<p dir="ltr">This thesis presents a comprehensive exploration of advanced optimization strategies for addressing mixed-integer optimal control problems (MIOCPs) in aerospace applications, emphasizing the enhancement of convergence robustness, computational efficiency, and accuracy. The research develops a broad spectrum of optimization methodologies, including multi-phase approaches, parallel computing, reinforcement learning (RL), and distributed algorithms, to tackle complex MIOCPs characterized by highly nonlinear dynamics, intricate constraints, and discrete control variables.</p><p dir="ltr">Through discretization and reformulation, MIOCPs are transformed into general quadratically constrained quadratic programming (QCQP) problems, which are then equivalently converted into rank-one constrained semidefinite programs problems. To address these, iterative algorithms are developed specifically for solving such problems. Initially, two iterative search methods are introduced to achieve convergence: one is a hybrid alternating direction method of multipliers (ADMM) designed for large-scale QCQP problems, and the other is an iterative second-order cone programming (SOCP) algorithm developed to achieve global convergence. Moreover, to facilitate the convergence of these iterative algorithms and to enhance their solution quality, a multi-phase strategy is proposed. This strategy integrates with both the iterative ADMM and SOCP algorithms to optimize the solving of QCQP problems, improving both the convergence rate and the optimality of the solutions. To validate the effectiveness and improved computational performance of the proposed multi-phase iterative algorithms, the proposed algorithms were applied to several aerospace optimization problems, including six-degree-of-freedom (6-DoF) entry trajectory optimization, fuel-optimal powered descent, and multi-point precision landing challenges in a human-Mars mission. Theoretical analyses of convergence properties along with simulation results have been conducted, demonstrating the efficiency, robustness, and enhanced convergence rate of the optimization framework.</p><p dir="ltr">However, the iteration based multi-phase algorithms primarily guarantee only local optima for QCQP problems. This research introduces a novel approach that integrates a distributed framework with stochastic search techniques to overcome this limitation. By leveraging multiple initial guesses for collaborative communication among computation nodes, this method not only accelerates convergence but also enhances the exploration of the solution space in QCQP problems. Additionally, this strategy extends to tackle general nonlinear programming (NLP) problems, effectively steering optimization toward more globally promising directions. Numerical simulations and theoretical proofs validate these improvements, marking significant advancements in solving complex optimization challenges.</p><p dir="ltr">Following the use of multiple agents in QCQP problems, this research expand this advantage to address more general rank-constrained semidefinite programs (RCSPs). This research developed a method that decomposes matrices into smaller submatrices for parallel processing by multiple agents within a distributed framework. This approach significantly enhances computational efficiency and has been validated in applications such as image denoising, showcasing substantial improvements in both efficiency and effectiveness.</p><p dir="ltr">Moreover, to address uncertainties in applications, a learning-based algorithm for QCQPs with dynamic parameters is developed. This method creates high-performing initial guesses to enhance iterative algorithms, specifically applied to the iterative rank minimization (IRM) algorithm. Empirical evaluations show that the RL-guided IRM algorithm outperforms the original, delivering faster convergence and improved optimality, effectively managing the challenges of dynamic parameters.</p><p dir="ltr">In summary, this thesis introduces advanced optimization strategies that significantly enhance the resolution of MIOCPs and extends these methodologies to more general issues like NLP and RCSP. By integrating multi-phase approaches, parallel computing, distributed techniques, and learning methods, it improves computational efficiency, convergence, and solution quality. The effectiveness of these methods has been empirically validated and theoretically confirmed, representing substantial progress in the field of optimization.</p>

Le syndrome néphrotique idiopathique (SNI) est une néphropathie glomérulaire définie par une protéinurie massive associée à une hypoalbuminémie, sans lésions inflammatoires rénales, ni dépôts de complexes immuns circulants. Les travaux réalisés au cours de ma thèse concernent l’étude du rôle potentiel du gène c-mip dans la physiopathologie du SNI.Dans un premier temps, nous avons étudié la physiopathologie moléculaire de l’association maladie de Hodgkin et SNI. Nous avons démontré que cette association était liée à une forte induction de c-mip à la fois dans les cellules de Reed Sternberg (dont la présence signe le lymphome hodgkinien) et les podocytes qui sont des cellules spécialisées du glomérule rénal (Audard, et al. 2010). Nous avons montré que l’induction de c-mip résultait d’un défaut quantitatif et/ou qualitatif du gène Fyn, à la fois chez les patients et dans un modèle de souris déficiente en Fyn. Nous avons trouvé que c-mip était fortement induit dans les podocytes au cours du SNI ainsi que dans la glomérulopathie extramemenbraneuse (GEM). La surexpression de c-mip par transgénèse chez la souris déclenche une protéinurie néphrotique dont le mécanisme implique une rupture, médiée par c-mip, de la voie de signalisation de la néphrine (Science Signaling, 2010 co-auteur). L’étude de la néphrite de Heyman, le modèle expérimental de la GEM humaine, a permis de montrer que l’induction de c-mip coincidait avec l’apparition de la protéinurie et était associée à l’inhibition de l’activité RhoA, à une perte de la synaptopodine, à une diminution du VEGF tandis que l’expression de la DAPK (death-associated protein kinase) est fortement augmentée (Audard et al, manuscrit soumis 1). Nous avons recherché si l’hypogammaglobulinémie au cours du SNI était associée à des anomalies fonctionnelles des lymphocytes B (LB). Nous avons trouvé que c-mip interagit avec la sous unité régulatrice de la PI3 kinase et empêche la dissociation de la sous unité catalytique, p110, nécessaire à l’activation de la PI3 kinase. Enfin, l’expression de l’IL 21, une cytokine–clé secrétée par les lymphocytes T et intervenant dans la commutation isotypique, était fortement réduite dans le SNI (Audard et al, manuscrit en préparation 2). Ces résultats donnent un éclairage nouveau sur la physiopathologie moléculaire du SNI et suggèrent un rôle crucial de c-mip dans les anomalies lymphocytaires et podocytaires observées chez les patients / Idiopathic nephrotic syndrome comprises several podocyte diseases of unknown origin, affecting the glomerular podocyte, which plays a key role in controlling the permeability of the kidney filter to proteins. It is characterized by massive proteinuria and hypoalbuminemia, with no inflammatory lesions or cell infiltration. This works focused on the potential role of c-mip in the pathogenesis of INS. We showed that occurrence of minimal change nephrotic syndrome in the course of Hodgkin lymphoma (cHL-MCNS) is closely related to the induction of c-mip in both Hodgkin-Reed Sternberg cells and podocytes (Audard, et al. 2010), which is caused by a qualitative and/or quantitative defect in Fyn in both HRS and podocytes cells. We found that c-mip is upregulated in podocytes of patients with membranous nephropathy (MN). Transgenic mice overproducing c-mip in the podocytes developed heavy proteinuria without morphological alterations, inflammatory lesions or cell infiltrations. We showed that c-mip turned off podocyte proximal signaling by preventing the interaction between Fyn and nephrin, resulting in the inhibition of nephrin signaling pathway (Science signaling, 2010 coauthor). Moreover, the induction of c-mip in passive type Heymann nephritis (the experimental model of MN) was concomitant to proteinuria occurrence and is associated with reduction of RhoA activity, downregulation of synaptopodin and VEGF expression whereas DAPK expression is significantly increased (Audard et al manuscript submitted 1).We demonstrated that hypogammaglobulinemia, a common feature in INS patients, may result from a defect in B lymphocytes. We found that c-mip interacts with p85 regulatory subunit and prevent its dissociation from p110 catalytic subunit, resulting in inactivation of PI3 kinase. Finally, the expression of IL21, a key cytokine involved in class switching recombination, is repressed in active phases of INS, which may contribute for immunoglobulin disorders commonly observed in these patients (Audard et al manuscript in progress 2).Altogether, these results suggest that c-mip is a major player of lymphocyte and podocytes dysfunction observed in patients with INS

Ces travaux de thèse portent sur l'analyse des mélanges réels et synthétiques complexes composés de petites molécules à l'aide de la RMN HRMAS. Dans une première partie, une approche RMN HRMAS basée sur l'analyse métabolomique en combinaison avec des techniques de reconnaissance des formes (PCA et O-PLS-DA) a été appliquée pour le diagnostic des lésions thyroïdiennes indéterminées et étudier également les effets biologiques négatifs des nanoparticules d'aluminium sur pseudomonas brassicacearum. Dans une seconde partie, nous avons étudié la RMN chromatographique en utilisant la silice comme matrice de support qui pourrait fournir une alternative rapide et complète de la LC pour la caractérisation de mélanges complexes. En outre, l'exigence de la suppression du signal dans l'extrait de plantes naturelles et d'hydrocarbures aromatiques conduit à l'élaboration d'une méthode rapide et précise en utilisant des polymères à empreintes moléculaires avec une excellente sélectivité. La sélectivité des polymères à empreintes moléculaires à travers la capture d'une cible moléculaire spécifique est exploitée ici pour éliminer efficacement les signaux RMN. / This thesis work deals with the analysis of natural and synthetic complex mixtures composed of small molecules using HRMAS NMR. In a first part, an integrated HRMAS-NMR based metabolomic analysis in combination with pattern recognition techniques (PCA and O-PLS-DA) has been applied for the diagnosis of indeterminate thyroid lesions and also studied the potential adverse biological effects of aluminium nanoparticles on pseudomonas brassicacearum. In a second part we investigated that chromatographic NMR using silica as the matrix support could provide a quick alternative and complement to LC for the characterization of complex mixtures. In addition, requirement for signal suppression in natural plant extract and aromatic hydrocarbons led to the development of a rapid and accurate method using molecularly imprinted polymers with excellent selectivity. The selectivity of Molecularly Imprinted polymers towards capturing a specific molecular target is exploited here to efficiently remove NMR signals.

O preparo da amostra é uma das etapas mais importantes em toda a análise química. O isolamento e a concentração dos componentes da amostra são cruciais e busca-se sempre que essas etapas sejam as mais simples e consumam o mínimo possível de tempo e reagentes. Nos últimos anos, um tipo de material tem se mostrado bastante útil para análises químicas a partir de fluidos biológicos, os polímeros molecularmente impressos (MIPs). Os MIPs são sintetizados por reações de polimerização, na presença de uma molécula molde (template). A molécula molde se liga aos monômeros funcionais do polímero durante a reação de polimerização e permanece ligada à superfície das cadeias poliméricas quando a reação se completa. Terminada a polimerização, realiza-se a completa lavagem das moléculas molde, assim, restam na superfície polimérica cavidades tridimensionais complementares à molécula empregada como molde. Essas cavidades permitem a ligação reversível e preferencial da molécula molde ou outras com estrutura química semelhante. A Cannabis sativa é a droga ilícita mais consumida em todo o mundo e nos últimos anos muita atenção tem se voltado a seus efeitos toxicológicos no corpo humano e a aplicações medicinais. Nesta dissertação, foi sintetizado um MIP com a molécula molde catequina para a extração e posterior análise por LC-MS/MS dos canabinóides &Delta;9-tetrahidrocanabinol (THC), 11-hidroxi-&Delta;9-tetrahidrocannabinol (THC-OH) e 11-nor-&Delta;9-tetrahidrocannabinol-9-ácido carboxílico (THC-COOH) em amostras de urina. O MIP produzido foi empacotado em microdispositivo e empregado no preparo das amostras de urina por microextração por sorvente empacotado (MEPS). O método desenvolvido apresentou boa linearidade (valores de r de 0,977 para o THC e 0,994 para THC-OH e THC-COOH). Os limites de detecção e de quantificação foram respectivamente de 5 ng mL-1 e 20 ng mL-1, para os compostos THC e THC-OH, na faixa linear de 25 a 250 ng mL-1. Para o composto THC-COOH os limites de detecção e quantificação alcançados foram de 1 ng mL-1 e 5 ng mL-1, respectivamente, na faixa linear de 5 a 170 ng mL-1. O método apresentou valores razoáveis de precisão entre 3,2% (THC-COOH) e 25,1% (THC) e de exatidão, que variou entre -18,4 e 17,4 (ambos para o THC). O MIP empregado no preparo da amostra mostrou-se mais seletivo e específico do que materiais normalmente empregados para a extração dos canabinoides das amostras de urina, além de a técnica de extração por MEPS apresentar baixo consumo de solventes e amostra para a extração dos analitos e posterior análise por LC-MS/MS. / The sample preparation is one of the most important steps in every chemical analysis. The isolation and concentration of the sample components are crucial and it is always sought that these steps are simple and consume the lowest amount of time and reagents. In the recent years, a type of material has proved to be very useful for chemical analyzes of biological fluids, the molecularly imprinted polymers (MIPs). MIPs are synthesized by polymerization reactions in the presence of a template molecule. The template molecule binds to the functional monomers of the polymer during the polymerization reaction and remains bonded to the surface of the polymeric chains after the reaction is complete. After the polymerization is finished, the complete washing of the template molecules is carried out, thus, three-dimensional cavities, complementary to the molecule used as a template, remain on the polymer surface. These cavities allow the reversible and preferential bonding of the template molecule or others with similar chemical structure. Cannabis sativa is the most commonly consumed illicit drug in the world and in recent years much attention has focused on its toxicological effects on human body and for medical applications. In this master dissertation, a MIP was synthesized with the catechin molecule as template, for extraction and subsequent analysis by LC-MS/MS of the cannabinoids &Delta;9-tetrahydrocannabinol (THC), 11-hydroxy-&Delta;9-tetrahydrocannabinol (THC-OH), and 11-nor-&Delta;9-tetrahydrocannabinol-9-carboxylic acid (THC-COOH) in urine samples. The MIP produced was packed in a microdevice and used in the preparation of the urine samples by microextraction by packed sorbent (MEPS). The developed method showed good linearity (r values of 0.977 for THC and 0.994 for THC-OH and THC-COOH). The detection and quantification limits were respectively 5 ng mL-1 and 20 ng mL-1 for THC and THC-OH in the linear range from 25 to 250 ng mL-1. For the compound THC-COOH the limits of detection and quantification achieved were 1 ng mL-1 and 5 ng mL-1, respectively, in the linear range from 5 to 170 ng mL-1. The method presented reasonable values of precision, between 3.2% (for THC-COOH) and 25.1% (for THC) and displayed accuracy ranging from -18.4 to 17.4 (both for THC). The MIP used in the sample preparation was more selective and specific than other materials usually employed for the extraction of the cannabinoids from the urine samples. The MEPS technique also showed low consumption of solvents and sample for sample preparation, extraction of analytes and subsequent analysis by LC-MS/MS.

Este trabalho aborda o problema multiestágio de planejamento e programação da produção em indústrias cervejeiras. O processo de fabricação de cerveja pode ser dividido em duas etapas principais: preparação do líquido e envase. A primeira etapa ocorre, na maior parte do tempo, dentro de tanques de fermentação e maturação. A segunda ocorre nas linhas de envase, podendo ter início assim que o líquido estiver pronto nos tanques. O tempo de preparação do líquido demora vários dias, enquanto que na maioria das indústrias de bebidas carbonatadas este tempo é de no máximo algumas horas. O objetivo deste estudo é obter planos de produção viáveis que visam otimizar as decisões de programação envolvidas nestes processos. Visitas a cervejarias no Brasil e em Portugal foram realizadas para uma maior familiaridade do processo de produção e dados foram coletados. Modelos de programação inteira mista para representar o problema foram desenvolvidos, baseados em abordagens CSLP (The Continuous Setup Lot-Sizing Problem), GLSP (General Lot Sizing and Scheduling Problem), SPL (Simple Plant Location Problem) e ATSP (Asymmetric Travelling Salesman Problem). Os resultados mostram que os modelos são coerentes e representam adequadamente o problema, entretanto, mostram-se difíceis de serem resolvidos na otimalidade. Esta dificuldade de resolução dos modelos motivou o desenvolvimento de procedimentos MIP-heurísticos, como também de uma metaheurística GRASP (Greedy Randomized Adaptive Search Procedure). As soluções obtidas pelos procedimentos heurísticos são de boa qualidade, quando comparadas ao melhor limitante inferior encontrado por meio da resolução dos modelos matemáticos. Os testes computacionais foram realizados utilizando instâncias geradas com base em dados reais. / This study deals with the multistage lot-sizing and scheduling problem in breweries. The brewing process can be divided into two main stages: preparation and filling of the liquid. The first stage occurs most of the time in fermentation and maturation tanks. The second stage occurs in the filling lines and it can start as soon as the liquid gets ready. The preparation time of the liquid takes several days, while in the carbonated beverage industries this time is at most a few hours. The purpose of this study is to obtain feasible production plans aimed at optimizing the decisions involved in these processes. Visits to brewery industries in Brazil and Portugal were held to a greater familiarity of the production process and data were collected. Mixed integer programming models have been developed to represent the problem, based on approaches for the CSLP (The Continuous Setup Lot-Sizing Problem), GLSP (General Lot Sizing and Scheduling Problem), SPL (Simple Plant Location Problem) and ATSP (Asymmetric Travelling Salesman Problem). The results show that the models are consistent and adequately represent the problem; however, they are difficult to be solved at optimality. This motivated the development of MIP-heuristic procedures, as well as a meta-heuristic GRASP (Greedy Randomized Adaptive Search Procedure). The obtained solutions by the heuristics are of good quality, when compared to the best lower bound found by solving the mathematical models. The tests were conducted using generated instances based on real data.

O preparo da amostra é uma das etapas mais importantes em toda a análise química. O isolamento e a concentração dos componentes da amostra são cruciais e busca-se sempre que essas etapas sejam as mais simples e consumam o mínimo possível de tempo e reagentes. Nos últimos anos, um tipo de material tem se mostrado bastante útil para análises químicas a partir de fluidos biológicos, os polímeros molecularmente impressos (MIPs). Os MIPs são sintetizados por reações de polimerização, na presença de uma molécula molde (template). A molécula molde se liga aos monômeros funcionais do polímero durante a reação de polimerização e permanece ligada à superfície das cadeias poliméricas quando a reação se completa. Terminada a polimerização, realiza-se a completa lavagem das moléculas molde, assim, restam na superfície polimérica cavidades tridimensionais complementares à molécula empregada como molde. Essas cavidades permitem a ligação reversível e preferencial da molécula molde ou outras com estrutura química semelhante. A Cannabis sativa é a droga ilícita mais consumida em todo o mundo e nos últimos anos muita atenção tem se voltado a seus efeitos toxicológicos no corpo humano e a aplicações medicinais. Nesta dissertação, foi sintetizado um MIP com a molécula molde catequina para a extração e posterior análise por LC-MS/MS dos canabinóides &Delta;9-tetrahidrocanabinol (THC), 11-hidroxi-&Delta;9-tetrahidrocannabinol (THC-OH) e 11-nor-&Delta;9-tetrahidrocannabinol-9-ácido carboxílico (THC-COOH) em amostras de urina. O MIP produzido foi empacotado em microdispositivo e empregado no preparo das amostras de urina por microextração por sorvente empacotado (MEPS). O método desenvolvido apresentou boa linearidade (valores de r de 0,977 para o THC e 0,994 para THC-OH e THC-COOH). Os limites de detecção e de quantificação foram respectivamente de 5 ng mL-1 e 20 ng mL-1, para os compostos THC e THC-OH, na faixa linear de 25 a 250 ng mL-1. Para o composto THC-COOH os limites de detecção e quantificação alcançados foram de 1 ng mL-1 e 5 ng mL-1, respectivamente, na faixa linear de 5 a 170 ng mL-1. O método apresentou valores razoáveis de precisão entre 3,2% (THC-COOH) e 25,1% (THC) e de exatidão, que variou entre -18,4 e 17,4 (ambos para o THC). O MIP empregado no preparo da amostra mostrou-se mais seletivo e específico do que materiais normalmente empregados para a extração dos canabinoides das amostras de urina, além de a técnica de extração por MEPS apresentar baixo consumo de solventes e amostra para a extração dos analitos e posterior análise por LC-MS/MS. / The sample preparation is one of the most important steps in every chemical analysis. The isolation and concentration of the sample components are crucial and it is always sought that these steps are simple and consume the lowest amount of time and reagents. In the recent years, a type of material has proved to be very useful for chemical analyzes of biological fluids, the molecularly imprinted polymers (MIPs). MIPs are synthesized by polymerization reactions in the presence of a template molecule. The template molecule binds to the functional monomers of the polymer during the polymerization reaction and remains bonded to the surface of the polymeric chains after the reaction is complete. After the polymerization is finished, the complete washing of the template molecules is carried out, thus, three-dimensional cavities, complementary to the molecule used as a template, remain on the polymer surface. These cavities allow the reversible and preferential bonding of the template molecule or others with similar chemical structure. Cannabis sativa is the most commonly consumed illicit drug in the world and in recent years much attention has focused on its toxicological effects on human body and for medical applications. In this master dissertation, a MIP was synthesized with the catechin molecule as template, for extraction and subsequent analysis by LC-MS/MS of the cannabinoids &Delta;9-tetrahydrocannabinol (THC), 11-hydroxy-&Delta;9-tetrahydrocannabinol (THC-OH), and 11-nor-&Delta;9-tetrahydrocannabinol-9-carboxylic acid (THC-COOH) in urine samples. The MIP produced was packed in a microdevice and used in the preparation of the urine samples by microextraction by packed sorbent (MEPS). The developed method showed good linearity (r values of 0.977 for THC and 0.994 for THC-OH and THC-COOH). The detection and quantification limits were respectively 5 ng mL-1 and 20 ng mL-1 for THC and THC-OH in the linear range from 25 to 250 ng mL-1. For the compound THC-COOH the limits of detection and quantification achieved were 1 ng mL-1 and 5 ng mL-1, respectively, in the linear range from 5 to 170 ng mL-1. The method presented reasonable values of precision, between 3.2% (for THC-COOH) and 25.1% (for THC) and displayed accuracy ranging from -18.4 to 17.4 (both for THC). The MIP used in the sample preparation was more selective and specific than other materials usually employed for the extraction of the cannabinoids from the urine samples. The MEPS technique also showed low consumption of solvents and sample for sample preparation, extraction of analytes and subsequent analysis by LC-MS/MS.

Les capteurs et les biocapteurs sont des moyens d’analyse en plein essor à la fois rapides, sensibles, sélectifs et peu coûteux, applicables à des domaines très variés (agroalimentaire, environnement, biomédical…). Dans ce travail de recherche, nous nous sommes intéressés au développement de trois dispositifs à savoir un nez et une langue électroniques à base de systèmes multicapteurs pour l’analyse des odeurs et des saveurs ainsi que les biocapteurs. Les deux premiers dispositifs ont permis dans un premier temps de caractériser et de détecter les fraudes dans les produits de l’industrie agroalimentaire. Ainsi, nous sommes parvenus à détecter les pratiques frauduleuses dans l’huile d’argan par le nez et par la langue électroniques. Nous avons, par ailleurs, réussi à caractériser des miels de différentes origines géographiques et botaniques et à détecter l’adultération du miel pur par l’utilisation d’une langue électronique voltammétrique. Enfin, nous avons démontré l’efficacité des systèmes de nez et de langue électroniques à discriminer cinq marques de lait pasteurisé marocain. Des limitations du système de nez et de langue électroniques ont été révélées quant à la discrimination du lait pasteurisé (Jawda) en fonction des jours de stockage. Par contre, la fusion des données des deux systèmes moyennant un niveau d'abstraction intermédiaire a permis d’améliorer cette discrimination. Dans une deuxième étape, nous avons développé deux biocapteurs, le premier est basé sur l’utilisation de Polymère à Empreinte Moléculaire (MIP) dédié à la détection de la tétracycline dans le miel. Alors que le second est un immunocapteur conçu pour la détection de l’ochratoxine A. Le MIP, utilisé dans le premier cas, a été synthétisé à la surface d’électrodes en or par électropolymérisation des nanoparticules d'or fonctionnalisées par le p-aminothiophénol en présence de la tétracycline comme molécule empreinte. Dans le deuxième cas, un nouveau biocapteur capacitif basé sur l’utilisation d’un substrat de Nitrure de Silicium (Si3N4) combiné avec une nouvelle structure de nanoparticules magnétiques (MNPs) pour la détection de l’ochratoxine A a été conçu. En effet, Les MNPs possédant une terminaison carboxylique ont été liés de manière covalente à la monocouche auto-assemblée du silane-amine (3-aminopropyl triéthoxysilane APTES). Enfin, les anticorps anti-ochratoxine A ont été immobilisés sur les MNPs par liaison amide. Les performances des deux biocapteurs (limite de détection, sélectivité, reproductibilité) ont ensuite été évaluées / Sensors and Biosensors are rapid, sensitive, selective and low-cost analytical devices of growing interest for a wide range of application fields (e.g. food, environment, health …). This research focused on the development of three devices namely an electronic nose and tongue and electrochemical biosensors with applications in food analysis. The first two devices allowed the characterization and detection of frauds in the food field. Thus, we have been able to detect fraudulent practices in argan oil by using an electronic nose and tongue systems. In addition, the electronic tongue has successfully classified honeys of different geographical and botanical origins and detects the adulteration of pure honey. Finally, we have demonstrated the ability of the electronic nose and tongue systems to classify five brands of Moroccan pasteurized milk and to discriminate against them based on their storage days. In the second stage, we have developed two biosensors. The first one is based on a molecularly imprinted polymer (MIP) for the detection of tetracycline in honey. The second one is based on an immunosensor devoted for the detection of ochratoxin A. For the first biosensor, imprinted gold nanoparticles composites are assembled on Au surfaces by the electropolymerization of p-amino-thiophenol functionalized gold nanoparticles in the presence of the imprint molecule. In the second case, we have developed a novel capacitance electrochemical biosensor based on silicon nitride substrate (Si3N4) combined with a new structure of mangnetic nanoparticles (MNPs). Indeed, The MNPs with terminated carboxylic acid were covalently bonded to Si3N4 through a Self-Assembled Monolayers (SAMs) of the silane-amine (3-Aminopropyl) triethoxysilane (APTES). Finally anti-ochratoxin A antibodies were immobilized on MNPs by amide bonding. The performances of the two biosensors (limit of detection, selectivity, reproducibility) were then evaluated and the results are generally satisfactory

Este trabalho aborda o problema multiestágio de planejamento e programação da produção em indústrias cervejeiras. O processo de fabricação de cerveja pode ser dividido em duas etapas principais: preparação do líquido e envase. A primeira etapa ocorre, na maior parte do tempo, dentro de tanques de fermentação e maturação. A segunda ocorre nas linhas de envase, podendo ter início assim que o líquido estiver pronto nos tanques. O tempo de preparação do líquido demora vários dias, enquanto que na maioria das indústrias de bebidas carbonatadas este tempo é de no máximo algumas horas. O objetivo deste estudo é obter planos de produção viáveis que visam otimizar as decisões de programação envolvidas nestes processos. Visitas a cervejarias no Brasil e em Portugal foram realizadas para uma maior familiaridade do processo de produção e dados foram coletados. Modelos de programação inteira mista para representar o problema foram desenvolvidos, baseados em abordagens CSLP (The Continuous Setup Lot-Sizing Problem), GLSP (General Lot Sizing and Scheduling Problem), SPL (Simple Plant Location Problem) e ATSP (Asymmetric Travelling Salesman Problem). Os resultados mostram que os modelos são coerentes e representam adequadamente o problema, entretanto, mostram-se difíceis de serem resolvidos na otimalidade. Esta dificuldade de resolução dos modelos motivou o desenvolvimento de procedimentos MIP-heurísticos, como também de uma metaheurística GRASP (Greedy Randomized Adaptive Search Procedure). As soluções obtidas pelos procedimentos heurísticos são de boa qualidade, quando comparadas ao melhor limitante inferior encontrado por meio da resolução dos modelos matemáticos. Os testes computacionais foram realizados utilizando instâncias geradas com base em dados reais. / This study deals with the multistage lot-sizing and scheduling problem in breweries. The brewing process can be divided into two main stages: preparation and filling of the liquid. The first stage occurs most of the time in fermentation and maturation tanks. The second stage occurs in the filling lines and it can start as soon as the liquid gets ready. The preparation time of the liquid takes several days, while in the carbonated beverage industries this time is at most a few hours. The purpose of this study is to obtain feasible production plans aimed at optimizing the decisions involved in these processes. Visits to brewery industries in Brazil and Portugal were held to a greater familiarity of the production process and data were collected. Mixed integer programming models have been developed to represent the problem, based on approaches for the CSLP (The Continuous Setup Lot-Sizing Problem), GLSP (General Lot Sizing and Scheduling Problem), SPL (Simple Plant Location Problem) and ATSP (Asymmetric Travelling Salesman Problem). The results show that the models are consistent and adequately represent the problem; however, they are difficult to be solved at optimality. This motivated the development of MIP-heuristic procedures, as well as a meta-heuristic GRASP (Greedy Randomized Adaptive Search Procedure). The obtained solutions by the heuristics are of good quality, when compared to the best lower bound found by solving the mathematical models. The tests were conducted using generated instances based on real data.

O Brasil é o maior produtor de cana-de-açúcar (Saccharum officinarum L.) do mundo e a cultura continua em expansão na safra 2014/15. A expansão das áreas de plantio de cana-de-açúcar e o sistema de colheita mecanizada de cana crua tem levado ao aumento proporcional de inúmeras pragas da cultura, destacando-se o bicudo da cana, Sphenophorus levis, considerado praga primária e limitante da cultura. Inicialmente, essa espécie foi considerada restrita a região de Piracicaba, mas desde 2010 foi encontrada nos estados de São Paulo, Minas Gerais, Paraná, Mato Grosso do Sul, Mato Grosso e Goiás. Suas larvas abrem galerias nos rizomas, causando danos diretos nos tecidos dos colmos, sobretudo nas épocas mais secas do ano, o que ocasiona a morte das touceiras, falhas na rebrota, além de promover danos indiretos, como o aumento do número de plantas invasoras que competem com a cultura. Os métodos químico, mecânico, cultural e biológico já foram testados até o momento, no entanto, o sucesso no seu controle ainda não foi obtido e a população dessa praga continua crescendo e se tornando mais constante nos canaviais pelo Brasil. Dessa forma, estudar estratégias para a melhoria do manejo de S. levis é fundamental para o estabelecimento de métodos de controle eficientes no contexto do MIP. Diante disso, objetivou-se com esse trabalho: 1) Determinar a distribuição espacial e a flutuação populacional de S. levis; 2) Testar o manejo da palha pós-colheita nas linhas de plantio na ocorrência de S. levis e; 3) Avaliar a eficácia da aplicação de Beauveria bassiana na população de S. levis. Em relação à flutuação populacional de S. levis foi observada uma elevação linear da última quinzena de outubro para a primeira de novembro, com queda subsequente e, aumento populacional maior à partir da última quinzena de dezembro. O pico populacional se deu no mês de março, seguido por uma rápida redução a partir de abril. A proporção de fêmeas foi similar a de machos em todos os tratamentos. Os dados sugerem que não houve efeito dos tratamentos na distribuição de adultos do bicudo. Os resultados da flutuação populacional, da distribuição de frequência e distribuição espacial se complementam, pois, em hipótese, a distribuição dos adultos está diretamente relacionada com a flutuação populacional, uma vez que em altas densidades a distribuição foi agregada. Além disso, as estratégias de afastamento da palha da linha de plantio não interferiram nos níveis de infestação de S. levis, quando comparado ao tratamento que não houve manejo da palha e, essa tática utilizada conjuntamente à aplicação de B. bassiana também não influenciou na densidade populacional de adultos de S. levis no período avaliado. Diante disso, concluiu-se que a distribuição de S. levis em cana-de-açúcar apresenta o padrão agregado em altas densidades populacionais; o manejo da palha entre as linhas de cana-de-açúcar não reduz a densidade populacional de S. levis em locais com elevado histórico de infestação e, são necessários mais estudos para ajustar a dose de B. bassiana adequada e eficaz no controle de S. levis. / Brazil is the world\'s largest sugarcane producer and the cultivated area continues to expand in 2014/2015 season. This expansion, along with the mechanical harvesting of green sugarcane leads to a proportional increase in the number of crop pests, especially the sugarcane weevil, Sphenophorus levis, considered a primary and limiting pest in this crop. Initially, this species was restricted to the Piracicaba region, however, since 2010 it has been found in several regions of São Paulo State, and also in other states in the country, such as Minas Gerais, Paraná, Mato Grosso do Sul, Mato Grosso and Goiás. The larvae penetrate the basal part of the plants and the rhizomes, building irregular galleries and causing damage to stem tissues, mainly in the driest season, leading to the death of clumps, failures in regrowth and, consequently, indirect damage due to the increase in the incidence of invasive plants that compete with the crop. Chemical, mechanical, cultural and biological methods have already been tested, however, the success in pest control has not yet been achieved and the population of S. levis continues to grow, becoming more constant in Brazilian sugarcane fields. Thus, the study of strategies to enhance the management of S. levis is essential to adopt effective control methods in the IPM framework. In this sense, this study aims to: 1) Determine the spatial distribution and population dynamics of S. levis; 2) Evaluate the postharvest management of straw on planting rows in occurrence of S. levis and; 3) Evaluate the efficacy of B. bassiana application in the S. levis population in sugarcane. Concerning population dynamics of S. levis, it was observed a linear increase beginning in the last quarter of October until November, with a subsequent decrease and, a highest population increase was observed in the last half of December. The population peak occurred in March, followed by a fast decline in early April. The proportion of females to males was similar in all treatments. The data suggest that there was no effect of treatments on the distribution of adult weevils. The results of population dynamics, frequency distribution and spatial distribution are complementary, because, hypothetically, the distribution of adults is directly related to population dynamics, since at high densities the distribution was aggregated. In addition, strategies for straw management in the planting row did not affect the infestation levels of S. levis and this procedure along with the addition of B. bassiana did not influence population density of adults during the study period. Hence, it is concluded that the distribution of S. levis in sugarcane exhibits an aggregated pattern at high population densities. Straw management between the cane rows does not reduce population density of S. levis in areas with high historical infestation. Therefore, further studies are needed to adjust the appropriate and effective concentration of B. bassiana in the control of S. levis.

Prediction of contaminant transport in concrete subjected to short cyclical wetting and drying processes is integrally bound to prediction of the moisture flux. The concrete is unsaturated and the non-linear contaminant and moisture fluxes are not described by simple constant diffusion methods. This thesis presents, and partially justifies, a thermodynamic model for prediction of moisture movement in concrete, at all moisture contents commonly encountered. The wetting process is examined with Nuclear Magnetic Resonance (NMR) images during a simple absorption (sorptivity) experiment. Diffusivity functions are derived via a novel analytical approach and a universal diffusivity is suggested. Water sorption and desorption isotherms are measured on large concrete samples. van Genuchten’s retention function is successfully used to model the results. The unrelia-bility of the water sorption method at high moisture contents is illustrated by comparison with Mercury Intrusion Porosimetry (MIP). The BJH method is exploited to provide a methodology for estimating the water sorption isotherm from MIP. Mualem’s conductivity model is assessed with the water retention and NMR results. This thorough validation of the model yields a tortuosity parameter that is different to that commonly assumed. An analytical relationship between the sorptivity and the saturated permeability suggests the experimental the long-term unsaturated permeability overesti-mates the unsaturated conductivity function, and as such should be used judiciously when predicting unsaturated flow processes. Mualem’s conductivity model is further exploited to provide unsaturated air and vapour functions that are experimentally justified. The thermodynamic description of water movement and the hydraulic functions that are developed in the thesis are incorporated into T r inCet , a transient heat and mass trans-fer model based on the Finite Element Method (FEM). The complex coupled behaviour of air, liquid, vapour and temperature are well handled under a variety of common cyclical boundary conditions. The thesis presents all necessary experimental results required for validation of a com-plex, but easily described, model for moisture movement. It covers disparate ground to provide a powerful numerical model of unsaturated moisture movement in concrete under short-term cyclical processes.