Spelling suggestions: "subject:"0ptimal control"" "subject:"aptimal control""
421 |
Modeling and real-time optimal energy management for hybrid and plug-in hybrid electric vehiclesDong, Jian 15 February 2017 (has links)
Today, hybrid electric propulsion technology provides a promising and practical solution for improving vehicle performance, increasing energy efficiency, and reducing harmful emissions, due to the additional flexibility that the technology has provided in the optimal power control and energy management, which are the keys to its success.
In this work, a systematic approach for real-time optimal energy management of hybrid electric vehicles (HEVs) and plug-in hybrid electric vehicles (PHEVs) has been introduced and validated through two HEV/PHEV case studies. Firstly, a new analytical model of the optimal control problem for the Toyota Prius HEV with both offline and real-time solutions was presented and validated through Hardware-in-Loop (HIL) real-time simulation. Secondly, the new online or real-time optimal control algorithm was extended to a multi-regime PHEV by modifying the optimal control objective function and introducing a real-time implementable control algorithm with an adaptive coefficient tuning strategy. A number of practical issues in vehicle control, including drivability, controller integration, etc. are also investigated. The new algorithm was also validated on various driving cycles using both Model-in-Loop (MIL) and HIL environment.
This research better utilizes the energy efficiency and emissions reduction potentials of hybrid electric powertrain systems, and forms the foundation for development of the next generation HEVs and PHEVs. / Graduate / laindeece@gmail.com
|
422 |
Contrôle quantique de la rotation moléculaire et de processus de Résonance Magnétique Nucléaire / Qauntum control of molecular rotation and of processes in Nuclear Magnetic ResonanceHamraoui, Khalid 17 April 2019 (has links)
L’objectif de cette thèse est d’appliquer des méthodes de contrôle quantique pour manipuler la dynamique rotationnelle de molécules et améliorer l’efficacité de processus en résonance magnétique nucléaire.Ces techniques ont été utilisées théoriquement et expérimentalement pour contrôler l’orientation d’une molécule toupie symétrique à l’aide de champ THz. Cette étude a été généralisée à une grande distance d’interaction entre le champ et l’échantillon. Dans ce cas, la molécule ne peut plus être considérée comme isolée. Nous avons également montré jusqu'à quel point l’évolution temporelle du degré d’orientation pouvait être mise en forme. Des méthodes de contrôle optimal ont permis de déterminer le champ THz pour atteindre cet état à la fois à températures nulle et non-nulle. Un autre chapitre présente un nouvel algorithme d’optimisation pour les dynamiques périodiques. Cet algorithme est appliqué à la maximisation du SNR en RMN. Un dernier chapitre est dédié à un article de vulgarisation sur l’effet de la raquette de tennis. Cet effet géométrique peut être observé dans tout corps rigide suffisamment asymétrique. / The goal of this thesis is to apply quantum control techniques to manipulate molecular rotation and to enhance the efficiency of processes in Nuclear Magnetic Resonance.These techniques have been used theoretically and experimentally to control the orientation of a symmetric top molecule by means of THz laser fields. This study has been extended to the case of a long interaction distance between the field and the sample. In this case, the molecule cannot be approximated as isolated. We have also shown the extend to which the time evolution of the degree of orientation can be shaped. Optimal control techniques were used to design the THz field which allows to reach the corresponding dynamics, both at zero and non zero temperatures. Another chapter proposes a new optimization algorithm in the case of periodic quantum dynamics. We apply this algorithm to the maximization of the SNR in NMR. A last chapter is dedicated to a popular paper about the tennis racket effect. This geometric effect can be observed in any asymetric rigid body.
|
423 |
Controlador dinâmico para o problema linear quadrático com saltos não observados / Dynamic controller for the linear quadratic jump problem without mode observationRomero, Luiz Henrique 04 June 2019 (has links)
Os Sistemas Lineares Sujeitos a Saltos Markovianos têm sido amplamente estudados nas últimas décadas pois fornecem modelos adequados para aplicações com mudanças bruscas de comportamento, possivelmente devido à falhas. Também é muito comum em aplicações do mundo real em que o chamado estado do sistema não seja observado de forma perfeita e instantânea. Com essa motivação, consideramos o problema linear quadrático e propomos um controlador independente da variável de salto, que é um componente de estado, o que é atraente para aplicações reais. Utilizamos dois métodos clássicos, Genético e Gradiente, e propomos derivados que combinam características favoráveis de ambos. Também consideramos o caso em que não observamos o estado de Markov diretamente, mas através de uma variável, um sensor, que provê informação sobre este parâmetro. / Markov Jump Linear Systems have been extensively studied in the last decades as they provide suitable models for applications featuring abrupt changes of behaviour. It is also quite common in real world applications that the so called state of the system is not perfectly and immediately observed. With this motivation, we consider the linear quadratic jump problem and we propose a controller that is irrespective of the jump variable (a component os the state), which is appealing for real world problems. We use classical Genetic and Gradient optimization methods and we propose variants combining favorable features of both of them; We also consider the case which we do not have direct access on the Markovian jump parameter, but a variable, a sensor, which provides information on this parameter.
|
424 |
Modular Multilevel Converter Control for HVDC Operation : Optimal Shaping of the Circulating Current Signal for Internal Energy Regulation / Commande adaptée pour le convertisseur modulaire multiniveaux pour les liaisons à courant continuesBergna Diaz, Gilbert 03 July 2015 (has links)
Dans le cadre du programme de croissance Européen 2020, la commission européenne a mis en place officiellement un chemin à long terme pour une économie à faible émission de carbone, en aspirant une réduction d’au moins 80% des émissions de gaz à effet de serre, d’ici 2050. Répondre à ces exigences ambitieuses, impliquera un changement majeur de paradigme, et notamment en ce qui concerne les infrastructures du réseau électrique. Les percées dans la technologie des semi-conducteurs et les avancées avec les nouvelles topologies d’électronique de puissance et leurs contrôle-commandes, ont contribué à l’impulsion donnée au processus en cours de réaliser un tel SuperGrid. Une percée technologique majeure a eu lieu en 2003, avec le convertisseur modulaire multi-niveaux (MMC ou M2C), présenté par le professeur Marquardt, et qui est actuellement la topologie d’électronique de puissance la plus adaptée pour les stations HVDC. Cependant, cette structure de conversion introduit également un certain nombre de défis relativement complexes tels que les courants “additionnels” qui circulent au sein du convertisseur, entrainant des pertes supplémentaires et un fonctionnement potentiellement instable. Ce projet de thèse vise à concevoir des stratégies de commande “de haut niveau” pour contrôler le MMC adaptées pour les applications à courant continue-haute tension (HVDC), dans des conditions de réseau AC équilibrés et déséquilibrés. La stratégie de commande optimale identifiée est déterminée via une approche pour la conception du type “de haut en bas”, inhérente aux stratégies d’optimisation, où la performance souhaitée du convertisseur MMC donne la stratégie de commande qui lui sera appliquée. Plus précisément, la méthodologie d’optimisation des multiplicateurs de Lagrange est utilisée pour calculer le signal minimal de référence du courant de circulation du MMC dans son repère naturel. / Following Europe’s 2020 growth program, the Energy Roadmap 2050 launched by the European Commission (EC) has officially set a long term path for a low-carbon economy, assuming a reduction of at least 80% of greenhouse gas emissions by the year 2050. Meeting such ambitious requirements will imply a major change in paradigm, including the electricity grid infrastructure as we know it.The breakthroughs in semi-conductor technology and the advances in power electronics topologies and control have added momentum to the on-going process of turning the SuperGrid into a reality. Perhaps the most recent breakthrough occurred in 2003, when Prof. Marquardt introduced the Modular Multilevel Converter (MMC or M2C) which is now the preferred power electronic topology that is starting to be used in VSC-HVDC stations. It does however, introduce a number of rather complex challenges such as “additional” circulating currents within the converter itself, causing extra losses and potentially unstable operation. In addition, the MMC will be required to properly balance the capacitive energy stored within its different arms, while transferring power between the AC and DC grids that it interfaces.The present Thesis project aimed to design adequate “high-level” MMC control strategies suited for HVDC applications, under balanced and unbalanced AC grid conditions. The resulting control strategy is derived with a “top-to-bottom” design approach, inherent to optimization strategies, where the desired performance of the MMC results in the control scheme that will be applied. More precisely, the Lagrange multipliers optimization methodology is used to calculate the minimal MMC circulating current reference signals in phase coordinates, capable of successfully regulating the capacitive arm energies of the converter, while reducing losses and voltage fluctuations, and effectively decoupling any power oscillations that would take place in the AC grid and preventing them from propagating into the DC grid.
|
425 |
Método variacional com atualização múltipla de ganhos para controle de sistemas lineares com parâmetros sujeitos a saltos Markovianos não observados / Variational method with multiple gains update for control of linear systems with parameters subject to unobserved Markov jumpOliveira, Larissa Tebaldi de 11 June 2014 (has links)
Neste trabalho foi estudado um problema de controle de sistemas lineares com saltos Markovianos sem observação da variável de salto, que pode ser escrito como um problema de otimização de considerável complexidade. As contribuições para a área estão divididas em três aspectos. Um dos avanços foi a elaboração de um contraexemplo para a conjectura de que há somente um mínimo local isolado para o problema. Além disso, foi estudado o problema de otimização intermediário, que consiste em fixar todas as variáveis do problema exceto duas matrizes de ganhos, e os resultados indicam que, com uma pequena alteração na formulação, este é um problema biquadrático. Por fim, novos algoritmos foram elaborados a partir de um método disponível na literatura, chamado de método Variacional, adaptando-o para atualizar os ganhos aos pares, levando a problemas intermediários biquadráticos. Três métodos foram implementados para a resolução destes problemas: dois métodos clássicos de descida, Newton e Gradiente, e uma adaptação do próprio método Variacional. Para a análise dos resultados foram utilizados exemplos gerados aleatoriamente a partir do Gerador de SLSM, que pode ser encontrado na literatura, e o método Variacional como referência para comparação com os métodos propostos / This work addresses a control problem arising in linear systems with Markov jumps without observation of the jump variable and advances in three different aspects. First, it is presented a counterexample to the conjecture that states about the uniqueness of local minimum. Second, the intermediary optimization problem, which sets all the variables of the problem except two arrays of gains, was studied and the results suggested that a slight modification in the formulation makes the intermediary problem a biquadratic one. Finally, new algorithms were developed based on a method available in the literature, which is frequently referred to as the Variational method, adapting it to update the gains in pairs, leading to biquadratic intermediary problems. Three methods were implemented to solve these intermediary problems: two classical descent methods, Newton and Gradient, and an adaptation of the Variational method. To evaluate the performance of the proposed methods, randomly generated examples were used and the Variational method was set as reference for comparing the results
|
426 |
Discrete-time jump linear systems with Markov chain in a general state space. / Sistemas lineares com saltos a tempo discreto com cadeia de Markov em espaço de estados geral.Figueiredo, Danilo Zucolli 04 November 2016 (has links)
This thesis deals with discrete-time Markov jump linear systems (MJLS) with Markov chain in a general Borel space S. Several control issues have been addressed for this class of dynamic systems, including stochastic stability (SS), linear quadratic (LQ) optimal control synthesis, fllter design and a separation principle. Necessary and sffcient conditions for SS have been derived. It was shown that SS is equivalent to the spectral radius of an operator being less than 1 or to the existence of a solution to a \\Lyapunov-like\" equation. Based on the SS concept, the finite- and infinite-horizon LQ optimal control problems were tackled. The solution to the finite- (infinite-)horizon LQ optimal control problem was derived from the associated control S-coupled Riccati difference (algebraic) equations. By S-coupled it is meant that the equations are coupled via an integral over a transition probability kernel having a density with respect to a in-finite measure on the Borel space S. The design of linear Markov jump filters was analyzed and a solution to the finite- (infinite-)horizon filtering problem was obtained based on the associated filtering S-coupled Riccati difference (algebraic) equations. Conditions for the existence and uniqueness of a stabilizing positive semi-definite solution to the control and filtering S-coupled algebraic Riccati equations have also been derived. Finally a separation principle for discrete-time MJLS with Markov chain in a general state space was obtained. It was shown that the optimal controller for a partial information optimal control problem separates the partial information control problem into two problems, one associated with a filtering problem and the other associated with an optimal control problem with complete information. It is expected that the results obtained in this thesis may motivate further research on discrete-time MJLS with Markov chain in a general state space. / Esta tese trata de sistemas lineares com saltos markovianos (MJLS) a tempo discreto com cadeia de Markov em um espaço geral de Borel S. Vários problemas de controle foram abordados para esta classe de sistemas dinâmicos, incluindo estabilidade estocástica (SS), síntese de controle ótimo linear quadrático (LQ), projeto de filtros e um princípio da separação. Condições necessárias e suficientes para a SS foram obtidas. Foi demonstrado que SS é equivalente ao raio espectral de um operador ser menor que 1 ou à existência de uma solução para uma equação de Lyapunov. Os problemas de controle ótimo a horizonte finito e infinito foram abordados com base no conceito de SS. A solução para o problema de controle ótimo LQ a horizonte finito (infinito) foi obtida a partir das associadas equações a diferenças (algébricas) de Riccati S-acopladas de controle. Por S-acopladas entende-se que as equações são acopladas por uma integral sobre o kernel estocástico com densidade de transição em relação a uma medida in-finita no espaço de Borel S. O projeto de filtros lineares markovianos foi analisado e uma solução para o problema da filtragem a horizonte finito (infinito) foi obtida com base nas associadas equações a diferenças (algébricas) de Riccati S-acopladas de filtragem. Condições para a existência e unicidade de uma solução positiva semi-definida e estabilizável para as equações algébricas de Riccati S-acopladas associadas aos problemas de controle e filtragem também foram obtidas. Por último, foi estabelecido um princípio da separação para MJLS a tempo discreto com cadeia de Markov em um espaço de estados geral. Foi demonstrado que o controlador ótimo para um problema de controle ótimo com informação parcial separa o problema de controle com informação parcial em dois problemas, um deles associado a um problema de filtragem e o outro associado a um problema de controle ótimo com informação completa. Espera-se que os resultados obtidos nesta tese possam motivar futuras pesquisas sobre MJLS a tempo discreto com cadeia de Markov em um espaço de estados geral.
|
427 |
Modelagem, controle e otimização de consumo de combustível para um veículo híbrido elétrico série-paralelo. / Modeling, control and application of dynamic programming to a series-parallel hydrid electric vehicle.Trindade, Ivan Miguel 16 May 2016 (has links)
O principal objetivo dos veículos híbridos é diminuir o consumo de combustível em relação a veículos convencionais. Para isso, existe a necessidade de realizar a integração dos diferentes sistemas do trem-de-força e coordenar o seu funcionamento através de estratégias de controle. Tais estratégias são desenvolvidas e simuladas em conjunto com um modelo computacional da planta do veículo antes de serem aplicadas em uma unidade de controle eletrônica. O presente estudo tem como objetivo analisar o gerenciamento de energia em um veículo híbrido elétrico não-plugin do tipo série-paralelo visando à diminuição de consumo de combustível. O método de otimização global é utilizado para encontrar as variáveis de controle que resultam no mínimo consumo de combustível em um determinado ciclo de condução. Na primeira etapa, um modelo computacional da planta do veículo e da estratégia de controle não-ótima são criados. Os resultados obtidos da simulação são então comparados com dados experimentais do veículo operando em dinamômetro de chassis. A seguir, o método de otimização global é aplicado ao modelo computacional utilizando programação dinâmica e tendo como objetivo a minimização do consumo de combustível total ao final do ciclo. Os resultados mostram considerável redução do consumo de combustível utilizando otimização global e tendo como variável de controle não só a razão de distribuição de torque mas também os pontos de operação do motor de combustão. Os modelos computacionais criados nesse trabalho são disponibilizados e podem ser usados para o estudo de diferentes estratégias de controle para veículos híbridos. / The main goal of hybrid electric vehicles is to decrease engine emission and fuel consumption levels. In order to realize this, one must perform the powertrain system integration and coordinate its operation through supervisory control strategies. These control strategies are developed in a simulation environment containing the plant model of the powertrain before they can be implemented in a real-time control unit. The goal of this work is to analyze the energy management strategy which minimizes the fuel consumption in a series-parallel non-plugin hybrid electric vehicle. Global optimization is used for finding the control variables that result in the minimum fuel consumption for a specific driving cycle. In a first stage, a computational model of vehicle plant and non-optimal control strategy are created. The results from the simulation are compared against experimental data from chassis dynamometer tests. Next, a global optimization strategy is applied using dynamic programming in order to minimize total fuel consumption at the end of the driving cycle. The results from the optimization show a considerable fuel consumption reduction having as control variables not only the torque-split strategy but also the engine operating points. As contribution from this work, the computational models are made available and can be used for analyzing different control strategies for hybrid vehicles.
|
428 |
Algoritmos de controle ótimo quadrático com restrições. / Algorithms for the solution of robust quadratic optimal control problems with restrictions.Barão, Renato Casali 12 December 1997 (has links)
O objetivo do trabalho é apresentar dois algoritmos para a solução de problemas de controle ótimo quadrático robusto com restrições, dentro de um contexto de controladores preditivos (MPC do inglês Model Predictive Control). Inicialmente apresentamos uma breve introdução aos algoritmos MPC, com ênfase na abordagem do controlador linear quadrático. Em seguida são apresentados os dois algoritmos de interesse, que utilizam técnicas de otimização LMI. Dessa forma as restrições e as incertezas podem ser colocadas em formas computacionalmente tratáveis. Por fim são realizadas simulações e comparações entre esses algoritmos, bem como com técnicas de MPC encontradas na literatura atual. / The goal of the work is to present two algorithms for the solution of robust quadratic optimal control problems with restrictions, within a model predictive control (MPC) setup. Initially we present a brief introduction of the MPC algorithms, emphasizing the linear quadratic controller approach. Next the two algorithms of interest, using LMI optimization techniques, are presented. By using this technique the restrictions and uncertainties can be written in a computational way. Finally some simulations and comparisons between these algorithms, as well as with MPC techniques found in the current literature, are performed.
|
429 |
Controle ótimo de sistemas com saltos Markovianos e ruído multiplicativo com custos linear e quadrático indefinido. / Indefinite quadratic with linear costs optimal control of Markov jump with multiplicative noise systems.Paulo, Wanderlei Lima de 01 November 2007 (has links)
Esta tese trata do problema de controle ótimo estocástico de sistemas com saltos Markovianos e ruído multiplicativo a tempo discreto, com horizontes de tempo finito e infinito. A função custo é composta de termos quadráticos e lineares nas variáveis de estado e de controle, com matrizes peso indefinidas. Como resultado principal do problema com horizonte finito, é apresentada uma condição necessária e suficiente para que o problema de controle seja bem posto, a partir da qual uma solução ótima é derivada. A condição e a lei de controle são escritas em termos de um conjunto acoplado de equações de Riccati interconectadas a um conjunto acoplado de equações lineares recursivas. Para o caso de horizonte infinito, são apresentadas as soluções ótimas para os problemas de custo médio a longo prazo e com desconto, derivadas a partir de uma solução estabilizante de um conjunto de equações algébricas de Riccati acopladas generalizadas (GCARE). A existência da solução estabilizante é uma condição suficiente para que tais problemas sejam do tipo bem posto. Além disso, são apresentadas condições para a existência das soluções maximal e estabilizante do sistema GCARE. Como aplicações dos resultados obtidos, são apresentadas as soluções de um problema de otimização de carteiras de investimento com benchmark e de um problema de gestão de ativos e passivos de fundos de pensão do tipo benefício definido, ambos os casos com mudanças de regime nas variáreis de mercado. / This thesis considers the finite-horizon and infinite-horizon stochastic optimal control problem for discrete-time Markov jump with multiplicative noise linear systems. The performance criterion is assumed to be formed by a linear combination of a quadratic part and a linear part in the state and control variables. The weighting matrices of the state and control for the quadratic part are allowed to be indefinite. For the finite-horizon problem the main results consist of deriving a necessary and sufficient condition under which the problem is well posed and a optimal control law is derived. This condition and the optimal control law are written in terms of a set of coupled generalized Riccati difference equations interconnected with a set of coupled linear recursive equations. For the infinite-horizon problem a set of generalized coupled algebraic Riccati equations (GCARE) is studied. In this case, a sufficient condition under which there exists the maximal solution and a necessary and sufficient condition under which there exists the mean square stabilizing solution for the GCARE are presented. Moreover, a solution for the discounted and long run average cost problems is presented. The results obtained are applied to solver a portfolio optimization problem with benchmark and a pension fund problem with regime switching.
|
430 |
Prefetching control for on-demand contents distribution : a Markov decision process study / Contrôle du préchargement pour la distribution de contenus à la demande : une approche par les processus de décision markoviensMorad, Olivia 17 September 2014 (has links)
Le contexte de la thèse porte sur le contrôle des réseaux de distribution de contenu à la demande. La performance des systèmes distribués interactifs dépend essentiellement sur la prévision du comportement de l'utilisateur et la bande passante en tant que ressource de réseau critique. Le préchargement est une approche prédictive bien connu dans le World Wide Web ce qui évite les délais de réponse en exploitant un temps d'arrêt que permet d'anticiper les futures demandes de l'utilisateur et prend avantage des ressources réseau disponibles. Le contrôle de préchargement est une opération vitale pour les systèmes à la demande interactifs où la réponse instantanée est le facteur crucial pour la réussite du système. Le contrôleur en ce type de système interactif fonctionne dans un environnement incertain et rend séquences de décisions à court et long terme effets stochastique. La difficulté est alors de déterminer à chaque état du système les contenus préchargés dans le cache. Le plan de préchargement pendant une session en flux continu interactif peut être modélisé comme un problème de décision séquentielle par les processus de décision de Markov (MDP). Nous nous concentrons sur le problème de contrôle de préchargement, dans lequel le contrôleur cherche à atteindre l'état du système à coût zéro aussi vite que possible. Nous modélisons ce problème de contrôle comme un problème de programmation dynamique stochastique négatif dans lequel nous minimisons le coût total prévu. Dans ce contexte, nous avons abordé les questions de recherche suivantes: 1) Comment fournir un politique de préchargement optimale/ approximative optimale qui maximise l'utilisation de la bande passante tout en minimisant les coûts de blocage et de la latence de l'utilisateur engagés sur le chemin? 2) Comment exploiter la structure du modèle de contrôle de préchargement pour aider efficacement calculer la politique de contrôle de préchargement avec la réduction des efforts de calcul et la mémoire de stockage? 3) Comment mener une étude d'évaluation pour évaluer le préchargement de différents algorithmes heuristiques basée sur le contexte de l'optimisation au lieu du cadre de l'empirique / simulation. Pour l'étude de notre problème de recherche, nous avons développé notre modèle MDP de préchargement, PREF-CT, nous avons établi ses propriétés théoriques et nous avons résolu par l'algorithme Value Iteration comme algorithme MDP pour calculer la politique de préchargement optimale. Pour calcul de la politique de préchargement optimale efficace, nous avons détecté une structure spéciale qui réalise un modèle de contrôle plus compact. Cette structure spéciale permet de développer deux algorithmes différents stratégiquement qui améliorent la complexité du calcul de la politique de préchargement optimale: - la première est « ONE-PASS » le second est « TREE-DEC ». Pour surmonter le problème de la dimensionnalité résultant du calcul de la politique de préchargement optimale, nous avons proposé l'algorithme de préchargement heuristique: « Relevant Blocks Prefetching » (RBP). Pour évaluer et comparer le préchargement politiques calculés par des algorithmes de préchargement heuristiques différents, nous avons présenté un cadre fondé sur des différentes mesures de performance. Nous avons appliqué le cadre proposé sous différentes configurations de coûts et différents comportements des utilisateurs pour évaluer les politiques de préchargement calculées par notre algorithme de préchargement proposé; RBP. Par rapport aux politiques de préchargement optimales, l'analyse expérimentale a prouvé des performances significatives des politiques de préchargement de l'heuristique du RBP algorithme. En outre, l'algorithme heuristique de préchargement; RBP se distingue par une propriété de clustériser qui est important pour réduire considérablement la mémoire nécessaire pour stocker la politique de préchargement. / The thesis context is concerned with the control of theOn-demand contents distribution networks. The performance of suchinteractive distributed systems basically depends on the prediction ofthe user behavior and the bandwidth as a critical network resource.Prefetching is a well-known predictive approach in the World Wide Webwhich avoids the response delays by exploiting some downtime thatpermits to anticipate the user future requests and takes advantage ofthe available network resources. Prefetching control is a vitaloperation for the On-demand interactive systems where the instantaneousresponse is the crucial factor for the system success. The controller insuch type of interactive system operates in an uncertain environment andmakes sequences of decisions with long and short term stochasticeffects. The difficulty, then, is to determine at every system statewhich contents to prefetch into the cache. The prefetching plan duringan interactive streaming session can be modeled as a sequential decisionmaking problem by a Markov Decision Process (MDP). We focus on theprefetching control problem in which the controller seeks to reach aZero-Cost system state as quickly as possible. We model this controlproblem as a Negative Stochastic Dynamic Programming problem in which weminimize the undiscounted total expected cost. Within this context, weaddressed the following research questions: 1) How to provide anoptimal/approximate-optimal prefetching policy that, maximizes thebandwidth utilization while minimizes the user's blocking and latencycosts incurred along the way? 2) How to exploit structure in theprefetching control model to help efficiently compute such prefetchingcontrol policy with both computational efforts and storage memoryreduction? 3) How to conduct a performance evaluation study to evaluatedifferent prefetching heuristic algorithms based on the context of thecontrol optimization rather than the context of theempirical/simulation. For studying our research problem, we developedour MDP prefetching control model, PREF-CT, we established itstheoretical properties and we solved it by the Value Iteration algorithmas MDP algorithm for computing the optimal prefetching policy. Forcomputing the optimal prefetching policy efficiently, we detected aspecial structure that achieves more compact control model. This specialstructure permits to develop two strategically different algorithmswhich improve the complexities of computing the optimal prefetchingpolicy: - the first one is the ONE-PASS which is based mainly on solvinga system of linear equations simultaneously in only one iteration,whereas the second is the TREE-DEC which is based on Markov decisiontree decomposition in which sequential sets of systems of equations aresolved. For overcoming the problem of the curse of dimensionalityresulting from the computation of the optimal prefetching policy, weproposed the prefetching heuristic algorithm: the Relevant BlocksPrefetching algorithm (RBP). For evaluating and comparing prefetchingpolicies computed by different prefetching heuristic algorithms, wepresented a framework based on different performance measures. Weapplied the suggested framework under different costs configurations anddifferent user behaviors to evaluate the prefetching policies computedby our proposed prefetching heuristic algorithm; the RBP. Compared tothe optimal prefetching policies, the experimental analysis provedsignificant performance of the prefetching policies of the RBP heuristicalgorithm. In addition, the RBP prefetching heuristic algorithm isdistinguished by a clustering property which is of importance to reducesignificantly the memory necessary to store the prefetching policy tothe controller.
|
Page generated in 0.086 seconds