Autonomic computing holds great promise for software systems of the future, but at the same time poses great challenges for Software Engineering. Autonomic computing research aims to design software systems that self-configure, self-repair, self-optimize and self-protect, so as to reduce software maintenance cost while improving performance. The aim of our research is to develop tool-supported methodologies for designing and operating autonomic systems. Like other researchers in this area, we assume that autonomic system architectures consist of monitoring, analysis/diagnosis, planning, and execution components that define a feedback loop and serve as the basis for system self-management. This thesis proposes an autonomic framework founded on models of requirements and design. This framework defines the normal operation of a software system in terms of models of its requirements (goal models) and/or operation (statechart models). These models determine what to monitor and how to interpret log data in order to diagnose failures. The monitoring component collects and manages log data. The diagnostic component analyzes log data, identifies failures, and pinpoints problematic components. We transform the diagnostic problem into a propositional satisfiability (SAT) problem solvable by off-the-shelf SAT solvers. Log data are preprocessed into a compact propositional encoding that scales well with growing problem size. For repair, our compensation component executes compensation actions to restore the system to an earlier consistent state. The framework repairs failures through reconfiguration when monitoring and diagnosis use requirements. The reconfiguration component selects a best system reconfiguration that contributes most positively to the system's non-functional requirements. It selects a reconfiguration that achieves this while reconfiguring the system minimally. The framework does not currently offer a repair mechanism when monitoring and diagnosis use statecharts. We illustrate our framework with two medium-sized, publicly-available case studies. We evaluate the framework's performance through a series of experiments on randomly generated and progressively larger specifications. The results demonstrate that our approach scales well with problem size, and can be applied to industrial sized software applications.

Autonomic computing holds great promise for software systems of the future, but at the same time poses great challenges for Software Engineering. Autonomic computing research aims to design software systems that self-configure, self-repair, self-optimize and self-protect, so as to reduce software maintenance cost while improving performance. The aim of our research is to develop tool-supported methodologies for designing and operating autonomic systems. Like other researchers in this area, we assume that autonomic system architectures consist of monitoring, analysis/diagnosis, planning, and execution components that define a feedback loop and serve as the basis for system self-management. This thesis proposes an autonomic framework founded on models of requirements and design. This framework defines the normal operation of a software system in terms of models of its requirements (goal models) and/or operation (statechart models). These models determine what to monitor and how to interpret log data in order to diagnose failures. The monitoring component collects and manages log data. The diagnostic component analyzes log data, identifies failures, and pinpoints problematic components. We transform the diagnostic problem into a propositional satisfiability (SAT) problem solvable by off-the-shelf SAT solvers. Log data are preprocessed into a compact propositional encoding that scales well with growing problem size. For repair, our compensation component executes compensation actions to restore the system to an earlier consistent state. The framework repairs failures through reconfiguration when monitoring and diagnosis use requirements. The reconfiguration component selects a best system reconfiguration that contributes most positively to the system's non-functional requirements. It selects a reconfiguration that achieves this while reconfiguring the system minimally. The framework does not currently offer a repair mechanism when monitoring and diagnosis use statecharts. We illustrate our framework with two medium-sized, publicly-available case studies. We evaluate the framework's performance through a series of experiments on randomly generated and progressively larger specifications. The results demonstrate that our approach scales well with problem size, and can be applied to industrial sized software applications.

Modern embedded systems are evolving in the direction of increased adaptivity and complexity. It is extremely important for a system with limited resource to be adaptive in order to maximize its efficiency of resource usage while guaranteeing a high level of fault tolerance and QoS. This report aims at exploring such a kind of system, i.e. Adaptive Embedded System (AES), which is featured by dynamic reconfiguration at runtime. Based on the investigation and analysis of a variety of case studies related with AES, we proposed the conceptual view and overall architecture of an AES by highlighting its predominant characteristics. We also made an incomplete but detailed summary of the most popular techniques that can be used to realize adaptivity. Those techniques are categorized into dynamic CPU/network resource re-allocation and adaptive fault tolerance. A majority of adaptive applications resort to one or more of those techniques. Besides, there is a separate discussion on dynamic reconfiguration and mode switch for AES. Finally, we classify adaptivity into different modeling problems at a higher abstraction level and build UPPAAL models for two different AESs, a smart phone and an object-tracking robot. Our UPPAAL models provide clear demonstration on how a typical AES works.

The design and regulation of power equipments which installed in distribution system are based on single direction power flow. When distributed generators (DG) are added into distribution system, it may cause some technical problems such as two-way current, fault capacity and power quality. In general, the utility should make sure that its power system could be operated safely and reliably before integrating DG into the system. If there are no complete measurements for DG, the capacity of DG would be restricted by fault current, short circuit capacity, feeder voltage or other problems. In this research, the focus is on the influence of DG operations in distribution system and the increase of DG integration capacity. The impacts of different combinations of DG generation profiles and control strategies are first analyzed, followed by the use of particle swarm optimization (PSO) technique to search for better feeder reconfigurations in order to increase DG integration capacity.

Reproduction de : Thèse de doctorat : Traitement des images et du signal : Université de Cergy-Pontoise : 2006. / Titre provenant de l'écran titre. Bibliogr. p.183-186. Index.

Le travail réalisé dans le cadre de cette thèse a deux objectifs principaux. Le premier est de contribuer `à la spécification de la reconfiguration dynamique des systèmes à base de composants. Le deuxième objectif est de s´sélectionner la configuration optimale parmi un ensemble de configurations qui fournissent des fonctionnalités identiques ou similaires. Le processus de sélection dépend des propriétés non-fonctionnelles du système. La propriété de reconfigurabilité est essentielle pour de nombreux systèmes à base de composants contemporains. En effet, cette propriété améliore la disponibilité, l'adaptabilité, l'évolutivité, la maintenabilité et la performance des systèmes tels que les systèmes avioniques, les commutateurs de télécommunications et les systèmes commerciaux. Pour ces systèmes, l'arrêt de longue durée n'est pas admissible pour des raisons s´sécuritaires ou économiques. L'adaptabilité et l'evolvabilité sont également des caractéristiques importantes pour ces systèmes qui ont besoin d'inclure des changements de l'environnement ou des nouvelles exigences des utilisateurs dans le logiciel. Toutes ces motivations plus montrent l'importance de permettre, dès la conception, la reconfiguration dynamique de systèmes. La reconfiguration est la capacité de modifier la structure ou le comportement d'un système à l'exécution et sans l'arrêter complétement. Le travail présenté dans cette thèse étudie les mécanismes et les techniques pour fournir la reconfigurabilité aux systèmes à base de composants. La fourniture de reconfigurabilité nécessite la prise en considération de la cohérence du système pendant et après la reconfiguration. Il y a deux sortes de cohérence : cohérence globale et cohérence locale. Dans cette thèse, nous proposons une approche pour préserver la cohérence globale d'un système à base de composants reconfigurable en utilisant un langage formel déclaratif, Alloy. Une autre approche est proposée pour préserver la cohérence locale en analysant la relation entre la dépendance indirecte et la reconfiguration dynamique. Enfin, la sélection de configuration consiste à choisir la configuration la plus optimale à partir d'un ensemble de choix dans le but de maximiser la satisfaction de l'utilisateur. Une approche proposée pour faire le meilleur choix en fonction des préférences de l'utilisateur exprimées sur des métriques non-fonctionnelles / The research of this thesis has two main goals. The first goal is to provide the reconfigurability feature to the component-based systems. The second goal is to select the optimal configuration from a set of configurations, which provide similar functionality. The selection process depends on the non-functional properties of the system. Reconfigurability is essential feature for many contemporary component-based systems. Reconfigurability enhances the continuous availability, the adaptability, the evolvability, the maintainability, and the performance. Avionics systems, telecommunications switches and some commercial systems require the high availability. For these systems, long shutting down is not allowable due to economical or safety reasons. The adaptability and the evolvability are also important features for those systems which need to accommodate the environmental changes or the new requirements of software users. The maintainability and the performance are important requirements for a large category of systems. All the previous motivations and more show the importance of having the reconfigurability. Reconfigurability is the ability to change the system structure or the system behavior at running time without stopping it. The work presented in this thesis investigates the required mechanisms and techniques in order to provide the reconfigurability feature to a component-based system. The provision of the reconfigurability feature requires preserving the system consistency during and after the reconfiguration. The consistency has two kinds: global consistency and local consistency. In this thesis, we propose an approach to preserve the global consistency of a reconfigurable component-based system using declarative formal language. Another approach is proposed to preserve the local consistency during the reconfiguration. The second approach investigates the relationship between the indirect dependency and the dynamic reconfiguration. Configuration selection is to select the most optimal configuration from a set of alternatives in order to maximize the end user satisfaction. The thesis proposes an approach to make the best selection depending on the user preferences

The need to achieve extended service life by battery powered Wireless Sensor Networks (WSNs) requires new concepts and technqiues beyond the state-of-the-art low-power designs based on fixed hardware platforms or energy-efficient protocols. This thesis investigates reconfiguration techniques that enable sensor hardware to adapt its energy consumption to external dynamics, by means of Dynamic Voltage Scaling (DVS), Dynamic Modulation Scaling (DMS), and other related concepts. For sensor node-level reconfiguration, an integration of DVS and DMS techniques was proposed to minimize the total energy consumption. A dynamic time allocation algorithm was developed, demonstrating an average of 55% energy reduction. For network-level reconfiguration, a node activation technique was presented to reduce the cost of recharging energy-depleted sensor nodes. Network operation combined with node activation was modeled as a stochastic decision process, where the activation decisions directly affected the energy efficiency of the network. An experimental test bed based on the Imote2 sensor node platform was realized, which demonstrated energy reduction of up to 50%. Such energy saving can be effectively translated into prolonged service life of the sensor network.

Todays protection schemes for shipboard applications, such as the zone-based differential protection, are efficient, fast, and reliable for fault detection. However, these protection schemes do not consider the system stability or power balance problems that occur with fault isolation and the system reconfiguration. This thesis focuses on developing the extended protection function including the fast reconfiguration function that intends to maintain the power balance of the unfaulted subsystem. Graph theory is utilized to represent the shipboard power system topology in matrices, and matrix operations are developed to represent the corresponding power system topology change and evaluate the outcome of the fault. Intelligent search algorithms are implemented to find the possible system configuration after fault isolation with balanced power generation and load through merging possible connected systems and priority-based load shedding. The algorithms are successfully implemented in MATLAB miles and tested on various shipboard power system configurations and fault scenarios.

The shipboard power system supplies energy to sophisticated systems for weapons, communications, navigation, and operation. After a fault is encountered, reconfiguration of a shipboard power system becomes a critical activity that is required to either restore service to a lost load or to meet some operational requirements of the ship. Reconfiguration refers to changing the topology of the power system in order to isolate system damage and/or optimize certain characteristics of the system related to power efficiency. When finding the optimal state, it is important to have a method that finds the desired state within a short amount of time, in order to allow fast response for the system. Since the reconfiguration problem is highly nonlinear over a domain of discrete variables, the genetic algorithm method is a suitable candidate. In this thesis, a reconfiguration methodology, using a genetic algorithm, is presented that will reconfigure a network, satisfying the operational requirements and priorities of loads. Graph theory is utilized to represent the shipboard power system topology in matrices. The reconfiguration process and the genetic algorithm are implemented in MATLAB and tested on an 8-bus power system model and on larger power system with distributed generators by considering different fault scenarios. Each test system was reconfigured in three different ways: by considering load priority, without considering load priority, and by combining priority factor and magnitude factor. The test results accuracy was verified through hand checking.

Electrical power systems are vulnerable to contingencies due to the possibility of fault or damage. When a serious fault (which cannot be removed immediately) occurs in the power system, the fault needs to be isolated, and the remaining system should be reconfigured to make efficient use of available energy. Reconfiguring the system by making an island can be one of the strategies to continue the energy supply using the remaining generation capacity. By analyzing the effect of contingencies on power systems in terms of voltage support, reliability and stability, survivability of the system can be predicted. A survivability index of the system can be developed based on voltage support, reliability and the stability index with predefined weight given to each performance criterion. These weights may vary for different system conditions. The purpose of this research work is to develop a survivability index for shipboard power system and rank reconfiguration choices based on the developed index.