Bayesian Optimization (BO) methods are often used to optimize an unknown function f(���) that is costly to evaluate. They typically work in an iterative manner. In each iteration, given a set of observation points, BO algorithms select k ��� 1 points to be evaluated. The results of those points are then added to the set of observations and the procedure is repeated until a stopping criterion is met. The goal is to optimize the function f(���) with a small number of experiment evaluations. While this problem has been extensively studied, most existing approaches ignored some real world constraints frequently encountered in practical applications. In this thesis, we extend the BO framework in a number of important directions to incorporate some of these constraints. First, we introduce a constrained BO framework where instead of selecting a precise point at each iteration, we request a constrained experiment that is characterized by a hyper-rectangle in the input space. We introduce efficient sequential and non-sequential algorithms to select a set of constrained experiments that best optimize f(���) within a given budget. Second, we introduce one of the first attempts in batch BO where instead of selecting one experiment at each iteration, a set of k > 1 experiments is selected. This can significantly speedup the overall running time of BO. Third, we introduce scheduling algorithms for the BO framework when: 1) it is possible to run concurrent experiments; 2) the durations of experiments are stochastic, but with a known distribution; and 3) there is a limited number of experiments to run in a fixed amount of time. We propose both online and offline scheduling algorithms that effectively handle these constraints. Finally, we introduce a hybrid BO approach which switches between the sequential and batch mode. The proposed hybrid approach provides us with a substantial speedup against sequential policies without significant performance loss. / Graduation date: 2013

O presente estudo teve por finalidade testar a hipótese de que o uso de uma taxa de recirculação variável do lixiviado de um reator tratando a fração orgânica dos resíduos sólidos urbanos domésticos, em um sistema em batelada híbrido, seria capaz de acelerar a degradação deste substrato em comparação com um sistema idêntico, porém que operasse com taxa de recirculação constante. Dois conjuntos de reatores foram operados, um com taxa de recirculação constante (sistema controle) e o outro com taxa de recirculação variável (sistema experimental). Os reatores foram mantidos em sala climatizada, com temperatura mantida em 35 ºC '+ OU -' 1 ºC. Cada conjunto era composto de um reator de sólidos, preenchido com o resíduo a ser digerido misturado a lixiviado de aterro sanitário, e um filtro anaeróbio, destinado a tratar o lixiviado que saía do primeiro. O substrato utilizado foi a fração orgânica dos resíduos sólidos urbanos domésticos padronizada (FORSUDp). Diversas variáveis foram medidas para o acompanhamento do processo: DQO de amostras filtradas e não filtradas, alcalinidade total, alcalinidade a ácidos voláteis, série de sólidos, nitrogênio total Kjeldahl e amoniacal, fósforo total, ácidos voláteis e composição do biogás por cromatografia gasosa. Também se acompanhou o desenvolvimento da microbiota anaeróbia por microscopia óptica de contraste de fase e fluorescência. A microscopia eletrônica de varredura indicou desenvolvimento de biofilme no meio suporte dos filtros anaeróbios. Concluiu-se que a taxa de recirculação variável consegue acelerar o processo, além de alterar de forma significativa o comportamento dos reatores e a forma de interação entre os mesmos / The objective of this study was to test the following hypothesis: varying the leachate recirculation rate in a hybrid batch reactor would make it possible to shorten the time required for the digestion of MSW, in comparison to a similar system operating with constant recirculation rate. Two sets of reactors were operated, one with constant recirculation rate (control system) and the other with varying rate (experimental system). The reactors were kept in a room which internal temperature was controlled. The temperature was kept at 35 ºC '+ OU -' 1ºC. Each set of reactors was composed of a solids reactor, filled with solid wastes and inoculated with landfill leachate, and an anaerobic filter, that should treat the leachate form the first one. The substract used was the standardized organic fraction of the municipal solid wastes (FORSUDp). Several parameters were measured for process monitoring: COD (in filtered and not filtered samples), total alkalinity, volatile acids alkalinity, solids, Kjeldahl nitrogen, ammonium nitrogen, total phosphorus, volatile acids by gas chromatography and biogas composition. The development of the anaerobic microorganisms was observed with phase-contrast and fluorescence microscopy. Scanning electron microscopy of the packed bed media showed biofilm development. It was concluded that the varying leachate recirculation rate proved to be able to enhance the degradation process. Besides, this operational strategy entirely changed the behaviour and interaction of the reactors

O presente estudo teve por finalidade testar a hipótese de que o uso de uma taxa de recirculação variável do lixiviado de um reator tratando a fração orgânica dos resíduos sólidos urbanos domésticos, em um sistema em batelada híbrido, seria capaz de acelerar a degradação deste substrato em comparação com um sistema idêntico, porém que operasse com taxa de recirculação constante. Dois conjuntos de reatores foram operados, um com taxa de recirculação constante (sistema controle) e o outro com taxa de recirculação variável (sistema experimental). Os reatores foram mantidos em sala climatizada, com temperatura mantida em 35 ºC '+ OU -' 1 ºC. Cada conjunto era composto de um reator de sólidos, preenchido com o resíduo a ser digerido misturado a lixiviado de aterro sanitário, e um filtro anaeróbio, destinado a tratar o lixiviado que saía do primeiro. O substrato utilizado foi a fração orgânica dos resíduos sólidos urbanos domésticos padronizada (FORSUDp). Diversas variáveis foram medidas para o acompanhamento do processo: DQO de amostras filtradas e não filtradas, alcalinidade total, alcalinidade a ácidos voláteis, série de sólidos, nitrogênio total Kjeldahl e amoniacal, fósforo total, ácidos voláteis e composição do biogás por cromatografia gasosa. Também se acompanhou o desenvolvimento da microbiota anaeróbia por microscopia óptica de contraste de fase e fluorescência. A microscopia eletrônica de varredura indicou desenvolvimento de biofilme no meio suporte dos filtros anaeróbios. Concluiu-se que a taxa de recirculação variável consegue acelerar o processo, além de alterar de forma significativa o comportamento dos reatores e a forma de interação entre os mesmos / The objective of this study was to test the following hypothesis: varying the leachate recirculation rate in a hybrid batch reactor would make it possible to shorten the time required for the digestion of MSW, in comparison to a similar system operating with constant recirculation rate. Two sets of reactors were operated, one with constant recirculation rate (control system) and the other with varying rate (experimental system). The reactors were kept in a room which internal temperature was controlled. The temperature was kept at 35 ºC '+ OU -' 1ºC. Each set of reactors was composed of a solids reactor, filled with solid wastes and inoculated with landfill leachate, and an anaerobic filter, that should treat the leachate form the first one. The substract used was the standardized organic fraction of the municipal solid wastes (FORSUDp). Several parameters were measured for process monitoring: COD (in filtered and not filtered samples), total alkalinity, volatile acids alkalinity, solids, Kjeldahl nitrogen, ammonium nitrogen, total phosphorus, volatile acids by gas chromatography and biogas composition. The development of the anaerobic microorganisms was observed with phase-contrast and fluorescence microscopy. Scanning electron microscopy of the packed bed media showed biofilm development. It was concluded that the varying leachate recirculation rate proved to be able to enhance the degradation process. Besides, this operational strategy entirely changed the behaviour and interaction of the reactors