This thesis proposes a novel approach to power electronics system design that is based on the open-architecture distributed digital controller and modular power electronics building blocks (PEBBs). The proposed distributed digital controller partitions the controller in three levels of control authority. The power stage controller, designated as hardware manager, is responsible for low-level hardware oriented tasks; the high level controller, designated as applications manager, performs higher-level application-oriented tasks; and the system level controller handles system control and monitoring functions. Communications between the hardware-oriented controller and the higher-level controller are implemented with the previously proposed 125 Mbits/sec daisy-chained fiber optic communication protocol. Real-time control and status data are communicated by means of communication protocol. The distributed controller on the power converter level makes the system open, flexible and simple to use. Furthermore, this work gives an overview and comparison of current state-of-the-art communication protocols for real-time control applications with emphasis on industrial automation and motion control. All of the studied protocols have been considered as local area networks (LAN) for system-level control in power converter systems. The most promising solution has been chosen for the system level communication protocol. This thesis also provides the details of design and implementation of the distributed controller. The design of both the hardware and software components are explained. A 100 kVA three-phase voltage source inverter (VSI) prototype was built and tested using the distributed controller approach to demonstrate the feasibility of the proposed concept. / Master of Science

Thesis (M. Tech. (Information Technology)) -- Central University of technology, Free State, 2013 / Markets for manufactured products are characterized by a fragmentation of the market (with regards to size and time), and by shorter product cycles. This is due to the occurrence of mass customization and globalization. In mass customization, the same basic products are manufactured for a broad market, but then consumers are given the liberty to choose the “finishing touches” that go with the product. The areas that manufacturers now compete for are higher quality products, low cost and timely response to market changes. Appropriate business strategies and manufacturing technologies must thus be used to implement these strategic dimensions. The paradigm of Reconfigurable Manufacturing System (RMS) has been introduced to respond to this new market oriented manufacturing environment. The design of RMS allows ease of reconfiguration as it has a modular structure in terms of software and hardware. This allows ease of reconfiguration as a strategy to adapt to changing market demands. Modularity will allow the ability to integrate/remove software/hardware modules without affecting the rest of the system. RMS can therefore be quickly reconfigured according to the production requirements of new models, it can be quickly adjusted to exact capacity requirements as the market grows and products change, and it is able to integrate new technology. In this research project, real-time, open controller is designed and developed for Reconfigurable Manufacturing Tools (RMTs). RMTs are the basic building blocks for RMS. Real time and openness of the controllers for RMT would allow firstly, for the modular design of RMTs (so that RMTs can be adapted easily for changing product demands) and secondly, prompt control of RMT for diagnosability.

Le soudage par friction malaxage (FSW) est un procédé de soudage innovant pour les matériaux à bas point de fusion (aluminium, cuivre…). Il a été breveté en 1992 par l’organisme anglais The Welding Institute (TWI). Depuis plusieurs années, celui-ci se développe dans l’industrie en cherchant à réduire son coût d’investissement. Le principe du FSW est de réaliser un cordon de soudure grâce à un outil animé d’un mouvement de rotation et d’avance. Les niveaux d’efforts et de précision requis contraignent à l’utilisation de machines cartésiennes de grande envergure. L’utilisation des robots industriels est un moyen de réduire les coûts, mais ils ne sont pas conçus pour ce genre d’applications et leur inconvénient majeur réside dans leur manque de rigidité. Ainsi, lorsque l’outil entre en contact avec les pièces à assembler, celui-ci peut dévier de plusieurs millimètres dans différentes directions de l’espace, rendant la mise en oeuvre d’une compensation de la trajectoire du robot obligatoire afin d’obtenir des soudures sans défauts. Le but de cette thèse a été de développer un procédé robotisé robuste. Le premier objectif est la mise en oeuvre d’une commande en effort robuste. En effet, en FSW, le maintien d’un effort axial constant est obligatoire. Le contrôle de cet effort permet de compenser la déviation axiale de l’outil et les défauts de mise en position des pièces à souder. Ainsi, une démarche d’identification et de modélisation afin de créer une commande en effort a été mise en oeuvre. La commande est définie de manière robuste afin d’éviter les réglages de l’asservissement lorsque les outils, les paramètres de soudage ou les trajectoires du robot changent. Une validation expérimentale complète a été réalisée dans le contexte du FSW. Le second objectif de cette thèse a été de développer une compensation de la déviation latérale de l’outil. Contrairement à l’objectif précédent, il n’y a pas d’effort à maintenir pour compenser cette déviation latérale. Dans l’industrie, cette déviation peut être compensée à l’aide d’un système de vision, mais ce dernier comporte de nombreux inconvénients en FSW (réflexion de l’aluminium, non visibilité du joint, coût, mise en oeuvre complexe). Ainsi, dans cette partie, un algorithme de compensation temps réel de la déviation latérale de l’outil a été mis en oeuvre. Celui-ci repose sur l’identification d’un modèle élasto-statique du robot. L’algorithme de compensation de la déviation latérale de l’outil a été couplé à la commande en effort et validé expérimentalement en FSW. La différence avec la majorité des travaux de recherche dans ce domaine est que les procédures d’identification n’utilisent pas de système de mesure 3D (photogrammétrie CCD ou laser de poursuite) dont le coût est un frein indéniable pour beaucoup d’industriels. La démarche est simple à mettre en oeuvre sur un robot industriel du marché actuel, et applicable pour d’autres procédés à contact comme l’usinage ou le polissage. / Friction Stir Welding (FSW) is an innovative welding process for materials with a low melting point (aluminium, copper…). It was patented in 1992 by the English organization The Welding Institute (TWI). For many years, an effort is done to reduce the investment cost for industrial applications. FSW process involves a rotating tool advancing along a path. Currently, gantry-type CNC systems are using for FSW manufacturing. These machines offer a high stiffness and can tolerate the forces during FSW in order to carry out a good weld quality. Industrials robots can reduce the investment cost; however they are not design for these applications. The main limitation is the low stiffness of the robot structure. Consequently, the robot deformation under the high process forces causes tool deviations about several millimeters. The robot path has to be compensated in order to obtain a good weld quality. The aim of this thesis is to develop a robust robotized process. The first goal is to realize a robust force control. During FSW, a constant axial forging force should be applied. Axial tool deviation is compensated with the force control approach. In this way, a modeling and identification method is done in order to design a force controller. The force controller is robust because no tuning is required, even if welding parameters or robot paths change. An experimental validation in FSW is done. The second goal is to realize a compensation of the lateral tool deviation. Unlike the axial deformation, there is no force to maintain for compensate this deviation. In industry, the lateral tool deviation could be compensated with a camera or laser sensor in order to track the weld seam path during welding. However, the cost of a seam tracking device, the aluminium reflexion and the lack of visibility in lap joint configuration are significant drawbacks. In this chapter, a compensation algorithm is designed. An elastostatic model of the robot is used to estimate in real time the deflection of the robot TCP. The compensation algorithm is coupled with the force controller defined previously. Compare with others research works about this topic, identification methods don’t need a 3D measurement system (CCD camera or laser tracker). The cost of such system is a main drawback for industrial applications. In this thesis, identification methods are easy to implement in an industrial robot and available for others processes like machining or polishing.

Abstract not available

In this thesis, a test platform based on real-time facilities and embedded software is designed to verify the performance of a controller model in real time. By the help of this platform, design errors can be detected earlier and possible problems can be solved cost-effectively without interrupting the development process. An unmanned combat air vehicle (UCAV) model is taken as a plant model due to its importance in current and future military operations. Among several autopilot modes, the altitude hold mode is selected since it is an important pilot-relief mode and widely used in aviation. A discrete PID controller is designed in MATLAB/Simulink environment for using in verification studies. To control the dynamic system in wide range, a gain scheduling is employed where the altitude and velocity are taken as scheduling variables. Codes for plant and controller model are obtained by using real time workshop embedded coder (RTWEC) and downloaded to two separate computers, in which xPC kernel and VxWorks operating system are run, respectively. A set of flight test scenarios are generated in Simulink environment. They are analyzed, discussed, and then some of them are picked up to verify the platform. These test scenarios are run in the setup and their results are compared with the ones obtained in Simulink environment. The reusability of the platform is verified by using a commercial aircraft, Boeing 747, and its controller models. The test results obtained in the setup and in Simulink environment are presented and discussed.

This work focuses on modeling, parameter estimation, and real-time position control of a two axis Gimbaled Mirror System (GMS) which is designed and manufactured to move an IR spot generated by an Infra Red Scene Generator System (IRSGS) in two orthogonal axes (elevation and azimuth) within the IR scene which is also generated by the IRSGS. Mathematical models of the GMS, the control system, and the disturbance torque originated from the movements of Flight Motion Simulator (FMS), on which the IRSGS will be mounted, are constructed using MATLAB&reg / /Simulink&reg / and MATLAB/Simulink/SimMechanics&reg / . Parameter estimations of the GMS and control system elements are achieved using MATLAB/Simulink Parameter Estimation Tool&reg / . The controller tuning is performed using the developed mathematical models in MATLAB/Simulink environment. Optimized digital PID controllers are implemented in the real-time control system. Performances of the controllers for both GMS axes are evaluated by both real system tests and simulation runs / and the results of these runs are compared. Controller performances under the effect of disturbances are analyzed by using the mathematical models developed in the MATLAB/ Simulink environment.

In semiconductor manufacturing, wafers are grouped into lots and sent to a separate facility for assembly and test (AT) before being shipped to the customer. Up to a dozen operations are required during AT. The facility in which these operations are performed is a reentrant flow shop consisting of several dozen to several hundred machines and up to a thousand specialized tools. Each lot follows a specific route through the facility, perhaps returning to the same machine multiple times. Each step in the route is referred to as a "pass." Lots in work in process (WIP) that have more than a single step remaining in their route are referred to as multi-pass lots. The multi-pass scheduling problem is to determine machine setups, lot assignments and lot sequences to achieve optimal output, as measured by four objectives related to key device shortages, throughput, machine utilization, and makespan, prioritized in this order. The two primary goals of this research are to develop a new formulation for the multipass problem and to design a variety of solution algorithms that can be used for both planning and real-time control. To begin, the basic AT model considering only single-pass scheduling and the previously developed greedy randomized adaptive search procedure (GRASP) along with its extensions are introduced. Then two alternative schemes are proposed to solve the multipass scheduling problem. In the final phase of this research, an efficient procedure is presented for prioritizing machine changeovers in an AT facility on a periodic basis that provides real-time support. In daily planning, target machine-tooling combinations are derived based on work in process, due dates, and backlogs. As machines finish their current lots, they need to be reconfigured to match their targets. The proposed algorithm is designed to run in real time. / text

This thesis describes the development of a decision support tool to be used in the operation of Vanderkloof Dam on the Orange River so that the supply of water to the lower Orange River can be optimised. The decision support tool is based on a hydrodynamic model that was customised to incorporate real time data recorded at several points on the river. By incorporating these data into the model the simulated flows are corrected to the actual flow conditions recorded on the river, thereby generating a best estimate of flow conditions at any given time. This information is then used as the initial conditions for forecast simulations to assess whether the discharge volumes and schedules from the dam satisfy the water demands of downstream users, some of which are 1400km or up to 8 weeks away. The various components of the decision support system, their functionality and their interaction are described. The details regarding the development of these components include: • The hydraulic model of the Orange River downstream of Vanderkloof Dam. The population and calibration of the model are described. • The modification of the code of the hydrodynamic engine so that real time recorded stage and flow data can be incorporated into the model • The development of a graphical user interface to facilitate the exchange of data between the real time network of flow gauging stations on the Orange River and the hydraulic model • The investigation into the effect of including the real time data on the simulated flows • Testing the effectiveness of the decision support system. / Thesis (M.Sc.)-University of Natal, Durban, 2002.

Thesis (Ph. D.)--Aerospace Engineering, Georgia Institute of Technology, 2008. / Committee Chair: Johnson, Eric; Committee Co-Chair: Calise, Anthony; Committee Member: Prasad, J.V.R.; Committee Member: Tannenbaum, Allen; Committee Member: Tsiotras, Panagiotis.

Alors que les préoccupations concernant le réchauffement climatique deviennent de plus en plus sérieuses, l'une de ses premières causes: la consommation d'électricité, continue à croître. Une des manières d'endiguer le phénomène pourrait être de mieux équilibrer la consommation et la production, afin d'allumer moins de gros groupes de production, et de permettre l'intégration de plus de sources renouvelables. Le nouveau paradigme du marché de l'électricité incite les consommateurs à réduire leur pic de consommation, et à différer leur consommation quand la demande est moindre, à l'aide d'incitations tarifaires. De nouveaux algorithmes d'optimisation, et de nouvelles méthodologies sont donc nécessaires pour optimiser la puissance d'électrique utilisée par les consommateurs. Schneider Electric propose, à travers le projet européen Arrowhead, d'étudier trois cas applicatifs : un ascenseur avec plusieurs sources énergétiques, une usine manufacturière et un réseau d'eau potable. Pour chacune de ces trois applications, une méthodologie pour optimiser les pics de puissance consommée (parfois à l'aide d'une fonction de coût de l'électricité) est donnée, ainsi que des algorithmes d'optimisation. Dans le cas de l'ascenseur multi-sources, deux contrôleurs couplés sont proposés : l'un au niveau stratégique résolvant un problème linéaire, et l'autre à base de règles au niveau tactique. Dans le cas de l'usine, la méthodologie que nous avons utilisée pour faire des mesures, construire des modèles énergétiques, et finalement optimiser est expliquée. De plus, trois formulations linéaires, ainsi qu'une procédure de recherche locale, et une formulation naïve de programmation par contraintes sont données afin de résoudre le problème d'ordonnancement NP-difficile. Dans le cas du réseau d'eau, une formulation à contraintes quadratiques est utilisée pour comparer des plans de pompage optimisés avec la tactique utilisée habituellement dans le réseau. Les méthodes proposées entraînent des gains sur la facture énergétique de 1.5% à 114%, dépendamment du contexte. De plus, elles permettent aux consommateurs de participer au nouveau marché de l'énergie. Finalement, les connaissances retirées de l'étude de ces trois pilotes sont résumées, et des lignes directrices sont données pour l'optimisation de la facture énergétique d'un système quelconque consommant de l'énergie. / While concerns about global warming have never been so important, one of its first causes: global electricity consumption, is still growing. One way to stem the phenomenon could be to better balance demand and production, in order to switch on less big production groups and to allow the integration of more renewable production sources. The new paradigm of electricity market incites customers to reduce their electricity consumption peak and to shift their consumption when the demand is lower, by introducing economical incentives. Thus, new optimization algorithms and methodologies are needed at the customers side to optimize power usage over time. Schneider Electric proposes, through the Arrowhead European project, to study three application use-cases: an elevator with multiple electricity sources, a manufac- turing plant, and a drinking water network. For each of these use-cases, a methodology to optimize power consumption peaks (sometimes through an electricity cost function) is given, as well as optimization algorithms. For the multisource elevator case, two coupled controllers are proposed: one at the strategic level solving a linear problem, the other one rule-based at the tactical level. For the manufacturing plant, the methodology we used to monitor, build energy models, and finally optimize is explained. Furthermore, three linear formulations are given, as well as a simple local search procedure and a naive constraint satisfaction formulation to handle the NP-hard scheduling problem. For the water network use-case, a quadratically constrained formulation is used to compare optimized pumping plans with the Business As Usual tactic. The methods proposed bring between 1.5% to 114% savings on the energy bill, depending on the context. Moreover, they allow electricity consumers to participate in the demand-response market. Finally, the knowledge extracted from our three use-cases is summarized, and guide- lines are given to optimize the electricity bill of any electricity consumer system.