kriptia.com
Búsqueda personalizada



Home > MATHEMATICS > OPERATIONAL RESEARCH >

CONTROL SYSTEMS

Español | Français | Deutsche
7 theses in 1 pages: 1
  • ANALYSIS OF THE DYNAMICS OF POWER ELECTRONIC CONVERTERS USING PWM BASED ON AVERAGING ZERO DYNAMICS OF THE ERROR (ZAD)
    Author: ANGULO GARCIA FABIOLA.
    Year: 2003.
    University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
    Place of defense: Aula 28.8 de l'ETSEIB Campus Sud.
    Place of preparation: ETSEIB, EDIFICI H Campus SUD.
    Summary: In this thesis the behaviour of the buck converter, driven with PWM and Zero Average Dynamic on error (ZAD) technique is studied in an analytical and numerical form. This technique for evaluating duty cycle in a PWM was reported for the first time in the literature in the year 2001. ZAD is widely studied in this thesis. Its main characteristics are: fixed frequency switching, robustness, and low error in the regulation case. These characteristics are very important in power converters. The main issues of this thesis are: To calculate the equilibrium point associated to the buck converter Poincaré map. To determine the stability of the 1âperiodic orbit, based on Lyapunov exponents, Floquet exponents and characteristic multipliers. Also an scheme of PWM with a delay time has been studied finding its stability limit. A study of transition to chaos has been made. It has confirmed analytically that the first bifurcation is flip type, the second bifurcation is corner collision type and third bifurcation is flip type, again. Also a controller for chaos is designed and tested and it has been compared with TDAS technique, showing lower error and very fast response. Applying Average theory, a bound for error in the regulation case and in stationary state has been found. This bound confirms the advantages of the ZAD technique. Finally the analytical and numerical results are confirmed in an experimental form.
  • METHODS GEOMETRIC CONTROL PROBLEMS IN OPTICAL UNIQUE. FUNDAMENTALS AND APPLICATIONS
    Author: DELGADO TELLEZ DE CEPEDA MARINA.
    Year: 2003.
    University: CARLOS III DE MADRID [www.uc3m.es].
    Place of defense: ESCUELA POLITECNICA SUPERIOR.
    Place of preparation: UNIVERSIDAD CARLOS III DE MADRID.
    Summary: This report deals with optimal control problems from the point of view geometric. The introduction outlines the problem from a historical point of view, based on the Principle of Maximum Pontryaguin enunciated in 1962. The second chapter establishes the framework geometric appropriate to the problem is the difference between optimal control problem of regular and singular and finally performs the calculus of variations in the appropriate spaces. In the third chapter works with the algorithm ligatures from the more general point of view, as a differential equation implicit unique to apply to the problem of optimal control singular seen as a differential equation implicit cuasilineal presimpléctica. It concludes with the Hamiltonian formulation of the algorithm incorporating calculating ligatures first and second class. It performs the calculation of comprehensive algorithm for the LQ problem and proposes a numerical algorithm, we bring empirical results very satisfactory. In the fourth chapter presents another algorithm, and eventually work for three unique problems more generally, gather the best braking of a car, and the problem of implicit control, showing results, samples, and numerical simulations.
  • NONLINEAR CONTROL OF UNCERTAIN SYSTEMS: SOME APPLICATION-ORIENTED ISSUES.
    Author: POZO MONTERO FRANCESC.
    Year: 2004.
    University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
    Place of defense: Escola Universitària d'E.T.I. de Barcelona,.
    Place of preparation: U FACULTAT DE MATEMATIQUES I ESTADISTICA SUD.
  • PREDICTIVE CONTROL FOR REPETITIVE PROCESSES.
    Author: RAMOS CUELI JOSÉ.
    Year: 2004.
    University: SEVILLA [www.us.es].
    Place of defense: ESCUELA SUPERIOR DE INGENIEROS.
    Place of preparation: ESCUELA SUPERIOR DE INGENIEROS.
    Summary: The purpose of this work is to obtain improvements in the control of two distinct types of processes. These processes can be defined around a point of operation and subjected to periodic disturbance type, or batch type processes or in batches. The problem has arisen in control applications such as the preparation phase of the extraction process of olive oil or a large number of industrial processes that operate batch (chemical polymerization, microelectronics, production of medicines, etc..). To improve control of these processes are designed optimal controllers, based on the application of the methodology of model-based predictive control in both cases. The Process Control subjected to periodic disturbances can be addressed, among other techniques, by an IMPC to which is added a block prediction optimal disturbance. In the case of genetic batch processes have been incorporated into the driver's ability to learn from previous batches using past information. The proposed controllers have been tested in simulation, linear and nonlinear, as well as real-world applications: a plant laboratory and the mill.
  • MODELING, SIMULATION AND CONTROL OF DOUBLY-FED INDUCTION MACHINE CONTROLLED BY BACK-TO-BACK CONVERTER
    Author: DORIA CEREZO ARNAU.
    Year: 2006.
    University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
    Place of defense: AULA 28.8, ETSEIB. CAMPUS SUD.
    Place of preparation: ETSEIB, EDIFICI H Campus SUD.
    Summary: This Thesis studies a complex multidomain system, the Flywheel Energy Storage System, including the control objectives specification, modeling, control design, simulation, experimental setup assembling and experimental validation stages. The port interconnection and control of electromechanical systems is studied. The port Hamiltonian formalism is presented in general, and particularized for generalized electromechanical systems, including variable structure systems (VSS). Interconnection and damping assignment-passivity based control (IDA-PBC) is a well known technique for port Hamiltonian systems (PCHS). In this Thesis we point out the kind of problems that can appear in the closed-loop structure obtained by IDA-PBC methods for relative degree one outputs, when nominal values are used in a system with uncertain parameters. To correct this, we introduce an integral control, which can be cast into the Hamiltonian framework. This Thesis also presents two new approaches which improve the range of applicability of the IDA-PBC technique. First, we show that the standard two-stage procedure used in IDA-PBC consisting of splitting the control action into the sum of energy-shaping and damping injection terms is not without loss of generality, and effectively reduces the set of systems that can be stabilized with IDA-PBC. To overcome this problem we suggest to carry out simultaneously both stages and refer to this variation of the method as SIDA-PBC. Secondly, we present an improvement of the IDA-PBC technique. The IDA-PBC method requires the knowledge of the full energy (or Hamiltonian) function. This is a problem because, in general, the equilibrium point which is to be regulated depends on uncertain parameters. We show how select the target port-Hamiltonian structure so that this dependence is reduced. This new approach allows to improve the robustness for higher relative degree outputs. The Flywheel Energy Storage System consists of a doubly-fed induction machine (DFIM), controlled through the rotor voltage by a power electronics subsystem (a back-to-back AC/AC converter (B2B)), and coupled to flywheel. The control objective is to optimally regulate the power flow between the DFIM and a local load connected to the grid, and this is achieved by commuting between different steady-state regimes. A police management based on the optimal speed for the DFIM is proposed. In this Thesis we propose a new control scheme for the DFIM that offers significant advantages, and is considerably simpler, than the classical vector control method. This controller allows an easy decomposition of the active and reactive powers on the stator side and their regulation, acting on the rotor voltage, via stator current control. This design was obtained applying the new robust IDA-PBC procedure. Other controllers are also designed along the dissertation. The classical vector control is studied. We also apply the classic IDA-PBC technique. It is shown that the partial differential equation that appears in this method can be circumvented by fixing the desired closed-loop total energy and adding new terms to the interconnection structure. Furthermore, to obtain a globally defined control law we introduce a state--dependent damping term that has the nice interpretation of effectively decoupling the electrical and mechanical parts of the system. This results in a globally convergent controller parameterized by two degrees of freedom. Finally, we also prove that with SIDA-PBC we can shape the total energy of the full (electrical and mechanical) dynamics of the DFIM. These different controllers (vector control, IDA-PBC, SIDA-PBC and robust IDA-PBC) are simulated and compared. The IDA-PBC robust controller is also experimentally tested and shown to work satisfactorily.
  • TRACKING ALGORITHMS FOR PRE-CRASH SYSTEMS USING ONLY RANGE MEASUREMENTS
    Author: SANS SANGORRIN JORGE.
    Year: 2006.
    University: JAUME I DE CASTELLON [www.uji.es].
    Place of defense: E.S.DE TECNO.Y CIE. EXPERIMENTALES.
    Place of preparation: E.S.DE TECNO.Y CIE. EXPERIMENTALES.
    Summary: This dissertation takes part of current research and development activities aimed to implement future and innovative safety related and advanced driver assistance applications that use vehicle surround sensing systems. In this research, algorithms that compute information to assess whether an accident may happen, i.e. prior to impact, have been developed and implemented, while taking the functional requirements of the applications into account. These algorithms can be used to implement Pre-Crash applications which are aimed to predict crashes, in order to allow activation of vehicle safety measures for reducing damages in road accidents. By making use of information about the vehicle surroundings provided by a sensing system, the approach implemented in this research allows the Pre-Crash system to track objects and perform predictions of crashes at the vehicle front. Impending frontal crashes can be detected and, in such cases, closing-velocity, time-to-impact and impact location estimates are computed. Algorithms have been properly implemented with a round-off to integer, satisfying the restrictions in terms of computational resources. Experiments are carried out in typical scenarios for validation of Pre-Crash applications using a sensing system consisting on two short range radars mounted in the vehicle front. Relevant contributions to internal projects at Robert Bosch GmbH, as well as to European projects with the goal of reducing road fatalities, which is a major concern of the European Commission, have been provided.
  • A CONCEPTUAL MODEL FOR THE IDENTIFICATION SYSTEMS
    Author: SEDANO FRANCO JAVIER.
    Year: 2006.
    University: LEÓN [www.unileon.es].
    Place of defense: ESCUELA DE INGENIERÍAS INDUSTRIAL E INFORMÁTICA.
    Place of preparation: ESCUELA DE INGENIERÍAS INDUSTRIAL E INFORMÁTICA.
    Summary: This thesis attempts to synthesize the coarse and dispersed knowledge in the system identification, seeking a compromise between east and gained from the experience. The thesis looking for the formulation of a domain, so that their exploitation by a system based on knowledge possible. The basic objective is to create a methodology, a process, ultimately providing a conceptual model, which will help the identifier and to provide intangible as intuition, insight, experience and the record of stages, guidelines, rules and tips so that the proposed system is representative of the system identified. To that end, three sets stage: Stage One: Development of the procedure of identification, this will be developed from existing knowledge, state of the art and benchmarking. Tata Our process of uniting the knowledge of great masters and experts as Iserman, Sà ¶ derstrà ¶ m, Stoica, Young, Rivera Haber, Nelles, Heuberger, Nà ¸ rgaard, Ljung, ... And be flexible and customizable with the system to identify. Second phase: Contrastación by experience and educción the rules associated to that end we must implement the identification procedure above, in a practical procedure (software identification) consisting of algorithms and commands and having excitation signals, methods and evaluation models and indicators. The practical procedure will be tested before systems, the experiences will be reflected in the behavior of software systems before this identification. The matching and educción the rules will occur if it meets the definition of Professor Ljung, that a good model that will give a good prediction, ie, to provide small errors when you apply the observed data. Third phase: Creation of the conceptual model, this will be developed on the basis of knowledge and the experimental basis educiendo knowledge and experiments, both in its design and in its development and in its analysis a set of rules, rules to follow more or less stages and patterns, in order to complete the procedure and know how to reinvest it.
7 theses in 1 pages: 1
Búsqueda personalizada
kriptia.com
E-mail