Dans ce travail, une approche d’optimisation dynamique hybride est développée et utilisée pour simuler et optimiser les procédés d’adsorption modulée en pression (PSA). Elle est principalement basée sur la formulation hybride du modèle du procédé et sur l’utilisation de la méthode du système adjoint.Le problème de simulation qui consiste à déterminer le régime stationnaire cyclique (CSS) est formulé comme un problème d’optimisation où le critère de performance est défini par la condition de CSS, les variables de décision sont données par les valeurs initiales des variables d’état, et les contraintes par le modèle hybride du procédé avec les conditions aux limites associées. En optimisation, le vecteur des variables de décision contient, en plus des valeurs initiales de l’état, les paramètres de dimensionnement et de fonctionnement. La condition de CSS devient, dans ce cas, une contrainte à satisfaire par chaque solution optimale. Plusieurs modèles de procédés, allant du plus simple au plus compliqué, sont ´étudiés.Il s’agit notamment de procédés isothermes et non isothermes avec et sans états gelés. Les critères de performance considérés sont la pureté, la récupération et l´énergie. Les résultats obtenus aussi bien au niveau des performances des procédés considérés que de la robustesse de l’algorithme d’optimisation mis en œuvre, sont tout `a fait intéressants et montrent le grand potentiel de l’approche développée pour le dimensionnement et le fonctionnement optimaux des procédés PSA / The objective of the work was to develop a hybrid dynamic optimisation approach for simulation and optimisation of pressure swing adsorption (PSA) processes. It is mainly based on the hybrid formulation of the process model and on the use of adjoint system method. The simulation problem which consists in determining the cyclic steady state (CSS) is formulated as an optimisation problem where the CSS condition is considered as the performance index, initial values of state variables as decision variables and process model along with associated conditions as constraints. In optimisation, the decision vector consists of design and operation parameters in addition to the initials values of state variables whereas the CSS condition is considered in this case as a constraint to be satisfied for each optimal solution. Several process models with a varied degree of complexity have been studied. These models are isothermal and non isothermal with and without frozen states. The performance index considered are purity, recovery and energy. The results obtained are interesting vis-a-vis the performance of the processes considered as well as the robustness of the optimisation algorithm and show the great potential of the approach developed for the optimal design and operation of PSA processes

Dans cette thèse, nous présentons un modèle à commutation de la dynamique de prolifération d’une plante aquatique envahissante : le Typha. Ce modèle appartient à la classe des systèmes hybrides qui sont relativement récents en biomathématique. II décrit la dynamique de colonisation de la plante en prenant en compte la saisonnalité de l’un des modes de reproduction qu’est la reproduction sexuée. Cette étude est motivée par le fait que durant cette dernière décennie, le Typha est parvenu à coloniser le Parc National des Oiseaux de Djoudj (PNOD), perturbant ainsi l’écosystème et encombrant considérablement les activités agricoles des populations locales. Il y a eu différentes formes de luttes expérimentées pour réduire sa prolifération.Toutefois, ces mesures se sont avérées peu efficaces et d’un coût financier considérable. Pourtant, il existe des modèles mathématiques sur le développement du Typha susceptibles de favoriser une lutte efficace contre cette plante envahissante. Mais ils sont phénologiques. Notre travail fait partie d’un effort de contribution écohydrologique pour la compréhension des rôles de chaque type de reproduction sur la dynamique de prolifération. Le travail mené dans cette thèse vise à construire un modèle mathématique en considérant des hypothèses biologiques sur la reproduction du Typha, à analyser le modèle afin de suggérer une stratégie de lutte inspirée par les mathématiques. Nous analysons les sous-modèles qui composent le modèle à commutation et en ajoutant certaines hypothèses sur les valeurs des paramètres du modèle. Nous étudions d’abord l’équilibre nul du modèle à commutation. Ensuite, nous analysons un modèle de dimension deux qui constitue le modèle réduit du modèle général pour confronter les résultats avec ceux qu’on ne pourrait démontrer avec le modèle général de dimension trois. Enfin, nous déterminons une condition d’existence de cycle limite du modèle réduit. Nous établissons, pour tous les cas étudiés, la stabilité asymptotique et globale de l’équilibre nul (équilibre sans plante) lorsque le taux de reproduction de base du système considéré est inférieur à 1: Nous obtenons également pour chacun des sous-modèles étudiés, une condition de stabilité asymptotique de l’équilibre positif lorsque son taux de reproduction de base est supérieur à 1. Dans le cas du modèle réduit, nous montrons que lorsque la moyenne pondérée des taux de reproduction de base des sous-modèles est inférieure à 1, les solutions convergent vers l’équilibre nul. Par contre, lorsque cette moyenne est supérieure à 1, nous montrons l’existence d’un cycle limite. / In this thesis, we propose and analyze a switching dynamics model of the proliferation of invasive aquatic plant : Typha. This model which belongs to the class hybrid systems is relatively new in the field biomathematics. It describes the colonization dynamics of the plant taking into account the seasonality of type of reproduction : the sexual reproduction. During the last decade, the plant has colonized PNOD, disrupting the ecosystem and also causing enormous problems for the local population. There had been several significant attrempts to reduce its proliferation.However, these attempts have been futile an inefficient due to the large financial cost. There are some few phenological mathematical models on development of Typha. The propose study is part of an eco-hydrological effort to contribute to the understanding of the roles of each type of reproducing on the proliferation dynamics of Typha. The three main goals of this thesis are : To construct a mathematical model based on biological hypotheses of the reproduction of Typha,– analyze the model and– suggest a proliferation combatting strategy.We analyze sub-models that make up the switching/commutation model by assumptions or considering some hypothesis on the values of the model parameters. We study the zero equilibrium of the switching model, and then we propose and analyze a two-dimensional model by reducing the general model to set the stage for the analysis of the more complicated general three- dimension model. Finally, we determine a condition for the existence of limit cycle of the model. In all the sub-models studies, we establish the local and glob al asymptotic stability of zero equilibrium (equilibrium without any Typha plant) when the basic reproduction rate of the system under consideration is less than unity. We also obtain the condition under which thepositive or non-zero equilibrium of the model/sub-models asymptotically stable when the basic reproduction rate is greater than unity. For the specific case of the reduced model, we show that when the weighted average of the breeding rate of this sub-model is less than 1, the solutions converge to the zero equilibrium. When this average is greater than 1, we prove the existence of a limit cycle.

A síntese e caracterização de sistemas hetero-híbridos gerados a partir da 3,6-bi-2-piridil-1,2,4,5-tetrazina (bptz) e interações com ouro nanoparticulado são abordados nesta tese. O bptz foi estudado por meio de métodos espectroscópicos e teóricos, focalizando principalmente o efeito SERS associado à adsorção em nanopartículas de ouro. O mecanismo de transferência de carga para metais macroscópicos foi transposto para a condição nanoparticulada, envolvendo ligações químicas entre bptz e as nanopartículas. Os complexos estudados possuem fórmula geral [LmRu(µ-bptz)RuLm]Xn, com “L” indicando os ligantes periféricos 5-cloro-1,10-fenantrolina (Clphen) ou 4’-(fenil)-2,2&#8217:6&#8217,2&#8221-terpirdina (ptpy) e “X” os contra-íons. Foram investigadas suas espectroeletroquímicas eletrônica e SERS, e as mudanças de perfil vibracional foram modeladas, incorporando o mecanismo de transferências de carga entre complexo e o ouro, além dos mecanismos ressonantes e eletromagnéticos / The synthesis and investigation of heterohybrid systems encompassing 3,6-bi-2-pyridyl-1,2,4,5-tetrazine (bptz) and its ruthenium complexes associated with gold nanoparticles are dealt with in this thesis. Bptz was characterized by spectroscopic and theoretical techniques, focusing on its SERS spectra after the adsorption onto nanoparticles. The charge transfer mechanism in the SERS spectra of macroscopic metals was transposed to the nanoparticle condition, assuming the formation of chemical bonds between bptz and the nanoparticles. Complexes of general formula [LmRu(µ--bptz)RuLm]Xn, “L” the peripheric ligants 5-chlorine-1,10-phenantroline or 4’-(phenyl)-2,2&#8217:6&#8217,2&#8221-terpyrdine (ptpy), and “X” counter-ions were also investigated, with special emphasis on their electronic and SERS spectroelectrochemistry. The changes in the vibrational profiles were successfully explained by the occurrence of charge transfer between the adsorbed complex and gold, in addition to the electromagnetic and resonance mechanisms.

A bioatividade de complexos metálicos é de grande interesse atual na área de pesquisa de novos medicamentos. Fármacos anti-inflamatórios não-esteróides (FAINEs) são amplamente consumidos para o tratamento de doenças inflamatórias e de dor, mas seu contínuo uso pode ocasionar sérios efeitos colaterais ao trato gastrointestinal (TGI). Metalofármacos alternativos de Cu-FAINEs causam menores danos ao TGI, sem perder a atividade anti-inflamatória. No entanto, alguns destes são pouco solúveis. Adicionalmente, o desenvolvimento de materiais híbridos de FAINEs/polímeros para liberação modificada dos fármacos é interessante do ponto de vista de se aumentar a eficácia terapêutica e minimizar problemas de toxicidade. O objetivo deste trabalho foi investigar metodologia, com emprego de secagem via spray-dryer, para a preparação de materiais híbridos contendo complexos de cobre(II) com os fármacos naproxeno e sulindaco incorporados em acetato de celulose. Dois tipos de materiais híbridos foram obtidos para cada metalofármaco: CuFAINE/AC e CuFAINE/AC/OM, sendo o segundo preparado em presença de óleo mineral (OM). A caracterização destes produtos foi efetuada por análises elementares, espectroscopia vibracional no infravermelho, difratometria de raios X, análise termogravimétrica acoplada a espectrômetro de massas, ressonância paramagnética eletrônica, microscopia eletrônica de varredura e espalhamento de luz dinâmico. Os métodos utilizados levaram à interação dos metalofármacos de CuFAINEs com a matriz polimérica biocompatível de acetato de celulose, gerando materiais híbridos bioinorgânicos-orgânicos. Duas espécies de CuFAINEs: uma dimérica mantendo a estrutura original ([Cu2(FAINE)4(Laxial)2]) e outra monomérica - provavelmente do tipo ([Cu(FAINE)2(Laxial)2]) - foram identificadas nestes materiais. As partículas apresentaram tamanho nanométrico. Os resultados aqui obtidos abrem novas perspectivas para futuros estudos que visem à investigação da viabilidade de usar sistemas como estes para liberação mofidicada dos metalofármacos de CuFAINEs / The bioactivity of metal complexes has current and high interest in the field of development of new drugs. Non-steroidal anti-inflammatory drugs (NSAIDs) are widely consumed to treat inflammatory diseases and pain. The continuous use of these drugs causes serious side-effects on the gastrointestinal tract (GT). Alternative copper(II)-metallodrugs (CuNSAIDs) show lower GT side-effects without loss of anti-inflammatory activity. However, some of them are poorly water-soluble. In addition, the development of NSAID/biocompatible polymeric hybrid materials for modified-release of pharmaceuticals is of interest to increase therapeutic efficacy and to lower toxicity of drugs. The aim of this work was to investigate a methodology adding the use of spray-drying technique to prepare hybrid materials containing copper(II) complexes with naproxen and sulindac drugs incorporated into cellulose acetate (CA). Two different materials were obtained for each metallodrug: CuNSAID/CA and CuNSAID/CA/MO, being the second one prepared in the presence of mineral oil (MO). The characterization of these products was conducted by elemental analysis, vibrational spectroscopy, powder x-ray diffraction, thermogravimetric analysis coupled to mass spectrometry, electron paramagnetic resonance, scanning electron microscopy and dynamic light scattering. The developed methodologies led to the interaction between the CuNSAIDs and the polymeric matrix generating bioinorganic-organic hybrid materials. Two CuNSAIDs species: a dimeric one that kept the complex original structure [Cu2(FAINE)4(Laxial)2] and a monomeric species, ([Cu(FAINE)2(Laxial)2] were identified in these materials. The particles have shown nanometric size. The obtained results open up new perspectives for future studies with the aim of investigating the availability of using this type of system for modified release of CuNSAIDs metallodrugs.

En automatique, l'obtention d'un modèle du système est la pierre angulaire des procédures comme la synthèse d'une commande, la détection des défaillances, la prédiction... Cette thèse traite de l'identification d'une classe de systèmes complexes, les systèmes dynamiques hybrides. Ces systèmes impliquent l'interaction de comportements continus et discrets. Le but est de construire un modèle à partir de mesures expérimentales d'entrée et de sortie. Une nouvelle approche pour l'identification de systèmes hybrides linéaires basée sur les propriétés géométriques des systèmes hybrides dans l'espace des paramètres est proposée. Un nouvel algorithme est ensuite proposé pour le calcul de la solution la plus parcimonieuse (ou creuse) de systèmes d'équations linéaires sous-déterminés. Celui-ci permet d'améliorer une approche d'identification basée sur l'optimisation de la parcimonie du vecteur d'erreur. De plus, de nouvelles approches, basées sur des modèles à noyaux, sont proposées pour l'identification de systèmes hybrides non linéaires et de systèmes lisses par morceaux / In automatic control, obtaining a model is always the cornerstone of the synthesis procedures such as controller design, fault detection or prediction... This thesis deals with the identification of a class of complex systems, hybrid dynamical systems. These systems involve the interaction of continuous and discrete behaviors. The goal is to build a model from experimental measurements of the system inputs and outputs. A new approach for the identification of linear hybrid systems based on the geometric properties of hybrid systems in the parameter space is proposed. A new algorithm is then proposed to recover the sparsest solutions of underdetermined systems of linear equations. This allows us to improve an identification approach based on the error sparsification. In addition, new approaches based on kernel models are proposed for the identification of nonlinear hybrid systems and piecewise smooth systems

Cette thèse se situe dans le cadre de l'analyse et de la commande des systèmes multidimensionnels. Ce sont des systèmes où l'information se propage dans plusieurs directions indépendantes les unes des autres (par exemple une dimension d'espace et une de temps). Les contributions présentées dans ce mémoire portent d'une part sur la commande des systèmes 2D discrets ou continus, à retards constants ou variables, et d'autre part sur la synthèse de loi de commande par retour d'état robuste des systèmes nD hybrides incertains dont l'incertitude est décrite sous forme de représentation rationnelle implicite (ILFR). Les travaux présentés utilisent deux approches, l'une basée sur le polynôme caractéristique et l'autre sur les techniques de Lyapunov. Pour les systèmes 2D à retards discrets ou continus nous avons utilisé l'approche basée sur des fonctionnelles de Lyapunov. Des conditions suffisantes de stabilité et de stabilisation par retour d'état, dépendantes du retard, sont établies. Outre la notion de stabilité, la notion de performance du type H∞ est traitée afin de résoudre le problème de rejet de perturbations pour cette classe de systèmes. Nous avons ensuite proposé un cadre assez général pour l'analyse en stabilité des systèmes nD hybrides, en utilisant la S-procédure, permettant l'obtention de conditions sous forme de LMIs faciles à exploiter numériquement. Nous avons également proposé des conditions de stabilité et de stabilisation robustes pour les systèmes nD hybrides incertains dont l'incertitude est du type LFR implicites. / This thesis deals with the analysis and the control of multidimensional systems. These systems can be defined as the classe of systems where the information is propagated in several independent direction. For instance, a 2D system with a dimension corresponding to space and, the other one to time. The contributions presented in this work focuses, on one hand, on the control of 2D discrete or continuous systems with constant or variable delays and on the other hand, on the synthesis of robust state feedback controllers for nD hybrid uncertain systems including parameter uncertainties complying with an implicit linear fractional representation (ILFR). Two approaches are used. One is based on the characteristic polynomial and the other on Lyapunov techniques. Sufficient conditions for stability and stabilization are established for 2D delayed discrete or continuous systems using Lyapunov approach. Conditions to insure both stability and a prescribed H∞ performance level are given for this class of systems. Then, a general framework for the establishment of computationally tractable LMI conditions to analyse the stability of nD hybrid systems is proposed. Robust stability and stabilization conditions are then established for nD hybrid uncertain systems. The uncertainties comply with an (ILFR) description.

A bioatividade de complexos metálicos é de grande interesse atual na área de pesquisa de novos medicamentos. Fármacos anti-inflamatórios não-esteróides (FAINEs) são amplamente consumidos para o tratamento de doenças inflamatórias e de dor, mas seu contínuo uso pode ocasionar sérios efeitos colaterais ao trato gastrointestinal (TGI). Metalofármacos alternativos de Cu-FAINEs causam menores danos ao TGI, sem perder a atividade anti-inflamatória. No entanto, alguns destes são pouco solúveis. Adicionalmente, o desenvolvimento de materiais híbridos de FAINEs/polímeros para liberação modificada dos fármacos é interessante do ponto de vista de se aumentar a eficácia terapêutica e minimizar problemas de toxicidade. O objetivo deste trabalho foi investigar metodologia, com emprego de secagem via spray-dryer, para a preparação de materiais híbridos contendo complexos de cobre(II) com os fármacos naproxeno e sulindaco incorporados em acetato de celulose. Dois tipos de materiais híbridos foram obtidos para cada metalofármaco: CuFAINE/AC e CuFAINE/AC/OM, sendo o segundo preparado em presença de óleo mineral (OM). A caracterização destes produtos foi efetuada por análises elementares, espectroscopia vibracional no infravermelho, difratometria de raios X, análise termogravimétrica acoplada a espectrômetro de massas, ressonância paramagnética eletrônica, microscopia eletrônica de varredura e espalhamento de luz dinâmico. Os métodos utilizados levaram à interação dos metalofármacos de CuFAINEs com a matriz polimérica biocompatível de acetato de celulose, gerando materiais híbridos bioinorgânicos-orgânicos. Duas espécies de CuFAINEs: uma dimérica mantendo a estrutura original ([Cu2(FAINE)4(Laxial)2]) e outra monomérica - provavelmente do tipo ([Cu(FAINE)2(Laxial)2]) - foram identificadas nestes materiais. As partículas apresentaram tamanho nanométrico. Os resultados aqui obtidos abrem novas perspectivas para futuros estudos que visem à investigação da viabilidade de usar sistemas como estes para liberação mofidicada dos metalofármacos de CuFAINEs / The bioactivity of metal complexes has current and high interest in the field of development of new drugs. Non-steroidal anti-inflammatory drugs (NSAIDs) are widely consumed to treat inflammatory diseases and pain. The continuous use of these drugs causes serious side-effects on the gastrointestinal tract (GT). Alternative copper(II)-metallodrugs (CuNSAIDs) show lower GT side-effects without loss of anti-inflammatory activity. However, some of them are poorly water-soluble. In addition, the development of NSAID/biocompatible polymeric hybrid materials for modified-release of pharmaceuticals is of interest to increase therapeutic efficacy and to lower toxicity of drugs. The aim of this work was to investigate a methodology adding the use of spray-drying technique to prepare hybrid materials containing copper(II) complexes with naproxen and sulindac drugs incorporated into cellulose acetate (CA). Two different materials were obtained for each metallodrug: CuNSAID/CA and CuNSAID/CA/MO, being the second one prepared in the presence of mineral oil (MO). The characterization of these products was conducted by elemental analysis, vibrational spectroscopy, powder x-ray diffraction, thermogravimetric analysis coupled to mass spectrometry, electron paramagnetic resonance, scanning electron microscopy and dynamic light scattering. The developed methodologies led to the interaction between the CuNSAIDs and the polymeric matrix generating bioinorganic-organic hybrid materials. Two CuNSAIDs species: a dimeric one that kept the complex original structure [Cu2(FAINE)4(Laxial)2] and a monomeric species, ([Cu(FAINE)2(Laxial)2] were identified in these materials. The particles have shown nanometric size. The obtained results open up new perspectives for future studies with the aim of investigating the availability of using this type of system for modified release of CuNSAIDs metallodrugs.

Made available in DSpace on 2015-03-05T13:56:58Z (GMT). No. of bitstreams: 0 Previous issue date: 23 / Nenhuma / Este trabalho apresenta um método de Inteligência Artificial para ser utilizado em operações de Restauração de Sistemas de Potência, com o desenvolvimento de um sistema hibrido. O sistema utiliza um Algoritmo Genético (AG), que codifca seqüências de operações de chaveamento. Durante a evolução do AG, as operações sugeridas são avaliadas por um programa de Fluxo de Potência, que fornece dados para a função objetivo deste AG. Para este AG, uma nova representação de genoma foi criada, capaz de codificar estágios compostos de blocos de operações sequenciais. Juntamente a esta nova representação, foram desenvolvidos operadores de cruzamento e de mutação / This work reports the development of an Artificial Intelligence system applied to Power Systems Restoration, with an hybrid approach. The system employs a Genetic Algorithm (GA) to generate sequences of operations which are analysed by a Power Flow program to verify and access their fitness. For this GA, a new genoma representation was developed, as well as two genetic operators, for crossover and mutation

O presente trabalho descreve um modelo híbrido para sistemas especialistas classificatórios que Integra os mecanismos de representação de conhecimento de frames e de redes neurais, assimilando qualidades inerentes de cada um dos dois paradigmas. O mecanismo de frames fornece tipos construtores flexíveis para a modelagem do conhecimento do domínio, enquanto as redes neurais possibilitam tanto a automação da aquisição de conhecimento, a partir de uma base de casos, quanto a implementação de aprendizado indutivo e dedutivo. Empregam-se dois tipos de frames para representar o conhecimento relativo aos problemas de classificação: frames de achado e frames de diagn6stico. Os primeiros descrevem os objetos elementares da aplicação que podem influenciar na detecção de certos diagnósticos. Os frames de diagnóstico, por sua vez, são descritos em termos de achados e correspondem as descrições dos problemas de classificação tratados pelo sistema especialista. E utilizado o Modelo Neural Combinatório (MNC), caracterizado por sua capacidade de aprendizado heurístico através do reconhecimento de regularidades observadas no mundo exterior. A integração do mecanismo de representação de frames e do MNC se da ao nível dos frames de diagn6stico, armazenando representações simbólicas do conhecimento das redes neurais que explicitam o conhecimento armazenado nas camadas intermediarias das Últimas. O motor de inferência do sistema opera de maneira integrada sobre ambos os paradigmas, o conexionista e o simbólico. A máquina de aprendizado fornece meios para a automação da aquisição de conhecimento a partir de um banco de casos, bem como para o refinamento do conhecimento do sistema. A máquina de aprendizado dividida em dois componentes principais: o componente de aprendizado indutivo e o componente de aprendizado dedutivo. O primeiro se volta para o componente indutivo e utiliza os métodos de aprendizado por casos de treinamento, descritos para o MNC, para refinar o conhecimento das redes neurais. O componente dedutivo faz use de uma estratégia de reorganização das redes neurais, para gerar novos modelos de resolução de problemas. 0 sistema implementado, denominado HYCONES (the HYbrid CONnectionist Expert System) é apresentado corn detalhamento da integragao entre os paradigmas simbOlico e conexionista. Para validar HYCONES, foram construfdas e testadas 5 versoes de uma base de conhecimento hfbrida para o diagnostico das 3 cardiopatias congenitas mais freqfjentes no Institute de Cardiologia - RS. A primeira delas (B1) baseou-se no conhecimento extrafdo de 22 especialistas em cardiologia pediatrica, de 4 instituicaes brasileiras. Os grafos medios de conhecimento destes especialistas foram diretamente traduzidos para redes neurais MNC, dando origem a primeria versao do sistema. A segunda versao (B2) baseou-se no refinamento da base de conhecimento da versao B1, atraves da reorganizacao e treinamento das redes neurais dos especialistas, corn aplicagao dos metodos de aprendizado dedutivo. A terceira versão (B3) baseou-se no conhecimento de 23 não especialistas, cardiologistas gerais, na sua maioria oriundos do Institute de Cardiologia - RS. A mesma técnica aplicada para a construção da versão B1 foi utilizada para construir esta versão da base de conhecimento hibrida. A quarta versao (B4) corresponde ao refinamento e reorganização da base de conhecimento da versão B3, com as mesmas técnicas de aprendizado dedutivo, aplicadas na construção da versão B2. Finalmente, na quinta versão (B5), a base de conhecimento foi construída, automaticamente, a partir de base de casos, valendo-se do método de explicitação automática de conhecimento do MNC. Os resultados obtidos na validação do sistema comprovaram a eficiência dos métodos de explicitação automática de conhecimento, a partir do banco de casos e de aprendizado dedutivo, por reorganização e treinamento das redes neurais. As principais contribuições deste trabalho foram: a definição do mecanismo para integração de frames a redes neurais; a especificação dos métodos de aquisição automática do conhecimento, a partir de um banco de casos e seu posterior treinamento e aprendizado, por métodos indutivos e dedutivos; finalmente, a incorporação das medidas de especificidade e sensibilidade no processo de explicitação automática do conhecimento do banco de casos. / This dissertation describes a hybrid model for classification expert systems - HYCONES (HYbrid CoNnectionist Expert System), which integrates the knowledge representation mechanism of frames and neural networks, incorporating qualities inherent to each of the knowledge representation paradigms. The mechanism of frames provides flexible constructs for modeling the domain knowledge. The neural networks provide the means for automatic knowledge acquisition and enable the implementation of deductive and inductive learning. The frames mechanism makes use of semantic modeling concepts to make the new model more generic and to provide flexible constructs for knowledge representation. It employs two basic types of frames to represent the knowledge involved in classification problems: finding frames and diagnosis frames. The finding frames describe the elementary objects of the application world that can influence in the detection of certain diagnoses. Diagnosis frames, on the other hand, are described in terms of findings and correspond to the descriptions of the classification problems addressed by the expert system. The Combinatorial Neural Model (CNM) was the architecture chosen for the neural network environment. This model is characterized by its capacity of learning through the recognition of regularities observed in the outside world. The integration of the knowledge representation mechanism of frames with the CNM happens at the diagnosis frames level. The diagnosis frames store symbolic representations of knowledge contained in the neural networks. The system's inference machine operates under an integrated manner with both knowledge representation paradigms. The learning machine provides the means for automatic knowledge acquisition and for knowledge refinement. It is composed by two modules: the inductive and the deductive learning component. The first one uses CNM methods to refine the neural network knowledge. The latter follows a strategy that reorganizes the neural networks, generating new problem solving models. The implemented HYCONES system is presented, detailing the integration between the symbolic and the connectionist approach. To validate HYCONES, five versions of a hybrid knowledge base (HKB) for the diagnosis of the three most common congenital heart diseases at the Institute de Cardiologia - RS were constructed and tested. The first version of the HKB, named B1 for validation purposes, is based on the mean knowledge graphs of 22 experts in congenital heart diseases, from four important Brazilian cardiological institutions. These KG were directly translated to CNM neural networks. The second version of the HKB, named B3 for validation purposes, contains the mean knowledge graphs of 23 non-experts, i.e., general cardiologists, from the Institute of Cardiology of RS. The same method used to built B1 was invoked here: direct translation from the knowledge graphs to CNM neural networks. The second and fourth versions of HKB, named B2 and B4 respectively, consist of trained versions of B1 and B3. To create these trained versions, the CNM neural networks of B1 and B3 were submitted to deductive learning procedures. Finally, the fifth version of HKB, B5, was automatically created, based on the training of the case database. The results obtained and presented in this document point out to the efficiency of the automatic knowledge acquisition methods and the deductive learning techniques implemented, through the reorganization and training of the neural networks. The main contributions of this research are the design of a mechanism for the integration of frames and neural networks, the specification of the learning methods for the hybrid model and the analysis of factors such as specificity and sensibility in the process of automatic knowledge acquisition.

No estudo de sistemas híbridos orgânico-inorgânico, o uso de materiais como polímeros conjugados e óxidos de metal de transição tem despertado grande interes- se. Em particular, destacam-se sistemas compostos de tiofenos e óxido de titânio, que encontram uma importante aplicação em células solares. Para um melhor entendimento da interação entre os dois sistemas, torna-se necessário conhecer a organização do polímero sobre o substrato inorgânico. Desse modo, investigamos neste trabalho a formação da interface entre oligômeros de tiofeno e a superfície (101) de TiO2-anatase utilizando um enfoque de multiformalismo, que inclui simulações de dinâmica molecular clássica, e uma combinação de cálculos de primeiros princípios segundo Hartree-Fock e Teoria do Funcional da Densidade (DFT) para a determinação de propriedades estruturais e eletrônicas. A deposição de oligômeros de tiofeno sobre TiO2, constituindo sistemas de milhares de átomos, foi simulada por meio de dinâmica molecular clássica. Como requisito do cálculo clássico para estes sistemas, realizamos a reparametrização do campo de forças Universal tanto para os oligômeros, cujas estruturas não são bem descritas pelos campos de força padrões, como para o cristal e a superfície de TiO2. Foi observada a formação de filmes desordenados e densos de quatertiofeno, com a presença de uma maioria de moléculas de orientação quase perpendicular em relação ao plano superficial. Na camada de interface também se encontram moléculas dispostas paralelamente ao substrato, aumentando o contato entre os sistemas orgânico e inorgânico. A deposição de oligômeros isolados de quatertiofeno e de hexatiofeno mostra ainda que as moléculas se dispõem paralelas na superfície, alinhadas segundo direções de periodicidade dos átomos da superfície. Estudamos desta forma as propriedades eletrônicas de um sistema composto de politiofeno sobre TiO2, com o polímero paralelo na superfície e disposto na direção preferencial, através de um formalismo ab initio DFT. Apesar do tratamento DFT apresentar problemas conhecidos quanto na definição do gap, o que é mais relevante ainda no nosso caso de sistemas híbridos, os resultados revelam um deslocamento do topo da banda de valência do material orgânico em relação ao inorgânico. Isto possibilita o aprisionamento de um buraco no polímero, condição necessária para o uso deste tipo de sistema em células fotovoltaicas. Verifica-se ainda o acoplamento entre átomos de enxofre do politiofeno e de oxigênio do TiO2 através da presença de um estado associado a uma densidade eletrônica que se estende do polímero na superfície. Nossos resultados indicam assim um bom acoplamento eletrônico da superfície (101) de TiO2-anatase com politiofenos. / In the study of organic-inorganic hybrid systems, the use of materials such as conjugated polymers and transition metal oxides has attracted great interest. In particular, it is worth mentioning systems composed by thiophenes and titanium oxide, which have an important application in solar cells. For a better understand- ing of the interaction between these systems, it is necessary to know the polymer organization over the inorganic substrate. Therefore, we investigated in this work the formation of the interface between thiophene oligomers and the (101) surface of TiO2-anatase by means of a multi-formalism approach, which includes classical molecular dynamics simulations, and a combination of ¯rst principles calculations based on Hartree-Fock and Density Functional Theory (DFT) for structural and electronic properties. The simulation of deposition of thiophene oligomers on TiO2, which demands systems with thousands of atoms, was performed by classical molecular dynamics. As a prerequisite for the classical calculation for these systems, we performed a re-parameterization of the Universal force ¯eld for the oligomers, whose structures are not well described by standard force ¯elds, and for the TiO2 bulk and surface. We observed the formation of disordered and dense quaterthiophene ¯lms, with presence of a majority of molecules oriented almost perpendicularly to the surface plane. In the ¯rst interfacial layer we ¯nd also molecules oriented parallel to the sub- strate, which increases the contact between the organic and the inorganic systems. The deposition of isolated quaterthiophene and sexithiophene oligomers resulted in molecules disposed parallel to the surface and aligned along directions of periodicity of the surface atoms. We therefore studied the electronic properties of a system composed of poly- thiophene on TiO2, with the polymer parallel to the surface and oriented along a preferential direction, by means of DFT formalism. Although DFT treatments present known problems in the de¯nition of the energy gap, even of more relevance in our case of hybrid systems, the results for the occupied states revealed a sizeable displacement of the top of the valence band of one system with respect to the other. The misalignment will prevent the passage of a hole from the polymer to the oxide, providing in this way the necessary condition for the use of this type of system in solar cells. It was also seen electronic coupling between sulfur atoms from polythio- phene, and oxygen atoms from TiO2 through the presence of a state associated with an electronic density extended from the polymer to the surface. Our results thus indicate there is good electronic coupling between the (101) surface of TiO2-anatase and polythiophenes.