Thèse numérisée par la Direction des bibliothèques de l'Université de Montréal.

In a single duct variable air volume (SDVAV) system, the supply air temperature is usually set as a constant value. Since this constant setpoint is selected to satisfy the maximum cooling load conditions, significant reheat will occur once the airflow reaches the minimum and the heating load increases. Resetting the supply air temperature (SAT) higher during the heating season can reduce the reheat. However, air flow will increase when the SAT is higher which consume extra fan power. Therefore, to minimize the total operating cost of a SDVAV system, the supply air temperature is typically reset based on outside air temperature (OAT) with a linear reset schedule. However, the linear reset schedule is often determined based on the engineer’s experience and it may not represent the optimal reset schedule for each building. This thesis documents a study to determine the optimized supply air temperature reset schedule for SDVAV systems and analyzes the influencing factors under different operation scenarios. The study was divided into five main sections. The first section introduces the research background and objective. Literature review is documented after the introduction. The third section describes the methodology used in this study and the fourth section develops an in-depth discussion and analysis of the impact of the key influencing factors: minimum air flow ratio; ratio of exterior zone area to total floor area (i.e., exterior area ratio); internal load and the prices of the electricity; the cooling and the heating energy. The simulation results using EnergyPlus Version 2.1.0 for various operation scenarios are investigated in this section. The last section is a conclusion of the whole study. The optimized supply air temperature can be set with respect to the OAT. The study found that instead of a simple linear relationship, the optimal reset schedule has several distinctive segments. Moreover, it is found that the optimal supply air temperature reset schedule should be modified with the change of operation conditions (e.g., different minimum flow ratios and internal loads). Minimum air flow ratio has a significant impact on energy consumption in a SDVAV system. Exterior area ratio determines zone load distribution and will change system load indirectly. For buildings with small internal load, a more aggressive supply air temperature reset tactic can be implemented. In addition, the cost of electricity, cooling and heating energy can determine which end use energy (i.e., reheat energy and fan power) should take the priority.

The potential of static pressure reset (SPR) control to save fan energy in variable air volume HVAC systems has been well documented. Current research has focused on the creation of reset strategies depending on specific system features. As the commissioning process has begun to require the prediction of savings, knowledge of the extent to which various SPR control strategies impact fan energy has become increasingly important. This research aims to document existing SPR control strategies and utilize building data and simulation to estimate fan energy use. A comprehensive review of the literature pertaining to SPR control was performed and the results were organized into a top-down flow chart tool. Based on the type of feedback available from a particular system, or lack thereof, this tool will facilitate the selection of a SPR control strategy. A field experiment was conducted on a single duct variable air volume system with fixed discharge air temperature and static pressure setpoints. Finally, an air-side model of the experimental system was created using detailed building design information and calibrated using field measurements. This model was used to estimate the fan energy required to supply the trended airflow data using fixed static pressure (FSP) and SPR control based on zone demand, system demand, and outside air temperature. While utilizing trend data from November 1, 2008 to February 12, 2009, the FSP control of the experimental system was used as the baseline for ranking the energy savings potential of nine different forms of duct static pressure control. The highest savings (73-74%) were achieved using zonal demand based SPR control. System demand based SPR control yielded savings ranging from 59 to 76%, which increased when the duct sensor was positioned near the fan discharge and under similar zone load conditions. The outside air temperature based SPR control yielded savings of 65% since the experimental system supplied primarily perimeter zones. Finally, increasing the FSP setpoint from 2 to 3 inWG increased fan energy by 45%, while decreasing the setpoint from 2 to 1 inWG decreased fan energy by 41%.

none

Cerný, em 1964, conjecturou que um autômato sincronizado com n estados possui uma palavra sincronizadora mínima de tamanho no máximo (n-1)². Esta conjectura permanece em aberto. Neste trabalho são apresentados algoritmos para obter palavras sincronizadoras e é feito um experimento comparativo entre os resultados obtidos por estes algoritmos em relação a algumas séries infinitas de autômatos. Por fim, é feito um breve histórico sobre os resultados parciais obtidos até a presente data e alguns destes trabalhos são apresentados em mais detalhes. / Cerný, on 1964, conjectured that a synchronizing automata with n states has a synchronizing word of size at most (n-1)². The conjecture remains open. We show some algorithms for obtaining synchronizing sequences and a comparative experiment between these algorithms with respect to some infinite series of automata. Furthermore, we briefly survey some of the partial results obtained until the present day.

Les réseaux de Petri et leurs extensions constituent un formalisme très connu pour la modélisation de la spécification des systèmes à évènements discrets temps réel. Pour ces systèmes, les exigences de vérification et de validation sont indispensables pour garantir leur bon fonctionnement car la moindre erreur peut entrainer des conséquences catastrophiques. Pour analyser le comportement du système, il est courant de construire le graphe d’états (ou espace d’états) qui énumère de façon exhaustive les différents états accessibles.Ce graphe permet de vérifier des propriétés génériques comme le caractère borné, l’accessibilité, la terminaison, la vivacité, l’absence de blocage, etc. Mais la construction de cet espace d’états est confrontée au problème d’explosion combinatoire du nombre d’états lié à la complexité et à la concurrence du système.L’une des alternatives pour contenir cette combinatoire est de conserver uniquement les ordres partiels entre les évènements. Le dépliage est une technique d’ordre partiel adaptée à la vérification, au diagnostic et à la planification. Cette thèse contribue à l’étude du dépliage des réseaux de Petri et, plus particulièrement,apporte une contribution au dépliage des réseaux de Petri avec des arcs de reset. De plus, le dépliage explicite les exécutions possibles du système qui sont, par définition, des processus. Il s’agit plus précisément des processus de branchement qui diffèrent des processus classiques. Ainsi, nous proposons une algèbre adaptée aux processus de branchement issus du dépliage des réseaux de Petri. / Petri nets and their extensions are a well-known formalism for modeling the specification of discrete events systems real-time. For these systems, the demands of verification and validation are essential to ensure their smooth functioning because the least error can lead to catastrophic consequences. To analyze the behavior of the system, it is common to construct the stategraph (or state space) which lists exhaustively the different accessible states. This graph allows to check the generic properties such as boundedness, accessibility, termination, vivacity, the absence of blocking, etc. But building this state space is confronted with the problem of combinatorial explosion of the number of states related to the complexity and concurrence of the system. One of the alternatives to contain this combination is to keep only partial orders between events. Unfolding is a partial order adapted to technical verification, diagnosis and planning. This thesis contributes to the study of the unfolding of Petri nets and, in particular, his contributes to the unfolding of Petri nets with reset arcs. Moreover, the unfolding explains the possible executions of the system which are, by definition, the processes. This is specifically the branching processes that differ from conventional processes. Thus, we propose an algebra adapted to the branching process from unfolding of Petri nets.

Cerný, em 1964, conjecturou que um autômato sincronizado com n estados possui uma palavra sincronizadora mínima de tamanho no máximo (n-1)². Esta conjectura permanece em aberto. Neste trabalho são apresentados algoritmos para obter palavras sincronizadoras e é feito um experimento comparativo entre os resultados obtidos por estes algoritmos em relação a algumas séries infinitas de autômatos. Por fim, é feito um breve histórico sobre os resultados parciais obtidos até a presente data e alguns destes trabalhos são apresentados em mais detalhes. / Cerný, on 1964, conjectured that a synchronizing automata with n states has a synchronizing word of size at most (n-1)². The conjecture remains open. We show some algorithms for obtaining synchronizing sequences and a comparative experiment between these algorithms with respect to some infinite series of automata. Furthermore, we briefly survey some of the partial results obtained until the present day.

This paper discusses how focus changes prosodic structure in Tokyo Japanese. It is generally believed that focus blocks the intonational process of downstep and causes a pitch reset. This paper presents experimental evidence against this traditional view by looking at the prosodic behavior of Wh words, which receive focus lexically in Japanese as in other languages. It is demonstrated, specifically, that the focused Wh element does not block downstep although it receives a much higher pitch than its preceding element. This suggests that presence of lexical focus does not trigger pitch reset in Japanese.

Made available in DSpace on 2014-06-12T18:04:10Z (GMT). No. of bitstreams: 2 arquivo7191_1.pdf: 588691 bytes, checksum: 2a4cdc6ab84afaa4aa34cb0cf4f67a47 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2007 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Uma ferramenta útil para a modelagem de dados em que a variável resposta assume continuamente valores no intervalo (0,1) é o modelo de regressão beta proposto por Ferrari & Cribari-Neto (2004). O modelo proposto é baseado na suposição que a variável resposta tem distribuição beta utilizando uma parametrização da lei beta que é indexada pela média e por um parâmetro de precisão. Neste modelo, supõe-se também que a variável resposta é relacionada a outras variáveis através de uma estrutura de regressão. O objetivo desta dissertação é propor um teste de erro de especificação correta em modelos de regressão beta, a partir do teste RESET proposto por Ramsey (1969). A avaliação numérica realizada revelou que o teste proposto é útil para detecção do uso de função de ligação a incorreta bem como de não-linearidades no predito linear. Mostramos que o teste proposto realizado através do teste escore apresentou, em geral, melhores resultados no que tange a tamanho e poder. Adicionalmente, mostramos que o melhor desempenho é alcançado quando se utiliza uma potência do preditor linear ajustado ou uma potência da resposta média estimada como variável de teste. O teste proposto também apresenta bom desempenho para pequenos tamanhos mostrais, apesar de ser baseado em aproximações assintóticas

Les contrôleurs à réinitialisation sont une classe de systèmes hybrides dont la valeur de tout ou partie des états peut être instantannément modifiée sous certaines conditions algébriques. Cette interaction entre dynamique temps-continu et temps-discret de ces contrôleurs permet souvent de dépasser les limites des contrôleurs temps-continu. Dans cette thèse, nous proposons des conditions constructives (sous forme d’Inégalités Matricielles Linéaires) pour analyser la stabilité et les performances de boucle de commande incluant un contrôleur à réinitialisation. En particulier, nous prenons en compte la présence de saturation en amplitude des actionneurs du système. Ces non-linéarités sont souvent source d'une dégradation des performances voir d’instabilité. Les résultats proposés permettent d’estimer le domaine de stabilité et un niveau de performance pour ces systèmes, en s’appuyant sur des fonctions de Lyapunov quadratiques ou quadratiques par morceaux. Au delà de l'aspect analyse, nous exposons deux approches pour améliorer la région de stabilité (nouvelle loi de réinitialisation et stratégie « anti-windup »). / Hybrid controllers are flexible tools for achieving system stabilization and/or performance improvement tasks. More particularly, hybrid controllers enrich the spectrum of achievable trade-offs. Indeed, the interaction of continuous- and discrete-time dynamics in a hybrid controller leads to rich dynamical behavior and phenomena not encountered in purely continuous-time system. Reset control systems are a class of hybrid controllers whose states are reset depending on an algebraic condition. In this thesis, we propose constructive conditions (Linear Matrix Inequalities) to analyze stability and performance level of a closed-loop system including a reset element. More particularly, we consider a magnitude saturation which could be the source of undesirable effects on these performances, including instability. Proposed results estimate the stability domain and a performance level of such a system, by using Lyapunov-like approaches. Constructive algorithms are obtained by exploiting properties of quadratic - or piecewise quadratic - Lyapunov functions. Beyond analysis results, we propose design methods to obtain a stability domain as large as possible. Design methods are based on both continuous-time approaches (anti-windup compensator) and hybrid-time approaches (design of adapted reset rules).