possam calcular qual a quantidade de subgrafos conexos ou mesmo o número de cliques em um grafo do tipo power law. No presente trabalho foi possível calcular experimentalmente para grafos power law com ate 17 vértices, sendo que também são apresentados resultados analíticos para grafos estrela (Kn−1,1 ). Os grafos estrela são uma aproximação aceitável, pois formam um hub, estrutura característica de grafos power law. Como trabalhos futuros podem ser citados: o mapeamento de domínios para os quais a restrição de clique seria adequada, além do desenvolvimento de um algoritmo que incorpore a restrição diretamente na contagem de coalizões validas. / The coalition structures generating problem (CSG) involves partitioning the set of agents in all possible subsets (or coalitions). What makes this problem challenging is the number of possible coalitions that grows exponentially as new agents are inserted. The number of coalitions is (2n − 1) where n is the number of agents. However, in many real-world applications, there are inherent limitations on possible coalitions: for example, some individuals may be prohibited from being in the same coalition or coalition structure may be required to contain coalitions of the same size. When we consider CSG restricted by graphs where the viability of a coalition is restricted by a synergy graph, the computational complexity can be maintained or eventually be smaller depending on what is considered a valid coalition. Synergy graphs are representations of the agents as being the vertices and their relationships are the edges. This work is a study on the use of restrictions involving graphs as a heuristic about coalitions for the problem coalition enumeration in order to consider a feasible coalition or not according to the density of the subgraph induced by the agents. Current works using the restriction graphs as heuristics to reduce the computational complexity, consider a coalition valid only if the subgraph formed by the agents of the coalition is connected. In this work it as experimentally verify for power law graphs, present in a variety of real graphs, that restricting availability coalition as a connected subgraph may in not prohibited a significant gain. However, they using a subgraphs with strong restrictions, in particular a clique, guarantees an exponential reduction in the number of considered coalition. There no are theorems calculate subgraphs or even the number of cliques on a type power law graph. In the present work it was possible to calculate values experimental for graphs of up to 17 vértices, being also presented analytics results for star graphs (Kn−1,1 ). Star graphs are an acceptable approximation, was they account for hubs, a characteristic structure of power law graphs. As future works can be cited the study of domains where the clique restriction is adequate as well as the development of an algorithm that incorporates the restriction for coalition counting.

[pt] Neste trabalho é desenvolvido um procedimento para a geração por inspeção das equações de estado e funções de transferência associadas a um grafo de ligação por meio de análises literal e numérica. O procedimento consiste na identificação de caminhos causais e a determinação de suas contribuições para as equações. É analisada a formulação matemática e discutida a implementação do procedimento em um código computacional. A aplicação do procedimento é ilustrada com exemplos, descreve-se detalhadamente as rotinas de entrada de dados, variáveis auxiliares, identificação dos caminhos e malhas causais, assim como a forma de apresentação dos resultados. / [en] In this work a procedure for the computer generation by inspection of the state equations and transfer functions, both in literal and numerical form, associated to a bond graph is discussed. The procedure consists of the identification of causal paths and the determination of its contributions for the equations. The mathematical background and the implementation of the procedure in a computational code are, also, presented. The application of the procedure is illustrated with examples; the routines for data entry, causal path and mesh identification and the used variables are described, as well as the form of presentation of the results.

A hydrostatic drive test stand was constructed to aid in the instruction of the characteristics of hydrostatic transmissions and to develop a time-varying load in a shaft to compare torque transducers. A mathematical model of the device was derived using the bond graph approach and was implemented using the Advanced Continuous Simulation Language (ACSL). The test stand was used to obtain experimental data to try to refine the model parameters. To facilitate the measurement of system flows, a microprocessor-based digital flow indicator was developed. Steady-state tests were performed at various constant pump swashplate angles; for the dynamic tests, the pump swashplate angle was cycled from zero degrees to 6.5 degrees. / Master of Science / incomplete_metadata

Les principaux objectifs de la thèse sont de proposer des modèles dynamiques pour décrire la circulation routière en adaptant des modèles de la mécanique des fluides et à partir de ces modèles de concevoir et mettre en œuvre des algorithmes de contrôle. La circulation routière est considérée comme un système complexe organisé dans une structure décentralisée à trois niveaux: au plus haut niveau il y a le réseau routier, décomposé, au niveau suivant, en objets routiers (tronçons et intersections), aussi décomposés, au dernier niveau, en éléments (segments) routiers.La thèse est organisée en six chapitres. Le premier chapitre présente les problèmes spécifiques aux réseaux de circulation routière, ainsi que des modèles et des algorithmes de contrôle trouvés dans la littérature. Le deuxième chapitre présente les principaux outils nécessaires pour décrire un modèle de trafic macroscopique et des façons pour représenter le flux de véhicules. Dans le troisième chapitre, des modèles dynamiques sont proposés pour différents segments et tronçons routiers. Le quatrième chapitre est consacré à la conception d’algorithmes pour le contrôle du trafic et l’optimisation de la circulation, y compris une analyse de robustesse pour évaluer la tolérance du système aux non-linéarités du modèle. Le cinquième chapitre présente une stratégie de contrôle pour le réseau routier, en utilisant la représentation par des modèles à compartiments et le concept de dynamique positive. Dans le dernier chapitre de la thèse, les contributions de l’auteur sont mises en évidence et quelques perspectives pour le développement ultérieur de la recherche sont présentées / The main objectives of this thesis are to propose dynamic models in order to describe road traffic by adapting models of fluid mechanics and based on these models to design and implement control algorithms. Road traffic is seen as a complex system organized in a decentralized structure. At the highest level of complexity there is the traffic network, considered as a part of a city’s road infrastructure. At the next level, the network is decomposed into traffic objects represented by road sections (streets) and intersections. At the last level of complexity there are the road segments that are parts of a road section.The thesis is organized into six chapters. The first chapter presents specific problems of road traffic networks, as well as control models and algorithms found in literature. The second chapter presents the main tools needed to describe a macroscopic traffic model, and some ways of representing the flow of vehicles. In the third chapter, dynamic models are proposed for different road segments and sections. The fourth chapter is devoted to designing algorithms for traffic control and optimization, including a robustness analysis to evaluate the system’s tolerance to model nonlinearities. The fifth chapter presents a control strategy for the road network, using the representation with compartmental models and the concept of positive dynamic. In the last chapter of the thesis, the author’s contributions are highlighted and some perspectives for further development of the research are presented

Les systèmes sur puce sont de plus en plus complexes : ils intègrent des parties numériques, desparties analogiques et des capteurs ou actionneurs. SystemC et son extension SystemC AMSpermettent aujourd’hui de modéliser à haut niveau d’abstraction de tels systèmes. Ces outilsconstituent de véritables atouts dans une optique d’étude de faisabilité, d’exploration architecturale etde vérification du fonctionnement global des systèmes complexes hétérogènes et multiphysiques. Eneffet, les durées de simulation deviennent trop importantes pour envisager les simulations globales àbas niveau d’abstraction. De plus, les simulations basées sur l’utilisation conjointe de différents outilsprovoquent des problèmes de synchronisation. Les modèles de bas niveau, une fois crées par lesspécialistes des différents domaines peuvent toutefois être abstraits afin de générer des modèles dehaut niveau simulables sous SystemC/SystemC AMS en des temps de simulation réduits. Une analysedes modèles de calcul et des styles de modélisation possibles est d’abord présentée afin d’établir unlien avec les durées de simulation, ceci pour proposer un style de modélisation en fonction du niveaud’abstraction souhaité et de l’ampleur de la simulation à effectuer. Dans le cas des circuits analogiqueslinéaires, une méthode permettant de générer automatiquement des modèles de haut niveaud’abstraction à partir de modèles de bas niveau a été proposée. Afin d’évaluer très tôt dans le flot deconception la consommation d’un système, un moyen d’enrichir les modèles de haut niveaupréalablement générés est présenté. L’attention a ensuite été portée sur la modélisation à haut niveaudes systèmes multiphysiques. Deux méthodes y sont discutées : la méthode consistant à utiliser lecircuit équivalent électrique puis la méthode basée sur les bond graphs. En particulier, nous proposonsune méthode permettant de générer un modèle équivalent au bond graph à partir d’un modèle de basniveau. Enfin, la modélisation d’un système éolien est étudiée afin d’illustrer les différents conceptsprésentés dans cette thèse. / Systems on chip are more and more complex as they now embed not only digital and analog parts, butalso sensors and actuators. SystemC and its extension SystemC AMS allow the high level modeling ofsuch systems. These tools are efficient for feasibility study, architectural exploration and globalverification of heterogeneous and multiphysics systems. At low level of abstraction, the simulationdurations are too important. Moreover, synchronization problems appear when cosimulations areperformed. It is possible to abstract the low level models that are developed by the specialists of thedifferent domains to create high level models that can be simulated faster using SystemC/SystemCAMS. The models of computation and the modeling styles have been studied. A relation is shownbetween the modeling style, the model size and the simulation speed. A method that generatesautomatically the high level model of an analog linear circuit from its low level representation isproposed. Then, it is shown how to include in the high level model some information allowing thepower consumption estimation. After that, the multiphysics systems modeling is studied. Twomethods are discussed: firstly, the one that uses the electrical equivalent circuit, then the one based onthe bond graph approach. It is shown how to generate a bond graph equivalent model from a low levelrepresentation. Finally, the modeling of a wind turbine system is discussed in order to illustrate thedifferent concepts presented in this thesis.

This thesis gives different views on the modeling and simulation of physical systems, especially together with embedded systems, forming mechatronic systems. The main considered application domain is automotive. One motivation behind the work is to find suitable representations of physical systems to be used in an architectural description language for automotive embedded systems, EAST-ADL2, which is implemented as a UML2 profile, and uses concepts from both UML and SysML. As a part of the thesis, several languages and tools are investigated, including bond graphs, MATLAB/Simulink, Ptolemy II, Modelica, MATLAB/Simscape and SysML. For SysML, the modeling of continuous-time systems and how it relates to MATLAB/Simulink and Modelica is evaluated. A case study of an electric power assisted steering is modeled to show the differences, the similarities and the usage of the above mentioned languages and tools. To be able to classify the tools and languages, five realization levels were developed: Physical modeling models Constraint models Continuous causal models Discretized models Discretized models with solver and platform implementation By using these realization levels, models, tools and modeling languages can be classified, and transformations between them can be set up and analyzed. As a result, a method to describe the simulation behavior of a MATLAB/Simulink model has been developed using SysML activity diagrams as an approach to achieve integrated system models. Another result is an evaluation of the parametric diagrams of SysML for continuous-time modeling, which shows that they do not enable “physical modeling”, i.e. modeling the topology of the system and getting the underlying equations out of this topology. By including physical ports and physical connectors to SysML internal block diagrams, this could be solved. The comparison also shows many similarities between the languages. The results led to a more detailed investigation on conjugate variables, such as force and velocity, and electric current and voltage, and how these are treated in various languages. The thesis also includes two industrial case studies: one of a twin-screw compressor, and one of a simulation environment for automotive fuel-cell systems. Conclusions are drawn from these models, referring to the realization levels. / QC 20100810

Les principaux objectifs de la thèse sont de proposer des modèles dynamiques pour décrire la circulation routière en adaptant des modèles de la mécanique des fluides et à partir de ces modèles de concevoir et mettre en œuvre des algorithmes de contrôle. La circulation routière est considérée comme un système complexe organisé dans une structure décentralisée à trois niveaux: au plus haut niveau il y a le réseau routier, décomposé, au niveau suivant, en objets routiers (tronçons et intersections), aussi décomposés, au dernier niveau, en éléments (segments) routiers.La thèse est organisée en six chapitres. Le premier chapitre présente les problèmes spécifiques aux réseaux de circulation routière, ainsi que des modèles et des algorithmes de contrôle trouvés dans la littérature. Le deuxième chapitre présente les principaux outils nécessaires pour décrire un modèle de trafic macroscopique et des façons pour représenter le flux de véhicules. Dans le troisième chapitre, des modèles dynamiques sont proposés pour différents segments et tronçons routiers. Le quatrième chapitre est consacré à la conception d'algorithmes pour le contrôle du trafic et l'optimisation de la circulation, y compris une analyse de robustesse pour évaluer la tolérance du système aux non-linéarités du modèle. Le cinquième chapitre présente une stratégie de contrôle pour le réseau routier, en utilisant la représentation par des modèles à compartiments et le concept de dynamique positive. Dans le dernier chapitre de la thèse, les contributions de l'auteur sont mises en évidence et quelques perspectives pour le développement ultérieur de la recherche sont présentées