Three-Phase Voltage Source Converters (VSCs) are finding widespread applications in grid integrated power conversion systems. The control systems of such VSCs are in an increasing number of these applications required to operate during voltage disturbances and unbalanced conditions. Control systems designed for grid side voltagesensor- less operation are at the same time becoming attractive due to the continuous drive for cost reduction and increased reliability of VSCs, but are not commonly applied for operation during unbalanced conditions. Methods for voltage-sensor-less grid synchronization and control of VSCs under unbalanced grid voltage conditions will therefore be the main focus of this Thesis. Estimation methods based on the concept of Virtual Flux, considering the integral of the converter voltage in analogy to the flux of an electric machine, are among the simplest and most well known techniques for achieving voltage-sensor-less grid synchronization. Most of the established techniques for Virtual Flux estimation are, however, either sensitive to grid frequency variations or they are not easily adaptable for operation under unbalanced grid voltage conditions. This Thesis addresses both these issues by proposing a simple approach for Virtual Flux estimation by utilizing a frequency-adaptive filter based on a Second Order Generalized Integrator (SOGI). The proposed approach can be used to achieve on-line frequency-adaptive varieties of conventional strategies for Virtual Flux estimation. The main advantage is, however, that the SOGI-based Virtual Flux estimation can be arranged in a structure that achieves inherent symmetrical component sequence separation under unbalanced conditions. The proposed method for Virtual Flux estimation can be used as a general basis for voltage-sensor-less grid synchronization and control during unbalanced conditions. In this Thesis, the estimated Virtual Flux signals are used to develop a flexible strategy for control of active and reactive power flow, formulated as generalized equations for current reference calculation. A simple, but general, implementation is therefore achieved, where the control objective and the power flow characteristics can be selected according to the requirements of any particular application. Thus, the same control structure can be used to achieve for instance balanced sinusoidal currents or elimination of double frequency active power oscillations during unbalanced conditions. In case of voltage sags, current references corresponding to a specified active or reactive power flow might exceed the current capability of the converter. The limits for active and reactive power transfer during unbalanced conditions have therefore been analyzed, and generalized strategies for current reference calculation when operating under current limitations have been derived. The specified objectives for active and reactive power flow characteristics can therefore be maintained during unbalanced grid conditions, while the average active and reactive power flow is limited to keep the current references within safe values. All concepts and techniques proposed in this Thesis have been verified by simulations and laboratory experiments. The SOGI-based method for Virtual Flux estimation and the strategies for active and reactive power control with current limitation can also be easily adapted for a wide range of applications and can be combined with various types of inner loop control structures. Therefore, the proposed approach can potentially be used as a general basis for Virtual Flux-based voltage-sensor-less operation of VSCs under unbalanced grid voltage conditions.

Apparu en 1970, le calcul parallèle permet, contrairement aux applications classiques qui exécutent un algorithme de manière séquentielle, d'exécuter des tâches d'une même application sur plusieurs processeurs en même temps. Les premières architectures -- les supercalculateurs -- qui regroupaient des milliers de processeurs au sein de la même machine, ont fait place aux grappes, à la fin des années 1970 : une interconnexion d'ordinateurs standard par un réseau rapide. Ces architectures s'étant développées un peu partout, les grilles ont fait leur apparition au début des années 1990, de manière à fédérer les ressources de différentes entités en les interconnectant et ainsi disposer d'une plus grande puissance de calcul globale. La grille, telle que nous la considérons dans ce manuscrit sera donc définie comme une interconnexion de grappes par un réseau longue distance.<br /> Les applications parallèles s'appuient la plupart du temps sur le standard MPI qui fonctionne par passage de message. Initialement destiné aux grappes, celui-ci est toujours utilisé pour programmer les communications des applications s'exécutant sur les grilles. Cela permet la réutilisation d'anciennes applications.<br /> Alors que différents problèmes ont été résolus pour les communications au sein des grappes, le réseau longue distance de la grille pose plusieurs problèmes. Tout d'abord, les messages MPI sont transmis de manière fiable sur le réseau longue distance via le protocole TCP. Or TCP, qui reste le protocole de transport utilisé dans la plupart des grilles, est basé sur un transfert de données à l'aide de flux ; il est donc peu adapté aux communications MPI. Ensuite, la grande latence du réseau longue distance implique des communications et des retransmissions de paquets perdus qui sont coûteuses. Enfin, le débit disponible sur le lien d'accès à ce réseau est généralement inférieur à la somme des débits nécessaires si tous les processus communiquent en même temps sur ce lien. Ceci crée de la congestion à la fois au sein d'une même application et à la fois avec les autres applications qui l'utilisent, et il devient nécessaire de gérer ce goulot d'étranglement.<br /> L'objectif principal de cette thèse est d'étudier en détail les interactions entre les applications parallèles et la couche de transport dans les réseaux longue distance des grilles de calcul, puis de proposer des solutions à ces problèmes.

This study on the neutron-induced fission of 234U was carried out at the 7 MV Van de Graaff accelerator of IRMM in Belgium. A Twin Frisch-Grid Ionization Chamber (TFGIC) was used to study 234U(n,f) between En = 0.2 and 5.0 MeV. The reaction is important for fission modelling of the second-chance fission in 235U(n,f). The fission fragment (FF) angular-, energy and mass distributions were determined using the 2E-method highlighting especially the region of the vibrational resonance at En = 0.77 MeV. The experiment used both conventional analogue and modern digital acquisition systems in parallel. Several advantages were found in the digital case, especially a successful pile-up correction. The shielding limitations of the Frisch-grid, called "grid-inefficiency", result in an angular-dependent energy signal. The correction of this effect has been a long-standing debate and a solution was recently proposed using the Ramo-Shockley theorem. Theoretical predictions from the latter were tested and verified in this work using two different grids. Also the neutron-emission corrections as a function of excitation energy were investigated. Neutron corrections are crucial for the determination of FF masses. Recent theoretical considerations attribute the enhancement of neutron emission to the heavier fragments exclusively, contrary to the average increase assumed earlier. Both methods were compared and the impact of the neutron multiplicities was assessed. The effects found are significant and highlight the importance of further experimental and theoretical investigation. In this work, the strong angular anisotropy of 234U(n,f ) was confirmed. In addition, and quite surprisingly, the mass distribution was found to be angular-dependent and correlated to the vibrational resonances. The anisotropy found in the mass distribution was consistent with an anisotropy in the total kinetic energy (TKE), also correlated to the resonances. The experimental data were parametrized assuming fission modes based on the Multi-Modal Random Neck-Rupture model. The resonance showed an increased yield from the Standard-1 fission mode and a consistent increased TKE. The discovered correlation between the vibrational resonances and the angular-dependent mass distributions for the asymmetric fission modes may imply different outer fission-barrier heights for the two standard modes.

Sensor network web services have recently emerged as promising tools to provide remote management, data collection and querying capabilities for sensor networks. They can be utilized in a large number of elds among which Demand-Side Energy Management (DSEM) is an important application area that has become possible with the smart electrical power grid. DSEM applications generally aim to reduce the cost and the amount of power consumption. In the traditional power grid, DSEM has not been implemented widely due to the large number of households and lack of ne-grained automation tools. However by employing intelligent devices and implementing communication infrastructure among these devices, the smart grid will renovate the existing power grid and it will enable a wide variety of DSEM applications. In this thesis, we analyze various DSEM scenarios that become available with sensor network web services. We assume a smart home with a Wireless Sensor Network (WSN) where the sensors are mounted on the appliances and they are able to run web services. The web server retrieves data from the appliances via the web services running on the sensor nodes. These data can be stored in a database after processing, where the database can be accessed by the utility, as well as the inhabitants of the smart home. We showthat our implementation is e cient in terms of running time. Moreover, the message sizes and the implementation code is quite small which makes it suitable for the memory-limited sensor nodes. Furthermore, we show the application scenarios introduced in the thesis provide energy saving for the smart home.

In recent years, significant advances have been achieved in communication networks and electric power systems. Communication networks are developed to provide services within not only well-connected network environments such as wireless local area networks, but also challenged network environments where continuous end-to-end connections can hardly be established between information sources and destinations. Delay tolerant network (DTN) is proposed to achieve this objective by utilizing a store-carry-and-forward routing scheme. However, as the network connections in DTNs are intermittent in nature, the management of network resources such as communication bandwidth and buffer storage becomes a challenging issue. On the other hand, the smart grid is to explore information and communication technologies in electric power grids to achieve electricity delivery in a more efficient and reliable way. A high penetration level of electric vehicles and renewable power generation is expected in the future smart grid. However, the randomness of electric vehicle mobility and the intermittency of renewable power generation bring new challenges to the resources management in the smart grid, such as electric power, energy storage, and communication bandwidth management. This thesis consists of two parts. In part I, we focus on the resource management in DTNs. Specifically, we investigate data dissemination and on-demand data delivery which are two of the major data services in DTNs. Two kinds of mobile nodes are considered for the two types of services which correspond to the pedestrians and high-speed train passengers, respectively. For pedestrian nodes, the roadside wireless local area networks are used as an auxiliary communication infrastructure for data service delivery. We consider a cooperative data dissemination approach with a packet pre-downloading mechanism and propose a double-loop receiver-initiated medium access control scheme to resolve the channel contention among multiple direct/relay links and exploit the predictable traffic characteristics as a result of packet pre-downloading. For high-speed train nodes, we investigate on-demand data service delivery via a cellular/infostation integrated network. The optimal resource allocation problem is formulated by taking account of the intermittent network connectivity and multi-service demands. In order to achieve efficient resource allocation with low computational complexity, the original problem is transformed into a single-machine preemptive scheduling problem and an online resource allocation algorithm is proposed. If the link from the backbone network to an infostation is a bottleneck, a service pre-downloading algorithm is also proposed to facilitate the resource allocation. In part II, we focus on resource management in the smart grid. We first investigate the optimal energy delivery for plug-in hybrid electric vehicles via vehicle-to-grid systems. A dynamic programming formulation is established by considering the bidirectional energy flow, non-stationary energy demand, battery characteristics, and time-of-use electricity price. We prove the optimality of a state-dependent double-threshold policy based on the stochastic inventory theory. A modified backward iteration algorithm is devised for practical applications, where an exponentially weighted moving average algorithm is used to estimate the statistics of vehicle mobility and energy demand. Then, we propose a decentralized economic dispatch approach for microgrids such that the optimal decision on power generation is made by each distributed generation unit locally via multiagent coordination. To avoid a slow convergence speed of multiagent coordination, we propose a heterogeneous wireless network architecture for microgrids. Two multiagent coordination schemes are proposed for the single-stage and hierarchical operation modes, respectively. The optimal number of activated cellular communication devices is obtained based on the tradeoff between communication and generation costs.

With the rapid growth of data and computational needs, distributed systems and computational Grids are gaining more and more attention. Grids are playing an important and growing role in today networks. The huge amount of computations a Grid can fulfill in a specificc time cannot be done by the best super computers. However, Grid performance can still be improved by making sure all the resources available in the Grid are utilized by a good load balancing algorithm. The purpose of such algorithms is to make sure all nodes are equally involved in Grid computations. This research proposes two new distributed swarm intelligence inspired load balancing algorithms. One is based on ant colony optimization and is called AntZ, the other one is based on particle swarm optimization and is called ParticleZ. Distributed load balancing does not incorporate a single point of failure in the system. In the AntZ algorithm, an ant is invoked in response to submitting a job to the Grid and this ant surfs the network to find the best resource to deliver the job to. In the ParticleZ algorithm, each node plays a role as a particle and moves toward other particles by sharing its workload among them. We will be simulating our proposed approaches using a Grid simulation toolkit (GridSim) dedicated to Grid simulations. The performance of the algorithms will be evaluated using several performance criteria (e.g. makespan and load balancing level). A comparison of our proposed approaches with a classical approach called State Broadcast Algorithm and two random approaches will also be provided. Experimental results show the proposed algorithms (AntZ and ParticleZ) can perform very well in a Grid environment. In particular, the use of particle swarm optimization, which has not been addressed in the literature, can yield better performance results in many scenarios than the ant colony approach.

To ensure the robustness of an integrated circuit design, its power distribution network (PDN) must be validated beforehand against any voltage drop on VDD nets. However, due to the increasing size of PDNs, it is becoming difficult to verify them in a reasonable amount of time. Lately, much work has been done to develop Model Order Reduction (MOR) techniques to reduce the size of power grids but their focus is more on simulation. In verification, we are concerned about the safety of nodes, including the ones which have been eliminated in the reduction process. This work proposes a novel approach to systematically reduce the power grid and accurately compute an upper bound on the voltage drops at power grid nodes which are retained. Furthermore, a criterion for the safety of nodes which are removed is established based on the safety of other nearby nodes and a user specified margin.

To ensure the robustness of an integrated circuit design, its power distribution network (PDN) must be validated beforehand against any voltage drop on VDD nets. However, due to the increasing size of PDNs, it is becoming difficult to verify them in a reasonable amount of time. Lately, much work has been done to develop Model Order Reduction (MOR) techniques to reduce the size of power grids but their focus is more on simulation. In verification, we are concerned about the safety of nodes, including the ones which have been eliminated in the reduction process. This work proposes a novel approach to systematically reduce the power grid and accurately compute an upper bound on the voltage drops at power grid nodes which are retained. Furthermore, a criterion for the safety of nodes which are removed is established based on the safety of other nearby nodes and a user specified margin.

In this body of work, we present the details of a novel method for passive resource discovery in cluster grid environments where resources constantly utilize inter-node communication. Our method offers the ability to non-intrusively identify resources that have available memory or CPU cycles; this is critical for lowering queue wait times in large cluster grid networks, and for memory-intensive cluster grid applica-tions such as Gaussian (computational chemistry package) and the Weather Research and Forecasting (WRF) modeling package. The benefits include: (1) low message complexity, (2) scalability, (3) load bal-ancing support, and (4) low maintainability. Using several test-beds (i.e., a small local test-bed and a 50-node Deterlab test-bed), we demonstrate the feasibility of our method with experiments utilizing TCP, UDP and ICMP network traffic. Using this technique, we observed a correlation between memory or CPU load and the timely response of network traffic. In such situations, we have observed that in highly utilized (due to multi-programming) nodes there will be numerous, active processes which require context switching or paging. The latency associated with numerous context switches or paging manifests as a de-lay signature within the packet transmission process. Our method detects this delay signature to determine the utilization of network resources. The aforementioned delay signature is the keystone that provides a correlation between network traffic and the internal state of the source node. We characterize this delay signature due to CPU utilization by (1) identifying the different types of assembly language instructions that source this delay and (2) describing how performance-enhancing techniques (e.g., instruction pipelin-ing, caching) impact this delay signature by using the LEON3, implemented as a 40 MHz development board. At the software level, results for medium sized networks show that our method can consistently and accurately identify nodes with available memory or CPU cycles (< 70% availability). At the hardware level, our results show that excessive context switching in active applications increases the average mem-ory access time, thus adding additional delay to the execution of LD instructions. Additionally, internal use of these instructions in heavily utilized situations to send network packets induces the delay signature into network traffic.

Current Source Inverter (CSI) topology is gaining acceptance as a competitive alternative for grid interface of renewable energy systems due to its unique and advantageous features. Merits of CSI over the more popular Voltage Source Inverter (VSI) topology have been elaborated on by a number of researchers. However, there is a dearth of quality work in modeling and control of CSI topology interfacing renewable energy resources to the grid. To enrich the study focussing on application of CSI for renewable energy interface, this thesis develops a multilevel structure based on CSI for three-phase grid-connected Photovoltaic (PV) application. In the first part of research, a single-stage CSI interfacing to PV array is developed. The CSI-based PV system is equipped with Maximum Power Point Tracker (MPPT), DC-link current controller, and AC-side current controller. To eliminate the nonlinearity introduced by the PV array, a feed-forward control is introduced in the DC-link current controller. The AC-side current controller is responsible for maintaining unity power factor at the Point of Common Coupling (PCC). To verify the performance of the developed CSI-based PV system, a number of simulation studies are carried out in PSCAD/EMTDC environment. To illustrate the performance of the CSI-based PV system during transients on the grid side, simulation studies are carried out for four kinds of faults. Results obtained from fault studies are highly in favor of CSI topology and provide illustrative evidence for short-circuit current protection capability of the CSI. On the other hand, the VSI-based PV system performs poorly when subjected to similar grid transients. To extend the research on CSI-based PV system further, a multilevel structure based on CSI is developed. The multilevel structure is a parallel combination of $n$ CSI units and capable of producing $2n+1$ levels of current at the terminal of the inverter. Each unit in the multilevel structure has its own MPPT, DC-link current controller. However, on the AC-side a combined current controller is proposed. The design results in a high power rating with reduced number of filters, sensors and controllers. The developed multilevel structure can operate with PV arrays exposed to equal and unequal insolation level. However, when the PV arrays are operating under unequal insolation level, low order harmonics are generated in the sinusoidal current that is injected into the grid. Elimination of these harmonics is performed by implementing a modified control strategy in stationary reference frame that corresponds to the harmonic component that needs to be minimized. The modified control strategy operates in coordination with the existing DC-side and AC-side current controllers, and MPPTs. Therefore, real-time suppression of current harmonics can be ensured. Performance of the multilevel structure is verified by different transient studies.