Дисертації з теми "Real-time model order reduction"

Numerical simulation, the use of computers to run a program which implements a mathematical model for a physical system, is an important part of today technological world. It is required in many scientific and engineering fields to study the behaviour of systems whose mathematical models are too complex to provide analytical solutions and it makes virtual evaluation of systems responses possible (virtual twins). This drastically reduces the number of experimental tests for accurate designs of the real system that the numerical model represents. However these virtual twins, based on classical methods which make use of a rich representations of the system (ex. finite element method), rarely allows real-time feedback, even when considering high performance computing, operating on powerful platforms. In these circumstances, the real-time performance required in some applications are compromised. Indeed the virtual twins are static, that is, they are used in the design of complex systems and their components, but they are not expected to accommodate or assimilate data so as to define dynamic data-driven application systems. Moreover significant deviations between the observed response and the one predicted by the model are usually noticed due to inaccuracy in the employed models, in the determination of the model parameters or in their time evolution. In this thesis we propose different methods to solve these handicaps in order to perform real-time monitoring and control. In the first part Model Order Reduction (MOR) techniques are used to accommodate real-time constraints; they compute a good approximation of the solution by simplifying the solution procedure instead of the model. The accuracy of the predicted solution is not compromised and efficient simulations can be performed (digital twins). In the second part Data-Driven modelling are employed to fill the gap between the parametric solution computed by using non-intrusive MOR techniques and the measured fields, in order to make dynamic data-driven application systems, DDDAS, possible (Hybrid Twins).
La simulación numérica, el uso de ordenadores para ejecutar un programa que implementa un modelo matemático de un sistema físico, es una parte importante del mundo tecnológico actual. En muchos campos de la ciencia y la ingeniería es necesario estudiar el comportamiento de sistemas cuyos modelos matemáticos son demasiado complejos para proporcionar soluciones analíticas, haciendo posible la evaluación virtual de las respuestas de los sistemas (gemelos virtuales). Esto reduce drásticamente el número de pruebas experimentales para los diseños precisos del sistema real que el modelo numérico representa. Sin embargo, estos gemelos virtuales, basados en métodos clásicos que hacen uso de una rica representación del sistema (por ejemplo, el método de elementos finitos), rara vez permiten la retroalimentación en tiempo real, incluso cuando se considera la computación en plataformas de alto rendimiento. En estas circunstancias, el rendimiento en tiempo real requerido en algunas aplicaciones se ve comprometido. En efecto, los gemelos virtuales son estáticos, es decir, se utilizan en el diseño de sistemas complejos y sus componentes, pero no se espera que acomoden o asimilen los datos para definir sistemas de aplicación dinámicos basados en datos. Además, se suelen apreciar desviaciones significativas entre la respuesta observada y la predicha por el modelo, debido a inexactitudes en los modelos empleados, en la determinación de los parámetros del modelo o en su evolución temporal. En esta tesis se proponen diferentes métodos para resolver estas limitaciones con el fin de realizar un seguimiento y un control en tiempo real. En la primera parte se utilizan técnicas de Reducción de Modelos para satisfacer las restricciones en tiempo real; estas técnicas calculan una buena aproximación de la solución simplificando el procedimiento de resolución en lugar del modelo. La precisión de la solución no se ve comprometida y se pueden realizar simulaciones efficientes (gemelos digitales). En la segunda parte se emplea la modelización basada en datos para llenar el vacío entre la solución paramétrica, calculada utilizando técnicas de reducción de modelos no intrusivas, y los campos medidos, con el fin de hacer posibles los sistemas de aplicación dinámicos basados en datos (gemelos híbridos).
La simulation numérique, c'est-à-dire l'utilisation des ordinateurs pour exécuter un programme qui met en oeuvre un modèle mathématique d'un système physique, est une partie importante du monde technologique actuel. Elle est nécessaire dans de nombreux domaines scientifiques et techniques pour étudier le comportement de systèmes dont les modèles mathématiques sont trop complexes pour fournir des solutions analytiques et elle rend possible l'évaluation virtuelle des réponses des systèmes (jumeaux virtuels). Cela réduit considérablement le nombre de tests expérimentaux nécessaires à la conception précise du système réel que le modèle numérique représente. Cependant, ces jumeaux virtuels, basés sur des méthodes classiques qui utilisent une représentation fine du système (ex. méthode des éléments finis), permettent rarement une rétroaction en temps réel, même dans un contexte de calcul haute performance, fonctionnant sur des plates-formes puissantes. Dans ces circonstances, les performances en temps réel requises dans certaines applications sont compromises. En effet, les jumeaux virtuels sont statiques, c'est-à-dire qu'ils sont utilisés dans la conception de systèmes complexes et de leurs composants, mais on ne s'attend pas à ce qu'ils prennent en compte ou assimilent des données afin de définir des systèmes d'application dynamiques pilotés par les données. De plus, des écarts significatifs entre la réponse observée et celle prévue par le modèle sont généralement constatés en raison de l'imprécision des modèles employés, de la détermination des paramètres du modèle ou de leur évolution dans le temps. Dans cette thèse, nous proposons di érentes méthodes pour résoudre ces handicaps afin d'effectuer une surveillance et un contrôle en temps réel. Dans la première partie, les techniques de Réduction de Modèles sont utilisées pour tenir compte des contraintes en temps réel ; elles calculent une bonne approximation de la solution en simplifiant la procédure de résolution plutôt que le modèle. La précision de la solution n'est pas compromise et des simulations e caces peuvent être réalisées (jumeaux numériquex). Dans la deuxième partie, la modélisation pilotée par les données est utilisée pour combler l'écart entre la solution paramétrique calculée, en utilisant des techniques de réduction de modèles non intrusives, et les champs mesurés, afin de rendre possibles des systèmes d'application dynamiques basés sur les données (jumeaux hybrides).

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2018.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 177-183).
Biomolecular systems often involve reactions that take place on different time-scales, giving rise to 'slow' and 'fast' system variables. This property is widely used in the analysis of systems to obtain dynamical models with reduced dimensions. In deterministic systems, methods to obtain such reduced-order models are well defined by the singular perturbation or averaging techniques. However, model reduction of stochastic systems remains an ongoing area of research. In particular, existing model reduction methods for stochastic models of biomolecular systems lack rigorous error quantifications between the full and reduced dynamics. Furthermore, they only provide approximations for the slow variable dynamics, making the application of such methods to biomolecular systems difficult since the variables of interest are typically mixed (i.e., they encompass both fast and slow variables). In this thesis, we consider biomolecular systems modeled using the chemical Langevin equation (CLE) and the Linear Noise Approximation (LNA). Specifically, we consider biomolecular systems with linear propensity functions modeled by the CLE and systems with arbitrary propensity functions modeled by the LNA. For these systems, we obtain reduced-order models that approximate both the slow and fast variables under time-scale separation conditions. In particular, with suitable assumptions, we prove that the moments of the reduced-order models converge to those of the full systems as the time-scale separation becomes large. Our results further provide a rigorous justification for the accuracy of the stochastic total quasi-steady state approximation (tQSSA). We then consider two applications of these reduced-order models. In the first application, we analyze the trade-offs between modularity and signal noise in biomolecular networks. In the second application, we consider the application of the reduced-order LNA developed in this work to obtain reduced-order stochastic models for gene-regulatory networks.
by Narmada Kumari Herath.
Ph. D.

Engineering problems are very often characterised by a large ratio between the scale of the structure and the scale at which the phenomena of interest need to be described. In fracture mechanics, the initiation and propagation of cracks is the result of localised microscopic phenomena. This local nature of fracture leads to large numerical models. Projection-based reduced order modelling is an increasingly popular technique for the fast solution of parametrised problems. However, traditional model order reduction methods are unable to reliably deal with either the initiation or the propagation of a crack or a local zone with high damage concentration. In this thesis, we look at the general problem of applying model order reduction to fracture/ damage mechanics, in the pursuit of rationalising the computational time involved in these kind of simulations. The �rst contribution of this thesis is the development of a reduced-order modelling for computational homogenisation, which is a general multiscale method used to take microscopic data into account when deriving an engineeringscale model. A speci�c strategy is used to reduce the cost of solving the representative element volume (RVE) boundary value problem traditionally formulated in this method. The second contribution was made by developing a partitioned reduced-order procedure for the case of parametrised nonlinear material deformations involving a local lack of correlation, which typically happens with fracture. The method allows to reduce the regions undergoing little non-linearities whilst computational work can be concentrated on regions of high non-linearity.

The Best MPhil Thesis in the Faculties of Dentistry, Engineering, Medicine and Science (University of Hong Kong), Li Ka Shing Prize,2009-2010
published_or_final_version
Electrical and Electronic Engineering
Master
Master of Philosophy

L'objet de ce projet de recherche est de prédire la durée de vie d'éléments mécaniques qui sont soumis à des phénomènes de fatigue cyclique. L'idée est de développer un schéma numérique novateur pour prédire la rupture de structures sous de tels chargements. Le modèle est basé sur la mécanique des milieux continus qui introduit des variables internes pour décrire l'évolution de l'endommagement. Le défi repose dans le traitement des cycles de chargement pour la prédiction de la durée de vie, particulièrement pour la prédiction de la durée de vie résiduelle de structures existantes. Les approches traditionnelles de l'analyse de la fatigue sont basées sur des méthodes phénoménologiques utilisant des relations empiriques. De telles méthodes considèrent des approximations simplificatrices et sont incapables de prendre en compte aisément des géométries ou des charges complexes associées à des problèmes d'ingénierie réels. Une approche basée sur la description de l'évolution thermodynamique d'un milieu continu est donc utilisée pour modéliser le comportement en fatigue. Cela permet de considérer efficacement des problèmes d'ingénierie complexe et la détérioration des propriétés du matériau due à la fatigue peut être quantifiée à l'aide de variables internes. Cependant, cette approche peut être numériquement coûteuse et, par conséquent, des approches numériques sophistiquées doivent être utilisées.La stratégie numérique sur laquelle ce projet est basé est singulière par rapport aux schémas incrémentaux en temps usuellement utilisés pour résoudre des problèmes élasto-(visco)plastique avec endommagement dans le cadre de la mécanique des milieux continus. Cette stratégie numérique appelée méthode LATIN (Large Time Increment method) est une méthode non-incrémentale qui recherche la solution de manière itérative sur l'ensemble du domaine spacio-temporel. Une importante innovation de la méthode LATIN est d'incorporer une stratégie de réduction de modèle adaptative pour réduire de manière très importante le coût numérique. La Décomposition Propre Généralisée (PGD) est une stratégie de réduction de modèle a priori qui sépare les quantités d'intérêt spacio-temporelles en deux composantes indépendantes, l'une dépendant du temps, l'autre de l'espace, et estime itérativement les approximations de ces deux composantes. L'utilisation de l'approche LATIN-PGD a montré son efficacité depuis des années pour résoudre des problèmes élasto-(visco)plastiques. La première partie de ce projet vise à étendre cette approche aux modèles incorporant de l'endommagement.Bien que l'utilisation de la PGD réduise les coûts numériques, le gain n'est pas suffisant pour permettre de résoudre des problèmes considérant un grand nombre de cycles de chargement, le temps de calcul peut être très conséquent, rendant les simulations de problèmes de fatigue intraitables même en utilisant les techniques LATIN-PGD. Cette limite peut être dépassée en introduisant une approche multi-échelle en temps, qui prend en compte l'évolution rapide des quantités d'intérêt lors d'un cycle et leur évolution lente au cours de l'ensemble des cycles. Une description type « éléments finis » en temps est proposée, où l'ensemble du domaine temporel est discrétisé en éléments temporels, et seulement les cycles nodaux, qui forment les limites des éléments, sont calculés en utilisant la technique LATIN-PGD. Puis, des fonctions de forme classiques sont utilisées pour interpoler les quantités d'intérêt à l'intérieur des éléments temporels. Cette stratégie LATIN-PGD à deux échelles permet de réduire le coût numérique de manière significative, et peut être utilisée pour simuler l'évolution de l'endommagement dans une structure soumise à un chargement de fatigue comportant un très grand nombre de cycles
The motivation of the research project is to predict the life time of mechanical components that are subjected to cyclic fatigue phenomena. The idea herein is to develop an innovative numerical scheme to predict failure of structures under such loading. The model is based on classical continuum damage mechanics introducing internal variables which describe the damage evolution. The challenge lies in the treatment of large number of load cycles for the life time prediction, particularly the residual life time for existing structures.Traditional approaches for fatigue analysis are based on phenomenological methods and deal with the usage of empirical relations. Such methods consider simplistic approximations and are unable to take into account complex geometries, and complicated loadings which occur in real-life engineering problems. A thermodynamically consistent continuum-based approach is therefore used for modelling the fatigue behaviour. This allows to consider complicated geometries and loads quite efficiently and the deterioration of the material properties due to fatigue can be quantified using internal variables. However, this approach can be computationally expensive and hence sophisticated numerical frameworks should be used.The numerical strategy used in this project is different when compared to regular time incremental schemes used for solving elasto-(visco)plastic-damage problems in continuum framework. This numerical strategy is called Large Time Increment (LATIN) method, which is a non-incremental method and builds the solution iteratively for the complete space-time domain. An important feature of the LATIN method is to incorporate an on-the-fly model reduction strategy to reduce drastically the numerical cost. Proper generalised decomposition (PGD), being a priori a model reduction strategy, separates the quantities of interest with respect to space and time, and computes iteratively the spatial and temporal approximations. LATIN-PGD framework has been effectively used over the years to solve elasto-(visco)plastic problems. Herein, the first effort is to solve continuum damage problems using LATIN-PGD techniques. Although, usage of PGD reduces the numerical cost, the benefit is not enough to solve problems involving large number of load cycles and computational time can be severely high, making simulations of fatigue problems infeasible. This can be overcome by using a multi-time scale approach, that takes into account the rapid evolution of the quantities of interest within a load cycle and their slow evolution along the load cycles. A finite element like description with respect to time is proposed, where the whole time domain is discretised into time elements, and only the nodal cycles, which form the boundary of the time elements, are calculated using LATIN-PGD technique. Thereby, classical shape functions are used to interpolate within the time element. This two-scale LATIN-PGD strategy enables the reduction of the computational cost remarkably, and can be used to simulate damage evolution in a structure under fatigue loading for a very large number of cycles

El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado.
The present research work has as objective the application of Six Sigma DMAIC methodology in the production’s processes, the results will be manifested in increasing the efficiency of the production system and in reducing the delay in order delivery. In a metal mechanic company dedicated to the manufacture of electrical boards which focus is the terraced boards there was presented a fulfillment rate failure to deliver on time 46%. On the other hand, the delivery delay is generated because the current productivity does not supply what is required by the customer, therefore a minimum 394 units per month is needed to meet the requirements but currently only produce 226 units per month. For solve this problem it was proposed that Production model that merges the painting and baking areas and that generates an impact on the entire painting operation.

Cette thèse traite de modélisation des systèmes complexes. Dans ce cadre, l'approche est basée sur les Modèles Structurés en Second Ordre (MSSO). Afin d'utiliser cette classe de modèles, les propriétés telles que l'atteignabilité, l'observabilité et les grammiens, bien connues pour les réalisations d'états, sont étendues aux MSSO.Lors de la co-simulation d'un système, des éléments de natures différentes (physiques et logicielles) sont intégrés et la simulation est effectuée en temps réel. Or, les modèles d'ordre élevés sont couteux en temps de calcul, ce qui rend difficile ce type de simulation. Ainsi, des méthodes de réduction de modèle sont explorées. En particulier, de nouvelles méthodes, permettant de préserver la structure des modèles avec une bonne erreur d'approximation sont présentées.Ces développements sont appliqués à la co-simulation de modèles véhicules sous forme de MSSO. Le modèle créé est un modèle par blocs, complexe et non-linéaire. Afin d'appliquer les méthodes de réduction de modèle il est nécessaire de le linéariser. La structure par blocs permet de linéariser l'ensemble du modèle ou de ne linéariser que certaines sous parties du modèle.Ensuite, l'identification des paramètres est effectuée pour chaque sous-systèmes du véhicule. Une méthode d'interconnexion est ensuite proposée pour créer une représentation monobloc du modèle afin de réduire ce dernier. Au final, des essais en co-simulation de la partie arrière du véhicule sous forme de modèle interconnectée avec la partie avant du véhicule physiquement présente sur un banc de test, valide notre approche pour effectuer de la co-simulation temps réel avec matériel.x

Cette thèse a pour but d’évaluer l'utilisation des méthodes de réduction de modèles fondées sur des approches parcimonieuses pour atteindre des performances en temps réel dans la cadre de la chirurgie computationnelle. Elle se concentre notamment sur l’intégration de la simulation biophysique dans des modèles personnalisés de tissus et d'organes afin d'augmenter les images médicales et ainsi éclairer le clinicien dans sa prise de décision. Dans ce contexte, trois enjeux fondamentaux sont mis en évidence. Le premier réside dans l'intégration de la paramétrisation de la forme au sein du modèle réduit afin de représenter fidèlement l'anatomie du patient. Une approche non intrusive reposant sur un échantillonnage parcimonieux de l'espace des caractéristiques anatomiques est introduite et validée. Ensuite, nous abordons le problème de la complétion des données et de la reconstruction des images à partir de données partielles ou incomplètes via des à priori physiques. Nous explorons le potentiel de la solution proposée dans le cadre du recalage d’images pour la réalité augmentée en laparoscopie. Des performances proches du temps réel sont obtenues grâce à une nouvelle approche d'hyper-réduction fondée sur une technique de représentation parcimonieuse. Enfin, le troisième défi concerne la propagation des incertitudes dans le cadre de systèmes biophysiques. Il est démontré que les approches de réduction de modèles traditionnelles ne réussissent pas toujours à produire une représentation de faible rang, et ce, en particulier dans le cas de la simulation électrochirurgicale. Une alternative est alors proposée via la métamodélisation. Pour ce faire, nous étendons avec succès l'utilisation de méthodes de régression parcimonieuses aux cas des systèmes à paramètres stochastiques
This thesis investigates the use of model order reduction methods based on sparsity-related techniques for the development of real-time biophysical modeling. In particular, it focuses on the embedding of interactive biophysical simulation into patient-specific models of tissues and organs to enhance medical images and assist the clinician in the process of informed decision making. In this context, three fundamental bottlenecks arise. The first lies in the embedding of the shape parametrization into the parametric reduced order model to faithfully represent the patient’s anatomy. A non-intrusive approach relying on a sparse sampling of the space of anatomical features is introduced and validated. Then, we tackle the problem of data completion and image reconstruction from partial or incomplete datasets based on physical priors. The proposed solution has the potential to perform scene registration in the context of augmented reality for laparoscopy. Quasi-real-time computations are reached by using a new hyperreduction approach based on a sparsity promoting technique. Finally, the third challenge concerns the representation of biophysical systems under uncertainty of the underlying parameters. It is shown that traditional model order reduction approaches are not always successful in producing a low dimensional representation of a model, in particular in the case of electrosurgery simulation. An alternative is proposed using a metamodeling approach. To this end, we successfully extend the use of sparse regression methods to the case of systems with stochastic parameters

This thesis presents the study of the model order reduction for power grids and transmission networks. The specific focus has been the transient dynamics. A mathematical viewpoint has been adopted for model reduction. Power networks are huge and complex network, simulation for power grid analysis and design require large non-linear models to be solved. In the context of developing "Smart Grids" with the distributed generation of power, real time analysis of complex systems such as these needs fast, reliable and accurate models. In the current study we propose model order reduction methods both a-priori and a-posteriori suitable for dynamic models of power grids. The model that describes the transient dynamics of the power grids is complex non-linear swing dynamics model. The non-linearity of the swing dynamics model necessitates special attention to achieve maximum benefit from the model order reduction techniques. In the current research, POD and LATIN methods were applied initially with varying degrees of success. The method of TPWL has been proved as the best-suited model reduction method for swing dynamics model; this method combines POD with multiple linear approximations. For the transmission lines, a distributed parameters model in frequency-domain is used. PGD based reduced-order models are proposed for the DP model of transmission lines. A fully parametric problem with electrical parameters of transmission lines included as coordinates of the separated representation. The method was extended to present the solution of frequency-dependent parameters model for transmission lines.
Cette these présente l'étude de la réduction de modeles pour les réseaux électriques et les réseaux de transmission. Un point de vue mathématique a été adopté pour la réduction de modeles. Les réseaux électriques sont des réseaux immenses et complexes, dont l'analyse et la conception nécessite la simulation et la résolution de grands modeles non-linéaires. Dans le cadre du développement de réseaux électriques intelligents (smart grids) avec une génération distribuée de puissance, l'analyse en temps réel de systemes complexes tels que ceux-ci nécessite des modeles rapides, fiables et précis. Dans la présente étude, nous proposons des méthodes de réduction de de modeles a la fois a priori et a posteriori, adaptées aux modeles dynamiques des réseaux électriques. Un accent particulier a été mis sur la dynamique transitoire des réseaux électriques, décrite par un modele oscillant non­linéaire et complexe. La non-linéarité de ce modele nécessite une attention particuliere pour bénéficier du maximum d'avantages des techniques de réduction de modeles. lnitialement, des méthodes comme POD et LATIN ont été adoptées avec des degrés de succes divers. La méthode de TPWL, qui combine la POD avec des approximations linéaires multiples, a été prouvée comme étant la méthode de réduction de modeles la mieux adaptée pour le modele dynamique oscillant. Pour les lignes de transmission, un modele de parametres distribués en domaine fréquentiel est utilisé. Des modeles réduits de type PGD sont proposés pour le modele DP des lignes de transmission. Un probleme multidimensionnel entierement paramétrique a été formulé, avec les parametres électriques des lignes de transmission inclus comme coordonnées additionnelles de la représentation séparée. La méthode a été étendue pour étudier la solution du modele des lignes de transmission pour laquelle les parametres dépendent de la fréquence.
Esta tesis presenta un estudio de la reducción de modelos (MOR) para redes de transmisión y distribución de electricidad. El enfoque principal utilizado ha sido la dinámica transitoria y para la reducción de modelos se ha adoptado un punto de vista matemático. Las redes eléctricas son complejas y tienen un tamaño importante. Por lo tanto, el análisis y diseño de este tipo de redes mediante la simulación numérica, requiere la resolución de modelos no-lineales complejos. En el contexto del desarrollo de redes inteligentes, el objetivo es un análisis en tiempo real de sistemas complejos, por lo que son necesarios modelos rápidos, fiables y precisos. En el presente estudio se proponen diferentes métodos de reducción de modelos, tanto a priori como a posteriori, adecuados para modelos dinámicos de redes eléctricas. La dinámica transitoria de redes eléctricas, se describe mediante modelos dinámicos oscilatorios no-lineales. Esta no-linearidad del modelo necesita ser bien tratada para obtener el máximo beneficio de las técnicas de reducción de modelos. Métodos como la POD y la LATIN han sido inicialmente utilizados en esta problemática con diferentes grados de éxito. El método de TPWL, que combina la POD con múltiples aproximaciones lineales, ha resultado ser el mas adecuado para sistemas dinámicos oscilatorios. En el caso de las redes de transmisión eléctrica, se utiliza un modelo de parámetros distribuidos en el dominio de la frecuencia. Se propone reducir este modelo basándose en la PGD, donde los parámetros eléctricos de la red de transmisión son incluidos como coordenadas de la representación separada del modelo paramétrico. Este método es ampliado para representar la solución de modelos con parámetros dependientes de la frecuencia para las redes de transmisión eléctrica

Le corps humain est un système complexe, et l'œil humain n’y fait pas exception. Malgré les avancées médicales, de nombreuses questions pathologiques subsistent. Les modèles mathématiques et computationnels complètent les études cliniques en dévoilant des mécanismes physiopathologiques complexes.L'œil, accessible de manière non invasive, offre des marqueurs biologiques utiles pour diagnostiquer des maladies. Comprendre son comportement, ses pathologies et les traitements associés est donc essentiel.Cette thèse explore la modélisation et la simulation des flux oculaires, notamment le transfert de chaleur et le flux d’humeur aqueuse. Ces approches nécessitent des validations cliniques rigoureuses et tiennent compte de nombreux paramètres, spécifiques au patient ou externes. Une analyse de sensibilité globale évalue leur impact pour guider les cliniciens. Ces analyses, coûteuses en calculs, bénéficient de méthodes de réduction de modèle certifiées, permettant des simulations précises et plus rapides, favorisant l’intégration des modèles dans la pratique clinique
The human body is a complex system, and the human eye is no exception. Despite medical advances, many pathological questions remain. Mathematical and computational models complement clinical studies by revealing complex pathophysiological mechanisms.The eye, which can be accessed non-invasively, offers useful biological markers for diagnosing diseases. Understanding its behavior, pathologies and associated treatments is therefore essential.This thesis explores the modeling and simulation of ocular flows, notably heat transfer and aqueous humor flow. These approaches require rigorous clinical validation and take into account numerous parameters, both patient-specific and external. A global sensitivity analysis assesses their impact to guide clinicians. These computationally-intensive analyses benefit from certified model reduction methods, enabling accurate and faster simulations, favoring the integration of models into clinical practice

Submitted by Lucienne Costa (lucienneferreira@ufcg.edu.br) on 2018-05-02T20:48:28Z No. of bitstreams: 1 ALAN KELON OLIVEIRA DE MORAES – TESE (PPGCC) 2017.pdf: 2206784 bytes, checksum: 9e05994ae273d28b9871d8d16769dac8 (MD5)
Made available in DSpace on 2018-05-02T20:48:28Z (GMT). No. of bitstreams: 1 ALAN KELON OLIVEIRA DE MORAES – TESE (PPGCC) 2017.pdf: 2206784 bytes, checksum: 9e05994ae273d28b9871d8d16769dac8 (MD5) Previous issue date: 2017-08-31
O teste baseado em modelos é uma abordagem de teste de software que usa modelos abstratos de uma aplicação para gerar, executar e avaliar os testes. A geração de casos de testes exerce um papel importante no teste baseado em modelos. Como essa geração consiste na busca sistemática por casos de testes que possam ser extraídos dos modelos, o teste baseado em modelos geralmente produz suítes de testes que são caras demais para serem executadas completamente. Técnicas de redução de suítes de testes têm sido propostas para abordar este problema. O objetivo dessas técnicas é obter suítes de testes reduzidas que são mais baratas de serem executadas e tão efetivas na detecção de faltas quanto as suítes completas, dado que as suítes reduzidas mantém o mesmo nível de cobertura, definido por um critério de adequação de testes, da suíte completa. Esses critérios definem que partes do sistema serão testados, com que frequência e sob quais circunstâncias. Entretanto, pouca atenção tem sido dada ao impacto que a escolha do critério tem na redução de suítes de testes. Por outro lado, sistemas de tempo-real são sistemas reativos cujos comportamentos são restringidos pelo tempo. Consequentemente, faltas relacionadas ao tempo são específicas desses sistemas. Para lidar com isso, modelos para sistemas de tempo real devem trabalhar com tempo e, consequentemente, há critérios de adequação de testes específicos para eles. Contudo, a pesquisa sobre redução de suítes de testes não tem focado em sistemas de tempo-real, portanto o impacto de critérios de adequação de testes na redução de suítes é desconhecido. Nesta pesquisa de doutorado objetivamos investigar a influência de critérios de adequação de testes nos resultados da redução de suítes de testes no contexto de teste baseado em modelos de sistemas de tempo-real. Em particular, nós estamos interessados no modelo Timed Input-Output Symbolic Transition Systems (TIOSTS), porque ele é um modelo de sistema de transições no qual dados e tempo são definidos simbolicamente, já que sistemas de transição são a base para o teste de conformidade de sistemas de tempo real. Para alcançar o objetivo da pesquisa, primeiramente, nós definimos 19 critérios de adequação de testes para o modelo TIOSTS. Os critérios definidos incluem critérios baseados em transições, fluxo de dados e tempo. Depois nós formalizamos uma hierarquia com esses critérios, onde eles estão parcialmente ordenados pela relação de inclusão estrita. Segundamente, nós avaliamos empiricamente o custo-benefício de doze dos critérios definidos e cinco técnicas de redução de suítes de testes. Nós avaliamos o tamanho, o tempo de execução e a detecção de faltas das suítes de testes reduzidas de cada uma das 60 combinações de critério e técnica. No experimento, nós usamos modelos de especificação, em TIOSTS, de uma máquina de recarga de cartão do metrô, de um sistema de alarme anti-roubo e de um limitador automático de velocidade de carros. Além disso, usamos simulações das implementações, que geram rastros corretos para os modelos. Por fim, o teste de mutação foi usado para gerar mutantes dos modelos de especificação, que, por sua vez, foram traduzidos para simulações com a finalidade de simular modelos de implementações defeituosas. As evidências empíricas sugerem que os critérios de adequação de testes mais próximos do topo da hierarquia produziram suítes reduzidas com melhor custo-benefício com relação à detecção de faltas e tempo de execução. Com relação às técnicas de redução, a técnica aleatória obteve melhor custo-benefício dentre as técnicas avaliadas. Os resultados apontam que os critérios explicam mais a variação nos resultados do que as técnicas.
Model-based testing is a testing approach that relies on the existence of abstract models of an application to generate, execute and evaluate tests. Test case generation plays an important role in model-based testing. Since it consists of a systematic search for test cases that can be extracted from models, model-based testing usually generates large test suites which are too expensive to execute in full. Test suite reduction techniques have been proposed to address this problem. The goal of the techniques is to obtain reduced test suites that are both cheaper to execute and as effective at detecting faults as the original suite, given that the reduced test suites maintain the same coverage level of the complete test suite required by a test adequacy criterion. These criteria define which parts of the system are going to be tested, how often and under what circumstances. Nevertheless, little attention has been paid to the impact of the criterion choice in test suite reduction research. On the other hand, real-time systems are reactive systems whose behavior is constrained by time. Consequently, time-related faults are specific to these systems. In order to cope with this issue, models for real-time systems must deal with time and, consequently, there are specific test adequacy criteria for them. However, test suite reduction research has not focused on real-time systems, therefore the impact of test adequacy criteria for models of real-time systems on test suite reduction is unknown. In this doctoral research, we aim at investigating the influence of test adequacy criteria on the outcomes of test suite reduction techniques in the context of model-based testing of real-time systems. In particular, we are interested in the Timed Input-Output Symbolic Transition Systems (TIOSTS) model because it is an expressive transition system in which data and time are symbolically defined, and transition systems are the basis for conformance testing of real-time systems. In order to achieve the research objective, first, we defined 19 test adequacy criteria for TIOSTS models. The defined criteria include transition-based criteria, data-flow-oriented criteria and real-time systems criteria. Next, we formalized a hierarchy with these criteria which is partially ordered by the strict inclusion relation. Second, we evaluated the cost-effectiveness of twelve criteria and five test suite reduction techniques in empirical studies of test suite reduction. We evaluated the size, execution time and fault detection of reduced test suites obtained from each combination of criterion and technique. In the experiment, we used TIOSTS specification models of a refilling machine for charging the subway card, a burglar alarm system, and an automated car speed limiter; simulations of the implementations, which generate correct traces for the models; and mutation testing to generate mutants of the specification models, which were also translated to simulations in order to simulate faulty model implementations. Empirical evidence suggests that test adequacy criteria closer to the top of the family obtained reduced test suites with better costeffectiveness regarding fault detection and execution time. With respect to the test suite reduction techniques, the Random technique obtained better cost-effectiveness among the evaluated criteria. Results also suggests that the criteria explain more the variation in fault detection and execution time of reduced test suites than the techniques.

The main focus of this thesis is to extend and improve the applicability and the accuracy of the Unstructured Partial Element Equivalent Circuit (PEEC) method. The interest on this subject is spurred by the growing need of fast and efficient numerical methods, which may help engineers during the design and other stages of the production of new generation electric components. First, the PEEC method in its unstructured form is extended to magnetic media. In this regard, two formulations are developed and compared: the first one, based on the Amperian interpretation of the magnetization phenomena, is derived from the existing literature concerning the standard (structured) version of PEEC; the second one, based on the Coulombian interpretation of the magnetization phenomena, is proposed by the author with the aim of collocating PEEC in the context of Volume Integral Equation methods. Then, the application of low-rank compression techniques to PEEC is investigated. Two different methods are applied: the first is based on hierarchical matrices (H and H2 matrices) whereas the second is based on hierarchical-semi-separable (HSS) matrices. The two methods are compared and the main numerical issues which emerge by applying low-rank techniques to PEEC are analyzed. Finally, the developed unstructured PEEC method is combined with the Marching On-in Time scheme for the study of fast transient phenomena with wide range of harmonics. Moreover, two different stochastic PEEC methods are developed for uncertainty quantification analysis. The first is based on the Polynomial Chaos expansion while the second is based on the Parametric Model Order Reduction technique coupled with spectral expansion.
L'obiettivo principale di questa tesi è di estendere e migliorare l'applicabilità e l'accuratezza del metodo Partial Element Equivalent Circuit (PEEC) non strutturato (Unstructured PEEC). L'interesse riguardo tale argomento è stimolato dalla crescente necessità di metodi numerici rapidi ed efficienti, che possono aiutare gli ingegneri durante la progettazione e altre fasi della produzione di componenti elettrici ed elettronici di nuova generazione. Durante la prima fase della tesi, il metodo PEEC (nella sua forma non strutturata) è esteso ai mezzi magnetici. A questo proposito, vengono sviluppate e confrontate due formulazioni: la prima, basata sull'interpretazione amperiana dei fenomeni di magnetizzazione, deriva dalla letteratura esistente relativa alla versione standard (strutturata) del metodo PEEC; il secondo, basato sull'interpretazione coulombiana dei fenomeni di magnetizzazione, è proposto dall'autore con l'obiettivo di collocare il metodo PEEC nel contesto dei metodi di integrali di volume (Volume Integral Equation). Successivamente, la ricerca si focalizza sull’utilizzo di tecniche di compressione a basso rango al fine di risolvere problemi PEEC in maniera computazionalmente efficiente, salvaguardando tempo e memoria di calcolo. A tal proposito, vengono applicati due metodi diversi: il primo si basa su matrici gerarchiche (matrici H e H2 mentre il secondo si basa su matrici gerarchiche-semi-separabili (HSS). I due metodi vengono confrontati e vengono analizzati i principali problemi numerici che emergono applicando tali tecniche di compressione a basso rango al metodo PEEC. In seguito, il metodo PEEC non strutturato viene combinato con l’approccio Marching On-In Time (MOT) per lo studio di fenomeni transitori rapidi con un ricco contenuto armonico. Infine, sono stati sviluppati due diversi metodi PEEC stocastici per la quantificazione dell'incertezza. Il primo si basa sull'espansione Polynomial Chaos, mentre il secondo si basa sulla tecnica di riduzione d'ordine parametrica (Parametric Model Order Reduction) unita all'espansione spettrale.

Les systèmes temps-réel sont souvent validés par des analyses temporelles, qui sont mises en péril par des anomalies temporelles (AT). Une AT contre-intuitive a lieu quand une accélération locale conduit à un ralentissement global, et une AT d'amplification, quand un ralentissement local entraîne un ralentissement encore plus grand.Alors que les AT contre-intuitives menacent le bien-fondé ou la flexibilité des analyses, il n'existe pas d'outils pour les détecter de manière systématique. Nous proposons une structure formelle unifiée pour évaluer les définitions des AT, concluant au manque d'une définition pratique, principalement dû à l'absence de relations entre les effets temporels locaux et globaux. Nous y répondons par la causalité, que nous utilisons pour revoir la formalisation de ces AT. Nous proposons aussi une instance des notions spécialisée pour les pipelines out-of-order. Nous évaluons notre procédure de détection subséquente sur des exemples illustratifs et bancs de tests, montrant qu'elle permet de capturer précisément les AT.La complexité des systèmes exige que leurs analyses gèrent l'important espace d'états résultant. Une solution est de réaliser des analyses compositionnelles, précisément menacées par les AT d'amplification. Nous faisons progresser leur étude en montrant comment une abstraction spécialisée peut être adaptée pour un processeur industriel, en modélisant les caractéristiques temporelles clés avec des réductions appropriées. Nous illustrons aussi à partir de cette classe d'AT comment des stratégies de vérification peuvent être utilisées en vue de l'obtention de motifs d'AT
The timing behavior of real-time systems is often validated through timing analyses, which are yet jeopardized by execution phenomena called timing anomalies (TAs). A counter-intuitive TA manifests when a local speedup eventually leads to a global slowdown, and an amplification TA, when a local slowdown leads to an even larger global slowdown.While counter-intuitive TAs threaten the soundness/scalability of timing analyses, tools to systematically detect them do not exist. We set up a unified formal framework for systematically assessing the definitions of TAs, concluding the lack of a practical definition, mainly due to the absence of relations between local and global timing effects. We address these relations through the causality, which we further use to revise the formalization of these TAs. We also propose a specialized instance of the notions for out-of-order pipelines. We evaluate our subsequent detection procedure on illustrative examples and standard benchmarks, showing that it allows accurately capturing TAs.The complexity of the systems demands that their timing analyses be able to cope with the large resulting state space. A solution is to perform compositional analyses, specifically threatened by amplification TAs. We advance their study by showing how a specialized abstraction can be adapted for an industrial processor, by modeling the timing-relevant features of such a hardware with appropriate reductions. We also illustrate from this class of TAs how verification strategies can be used towards the obtainment of TA patterns

Il controllo predittivo basato su modello (MPC) è una tecnica di controllo avanzata che ha svolto un ruolo importante nella gestione di molti processi nel settore industriale. Oggi, nell’ottica di una gestione energetica efficiente degli edifici, l’utilizzo di questa strategia si sta dimostrando una soluzione promettente per ridurre al minimo i consumi e i costi energetici complessivi. Tuttavia, gli studi sulla sua fattibilità tecnica in edifici esistenti sono ancora in una fase iniziale. Pertanto, il risultato principale di questa tesi è la progettazione e lo sviluppo di un prototipo hardware e software per la verifica sul campo di un sistema di controllo predittivo, basato su modello, integrando un modello predittivo virtuale della porzione dell'edificio in esame, il controllore e l'interfaccia grafica per i dispositivi di monitoraggio e regolazione utilizzati. Inoltre, particolare attenzione è stata posta sulla fattibilità tecnica relativa all'implementazione di un tipico sistema MPC, che include un sottosistema di monitoraggio, un set di acquisizione dati e un metodo di identificazione del sistema per ottenere il modello per il controllore, mediante un approccio di modellazione grey-box. La fase di modellazione e l'approccio empirico sviluppato sono presentati nella prima parte di questa tesi di ricerca, mentre la parte centrale riguarda: lo sviluppo del prototipo di controllo predittivo, basato su modello, all'interno di uno strumento virtuale del software LabVIEW e la descrizione del test sperimentale, effettuato durante la stagione di riscaldamento, garantendo la normale operatività dell’edificio durante l'intero periodo di monitoraggio. Infine, è presentato lo studio sviluppato in ambiente di simulazione per indagare il potenziale della logica di controllo per la valutazione di scenari di riqualificazione. Il focus è sulla definizione dei principali componenti del simulatore MPC e sui risultati ottenuti testando uno degli scenari di intervento.
Model Predictive Control (MPC) is an advanced control technique which has played an important role in the management of many processes in the industry sector. Nowadays, in the perspective of an efficient building energy management, the exploitation of this strategy is proving to be a promising solution for minimising overall energy consumptions and costs. However, investigations on the feasibility of the technique in real existing buildings are at an initial stage. Hence, the main outcome of this dissertation is the design and development of a prototype hardware and software set up for on-field testing of a model-based predictive control system, integrating a virtual predictive model of the portion of the building under investigation, the controller and the interface to the monitoring and regulation devices used. Moreover, this research is addressed to investigate the technical feasibility of the development and deployment of a typical MPC system, which includes a monitoring sub-system, a data acquisition set up and a system identification method to obtain the model for the controller by means of a grey-box modelling approach. The modelling phase and the empirical approach developed are presented in the first part of this research thesis, while the core part concerns: the development of the MPC prototype, within a virtual instrument of LabVIEW software and the description of the experimental test, which was carried out during heating season, ensuring normal building operation during the entire monitoring period. Finally, this dissertation presents the study developed in simulation environment to investigate the potential of the control logic for the evaluation of retrofitting scenarios. The focus is on the definition of the main MPC simulator components and on the results obtained by testing one of the intervention scenarios.

Ce travail se place dans le cadre de l'assimilation de données en mécanique des structures. Il vise à développer de nouveaux outils numériques pour l'assimilation de données robuste et en temps réel afin d'être utilisés dans diverses activités d'ingénierie. Une activité cible est la mise en œuvre d'applications DDDAS (Dynamic Data Driven Application System) dans lesquelles un échange continu entre les outils de simulation et les mesures expérimentales est requis dans le but de créer une boucle de contrôle rétroactive sur des systèmes mécaniques connectés. Dans ce contexte, et afin de prendre en compte les différentes sources d'incertitude (erreur de modélisation, bruit de mesure,...), une méthodologie stochastique puissante est considérée dans le cadre général de l’inférence bayésienne. Cependant, un inconvénient bien connu d'une telle approche est la complexité informatique qu’elle engendre et qui rend les simulations en temps réel et l'assimilation séquentielle des données difficiles.Le travail de thèse propose donc de coupler l'inférence bayésienne avec des techniques numériques attrayantes et avancées afin d'envisager l’assimilation stochastique de données de façon séquentielle et en temps réel. Premièrement, la réduction de modèle PGD est introduite pour faciliter le calcul de la fonction de vraisemblance, la propagation des incertitudes dans des modèles complexes et l'échantillonnage de la densité a posteriori. Ensuite, l'échantillonnage par la méthode des Transport Maps est étudiée comme un substitut aux procédures classiques MCMC pour l'échantillonnage de la densité a posteriori. Il est démontré que cette technique conduit à des calculs déterministes, avec des critères de convergence clairs, et qu'elle est particulièrement adaptée à l'assimilation séquentielle de données. Là encore, l'utilisation de la réduction de modèle PGD facilite grandement le processus en utilisant les informations des gradients et hessiens d'une manière simple. Enfin, et pour accroître la robustesse, la correction à la volée du biais du modèle est abordée à l'aide de termes d'enrichissement fondés sur les données. Aussi, la sélection des données les plus pertinentes pour l’objectif d’assimilation est abordée.Cette méthodologie globale est appliquée et illustrée sur plusieurs applications académiques et réelles, comprenant par exemple le recalage en temps réel de modèles pour le contrôle des procédés de soudage, ou l’étude d'essais mécaniques impliquant des structures endommageables en béton instrumentées par mesures de champs
The work is placed into the framework of data assimilation in structural mechanics. It aims at developing new numerical tools in order to permit real-time and robust data assimilation that could then be used in various engineering activities. A specific targeted activity is the implementation of DDDAS (Dynamic Data Driven Application System) applications in which a continuous exchange between simulation tools and experimental measurements is envisioned to the end of creating retroactive control loops on mechanical systems. In this context, and in order to take various uncertainty sources (modeling error, measurement noise,..) into account, a powerful and general stochastic methodology with Bayesian inference is considered. However, a well-known drawback of such an approach is the computational complexity which makes real-time simulations and sequential assimilation some difficult tasks.The PhD work thus proposes to couple Bayesian inference with attractive and advanced numerical techniques so that real-time and sequential assimilation can be envisioned. First, PGD model reduction is introduced to facilitate the computation of the likelihood function, uncertainty propagation through complex models, and the sampling of the posterior density. Then, Transport Map sampling is investigated as a substitute to classical MCMC procedures for posterior sampling. It is shown that this technique leads to deterministic computations, with clear convergence criteria, and that it is particularly suited to sequential data assimilation. Here again, the use of PGD model reduction highly facilitates the process by recovering gradient and Hessian information in a straightforward manner. Eventually, and to increase robustness, on-the-fly correction of model bias is addressed using data-based enrichment terms.The overall cost-effective methodology is applied and illustrated on several academic and real-life test cases, including for instance the real-time updating of models for the control of welding processes, or that of mechanical tests involving damageable concrete structures with full-field measurements

Dans cette thèse nous proposons un modèle temporel dans le cadre des machines à états abstraits (ASM). Une extension du langage de spécification ASM est développé qui correspond à ce modéle temporel pour le temps continu. L'extension du langage avec des constructions de temps permet de diminuer la taille de la spécification et donc de réduire la probabilité d'erreurs. La sémantique de l'extension du langage ASM est fournie et prend en compte les définitions des fonctions externes, les valeurs des délais et les choix de résolution des non-déterminismes. Un sous-système de vérification des propriétés exprimées en logique FOTL (FirstOrder Timed Logic) est développé. Un simulateur d'ASMs temporisées est développé et implémenté, il comprend un analyseur syntaxique, un interprète du langage, un sous-système de vérification des propriétés ainsi qu'une interface graphique

Les automates cellulaires constituent le modèle de calcul parallèle et local par excellence.Comme pour tout modèle du parallélisme, leur programmation est réputée difficile. La puissance de calcul des automates cellulaires, modèle le plus simple du parallélisme, est attestée par le fait que nombre de problèmes significatifs sont calculés en temps minimal, appelé temps-réel, surautomate cellulaire.Principal résultat de cette thèse, on démontre des liens exacts (des équivalences) entre d’un côté la complexité descriptive, essentiellement la définissabilité en logique du second ordre existentiel sur des formules de Horn, de l’autre les classes de complexité en temps-réel des automates cellulaires.Au-delà de cette caractérisation en logique de la complexité en temps minimal, la thèse établit une méthode de programmation parallèle. Cette méthode consiste dans un premier temps à programmer dans nos logiques de Horn l’induction résolvant un problème, puis dans un secondtemps, à appliquer un processus automatique aboutissant au programme de l’automate cellulaire résolvant le problème. Pour justifier l’intérêt de la méthode, la thèse présente un ensemble de programmes logiques pour une variété représentative de problèmes classiques connus pour êtrecalculables en temps-réel sur automates cellulaires.Par ailleurs, toujours en passant par nos logiques, on prouve divers résultats liant le temps-réel des automates cellulaires et les grammaires formelles. Typiquement, tout langage généré par une grammaire algébrique et, plus généralement, une grammaire conjonctive d’Okhotin, est reconnu en temps-réel sur un automate cellulaire de dimension 2
Cellular automata constitute the model of parallel and local computation by excellence.As for any model of parallelism, their programming is known to be difficult. The computingpower of cellular automata, the simplest model of parallelism, is attested by the fact that manysignificant problems are computed in minimal time, called real-time, on cellular automata.The main result of this thesis is the demonstration of exact links (equivalences) between, on onehand, the descriptive complexity, essentially the definability in existential second order logic on Horn formulas, and, on the other hand, the real-time complexity classes of cellular automata.Beyond this characterization in logic of the complexity in minimal time, the thesis establishes a method of parallel programming. This method consists first of all in programming in our Horn ogics the induction solving a problem, then in a second step, in applying an automatic process leading to the program of the cellular automaton solving the problem. To justify the interest of the method, the thesis presents a set of logic programs for a representative variety of classical problems known to be computable in real-time on cellular automata.In addition, we prove various results linking the real time of cellular automata and formal grammars. Typically, any language generated by an algebraic grammar and, more generally, an Okhotin conjunctive grammar, is recognized in real-time on a 2-dimensional cellular automaton

博士
國立臺灣大學
電信工程學研究所
96
Hybrid methods combining time-domain full-wave electromagnetic simulation and Krylov subspace based model order reduction techniques are proposed for efficiently obtaining the late-time responses of closed systems. In general, model order reduction process is applied after the sources fade to zero for extracting the active modes in the system. Late-time responses are then constructed by the linear combination of the extracted modes. Taking advantage of the space information, only few direct time domain iterations are required after the sources fade to zero with additional round-trip time before the extraction of the excited modes. After the dominant modes are properly extracted, the late time response of the system can be easily reconstructed by analytic expressions. With the proposed hybridizing techniques, existing codes of time-domain simulation can be resorted. Several numerical examples are provided for the verification of the correctness, efficiency, convergence, and complexity of the proposed hybrid methods, which show that with only very few iterations of model order reduction, good agreement can be achieved between the results of the proposed hybrid methods and those obtained from direct time-domain iterations.

碩士
國立中山大學
電機工程學系研究所
93
It is well known that the finite difference time domain (FDTD) method is a powerful numerical analysis tool for solving electromagnetic problems. In a simulated area, in order to discretize an object which is much smaller than the others, a very small space increment is needed and hence the time step should be decreased too for stability consideration in traditional FDTD. The small space and time increments will respectively increase the memory requirement and calculation time. To overcome these problems, some numerical methods were developed, such as the subcell and nonuniform grid method, to handle the small feature size. This thesis describes an efficient method for generating FDTD subcell equations. We construct a second order macromodel system instead of the subcell region in conventional FDTD. The macromodel system can be reduced with model order reduction techniques (MOR) and then translated into new FDTD update equations. When the problem contains several objects of the same size and material properties, the MOR subcell has the advantage of reusability. This means that the reduce-order model of the object needs to be generated only once nonetheless can be applied to every position where the objects originally occupied.

碩士
大葉大學
資訊管理學系碩士在職專班
93
The major profit of an enterprise comes from order, however, many company face the problem of blind or busy orders. Blind orders may from the situation as the more orders you take, the more profit you will lose. In this case, if an enterprise can make all the information well to analyze order and decide accept the order or not in advance, then it can definitely keep the advantages of competition. The research of “orders selection” mostly bases on batch mode in nowadays. There are rarely research focus on “real-time” mode integrating the in-time information (ERP) system to make an initial judgment of an order. In this way, the sales person can speedily know the contribution of the order and the shortcoming or advantages in every structure. And then, they can take the resources of company to improve the disadvantage of this order and make the next step to interactive with customers. This research want to construct a single order accept model system which focus on the elements affecting an enterprise’s accepting order, and how to make the rules to evaluate the criterion. The research uses AHP, linguistic variables of Fuzzy theory and fuzzy mathematic rules to construct a model and it also apply real cases to make the suggestion let the executor make references.

碩士
國立高雄應用科技大學
化學工程與材料工程系
99
In the paper, the technique of adaptive control is used for real-time model order identification and parameter estimation for linear time-invariant systems. At first, we use central difference to approximate the differential operator. And then the regression model of the controlled process is derived and the model parameters are obtained by the real-time recursive least square algorithm. To identify the model order, the continued fraction expansion (CFE) of the Cauer second form is first used to represent the preset process model and the exact process model is then obtained by the truncated CFE. The simulation results by Matlab/Simulink software that the proposed method is able to quickly identify the model order and parameters of the controlled process.

碩士
國立臺灣海洋大學
河海工程學系
101
Due to the disaster of flooding caused by typhoon has been increasingly serious, and many cities have been highly developed with a great growing population that require protections; thus, the flooding and disaster prevention for urban areas has become an important issue on policy. The purpose of this study developed a real-time simulation model for the drainage system in urban areas by receiving real-time rainfall and tidal data, which can react upon the flooding situation in cities as to provide reference for broadcast of flooding warning. This study used SWMM model developed by the U.S. Environmental Protection Agency to simulate the water situation of sewer system and river system, and employed street model (Lee and Lin, 2011) to simulate urban surface which adopted the Preissmann implicit finite-difference scheme with the double-sweep method to solve de Saint-Venant equations.A modified kinematic-wave based on geomorphic instantaneous unit hydrograph model (Lee and Chang, 2005)and kinematic-wave V-shaped overland flow model(Chow et al. , 1988) was used to simulate runoff processes. The above models were combined to explore the effect of drainage of urban sewer system caused by tidal boundary. Nevertheless, such physical models take times as real-time formulations were calculated at the risk of the numerical divergences. Therefore, this study particularly used warning methods of design condition and current condition (Lee et al. 2011) for setting warning values of rainfall for every district of Taiwan and the study area. This study selectedLotong Township as watershed of applied model. The results showed that the tidal variation of the Dongshan River estuary had a slight effect upon sewer system of Lotong Township. However, this study specifically developed three schemes - revise pipe, diversion flow and detention pond - for most flooding regions to effectively promote drain ability of sewer system in Lotong Township. In addition, the results indicated that the situation of overflow of every chocking manhole has been effectively improved after adjusting sewer system of Lotong Township through the schemes. As a result, the scheme of inundation reduction can provide an example for flooding and disaster prevention for Lotong Township. Consequently, this study combined the results with flash flood guidance theory to establish FFG of chocking manholes as a preference of rainfall warning value in case of flooding disaster caused by Typhoon.

碩士
輔仁大學
統計資訊學系應用統計碩士班
106
Nowadays in Taiwan people find themselves hard to pay living expenses just by their salaries, and stocks became a popular choice to gain wealth. Stock Price varies with many unpredictable messages or some unperceivable complicated relations, so there are many variables to consider about. If there are ways good enough to reduce dimensions and get features that really changes stock price, it will be able to determine trends and get more remuneration. So, this research uses real time information of Taiwanese stock market, western and some Asian index along with information of Institutional investors related with Margin Trading and Short Selling in Taiwan Stock Exchange, and use autoencoder to reduce dimension and predict the stock prices of three targets: MediaTek, Getac and CTBC Financial Holding. However, the dimension reduces by autoencoder didn't take the effects by response variable into consideration. So we submitted a resolution by adding random forest and change point detection to select further outcomes made by autoencoder. Our results proved that compared to using all variables or use only autoencoder, using autoencoder to decrease dimensions and use random forest and change point detection can get lower RMSE.