Made available in DSpace on 2014-12-17T14:54:50Z (GMT). No. of bitstreams: 1 AndersonLOC.pdf: 927541 bytes, checksum: 73019c94d2a3901e38fc25a788ef32fd (MD5) Previous issue date: 2008-10-17 / This work addresses issues related to analysis and development of multivariable predictive controllers based on bilinear multi-models. Linear Generalized Predictive Control (GPC) monovariable and multivariable is shown, and highlighted its properties, key features and applications in industry. Bilinear GPC, the basis for the development of this thesis, is presented by the time-step quasilinearization approach. Some results are presented using this controller in order to show its best performance when compared to linear GPC, since the bilinear models represent better the dynamics of certain processes. Time-step quasilinearization, due to the fact that it is an approximation, causes a prediction error, which limits the performance of this controller when prediction horizon increases. Due to its prediction error, Bilinear GPC with iterative compensation is shown in order to minimize this error, seeking a better performance than the classic Bilinear GPC. Results of iterative compensation algorithm are shown. The use of multi-model is discussed in this thesis, in order to correct the deficiency of controllers based on single model, when they are applied in cases with large operation ranges. Methods of measuring the distance between models, also called metrics, are the main contribution of this thesis. Several application results in simulated distillation columns, which are close enough to actual behaviour of them, are made, and the results have shown satisfactory / Este trabalho aborda aspectos relacionados ? an?lise e ao desenvolvimento de controladores preditivos multivari?veis baseados em multi-modelos bilineares. O Controlador Preditivo Generalizado (GPC) Monovari?vel e Multivari?vel para o caso linear ? apresentado, sendo destacadas suas propriedades, caracter?sticas principais e aplica??es na ind?stria. O GPC bilinear, que ? o controlador base de todo o desenvolvimento desta Tese, ? apresentado atrav?s da abordagem da quasilineariza??o por degrau de tempo. Alguns resultados empregando este controlador s?o apresentados de forma a evidenciar o melhor desempenho do mesmo, quando comparado ao GPC linear, visto que os modelos bilineares representam melhor a din?mica de determinados processos. A quasilineariza??o por degrau de tempo, devido ao fato de ser uma aproxima??o, provoca um erro de predi??o, que limita o desempenho deste controlador ? medida que seu horizonte de predi??o aumenta. Devido ao referido erro de predi??o, o GPC bilinear com compensa??o iterativa ? mostrado de forma a minimizar o referido erro, buscando um melhor desempenho que o GPC bilinear cl?ssico. Alguns resultados utilizando o algoritmo de compensa??o iterativa s?o mostrados. O emprego dos multi-modelos ? abordado nesta Tese, buscando suprir a defici?ncia existente em controladores baseados em modelo ?nico, quando os mesmos s?o aplicados em processos com grandes faixas de opera??o. Formas de mensura??o de dist?ncia entre modelos, tamb?m chamadas de m?tricas, consistem na principal contribui??o desta Tese. Diversos resultados de aplica??o em colunas de destila??o simuladas, que se aproximam bastante do comportamento real das mesmas, foram realizados, e os resultados se mostraram satisfat?rios

Les bassins versants nord-méditerranéens sont fréquemment soumis à des crues extrêmes liées à des précipitations convectives intenses et aux caractéristiques hydrologiques locales. La région méditerranéenne est considérée comme une des régions les plus affectées par le réchauffement climatique, ce qui laisse présager des changements dans le cycle hydrologique. L’objectif de cette thèse CIFRE est d’évaluer les impacts du changement climatique sur les précipitations extrêmes à travers une méthode dite de « futurisation », dans laquelle une fonction de transfert est construite en comparant la distribution des quantiles de précipitations du climat présent et futur. Les impacts du changement climatique sur les précipitations extrêmes sont évalués à travers les simulations à haute résolution EMCORDEX. L’exercice se focalise sur le bassin versant de l’Orbieu dans le sud de la France. La méthode de futurisation est appliquée à six épisodes majeurs de précipitations ayant généré des crues éclair. Les impacts hydrologiques des équivalents statistiques futurs des épisodes de précipitations sont ensuite évalués à travers un modèle hydrologique évènementiel conceptuel. Une estimation des changements d’humidité du sol liés au changement climatique est réalisée et couplée à la quantification des impacts hydrologiques. Le choix d’une modélisation hydrologique conceptuelle a été motivé par ses futures applications opérationnelles. Les conséquences de ce choix sont évaluées à travers une comparaison avec un modèle hydrologique à base physique. Ce dernier est mis en place grâce à une caractérisation du fonctionnement hydrologique du bassin versant de l’Orbieu appuyée par plusieurs campagnes de terrain. / Northern mediterranean meso-scale river catchments are submitted to extremes floods events linked to intense convective precipitation and local hydrologic features. The Mediterranean region is known to be one of the most affected areas by global warming, and it is likely that changes can be expected in the hydrological cycle. The aim of this CIFRE thesis is to assess the climate change impacts on extreme precipitation events using a so-called “futurization” method, in which a transfer function is built by comparing the quantiles of distribution for both present and future climate precipitation. The climate change impact on extreme precipitation events is assessed over high-resolution EMCORDEX simulations. The focus is on the Orbieu catchment located in southwestern France. The futurization method is applied to six major events of precipitation that trigger flash floods. The hydrological impacts of those future statistical counterpart precipitation events are therefore assessed through a conceptual event-based hydrological model. An assessment of soil moisture changes under climate change is performed and coupled to the hydrological impact quantification. The conceptual hydrological model chosen, have been motivated by its future operational applications. The consequences of that choice are assessed through a comparison to a physically based hydrological model. It has been implemented through the hydrological functioning caracterisation of the Orbieu catchment supported by several field campaigns.