Dissertations / Theses on the topic 'Redresseur de type buck'

Cette thèse se concentre sur la thématique de l'éjection électromagnétique appliquée dans le cadre de l'industrie du recyclage des métaux. L'objectif de celle-ci est la modélisation et la réalisation d'une architecture de conversion d'énergie permettant la réalisation de cette éjection. Le champ magnétique variable généré est utilisé afin de séparer d'un flux de déchets les matériaux métalliques non ferromagnétique tel que l'aluminium ou le cuivre par le biais d'une force de Laplace engendrée par la conjonction entre le champ magnétique crée et le champ magnétique induit par les courants de Foucault dans les matériaux conducteurs. L'architecture développée se compose de plusieurs éléments: un redresseur, un onduleur et un inducteur. La partie redresseur, à large plage de fonctionnement connectée sur le réseau d'alimentation triphasée permet d'obtenir une tension continu réglable et assure un prélèvement sinusoïdale de courant en phase avec la tension. La partie onduleur, permet de contrôler la puissance transmise, en adaptant l'amplitude et la fréquence du courant traversant la dernière partie du système que forme l'inducteur, responsable de la création du champ magnétique alternatif. Le choix du redresseur s'est porté sur la structure classique du redresseur abaisseur de tension de type Buck en raison de la faible impédance de l'inducteur utilisé. Bien que cette structure permette d'abaisser la tension triphasée, sa plage de fonctionnement peut être facilement augmentée sans l'ajout de composant passif. Le contrôle classique de ce redresseur ne se base que sur ses grandeurs de sorties ce qui peut engendrer des oscillations non contrôlées causées par la mise en résonance du filtre LC d'entrée excité par les harmoniques générés par les commutations des transistors. Nous avons proposé dans cette thèse une nouvelle méthode de contrôle qui traite à la fois de ses grandeurs d'entrée et de sortie et qui permet de contrôler les éventuelles oscillations du filtre LC d'entrée tout en bénéficiant d'une meilleure réponse dynamique lorsque le système est soumis à un échelon de charge. Cette méthode de contrôle se base sur les propriétés de platitude des systèmes différentielles, ainsi elle ne dépend pas du point de fonctionnement et garantit la stabilité large signal du système. Le choix de l'onduleur monophasé s'est porté sur une structure en pont complet permettant l'application de trois niveaux de tension et un large choix de contrôle de l'amplitude, de la forme et de la fréquence du courant traversant l'inducteur. Différents contrôle de ce convertisseur ont été étudiés et comparés. Ceux-ci permettent de faire varier la puissance injectée dans l'inducteur, ont un impact sur le contenu harmonique du courant le traversant et sur les contraintes des différents composants du système. Une modélisation de l'inducteur ainsi qu'une estimation de la valeur du champ magnétique nécessaire à l'éjection est effectuée. Les différentes méthodes proposées sont validées par des résultats de simulations numérique mais également par le biais de tests expérimentaux réalisés sur le système complet
This thesis focuses on the subject of electromagnetic ejection applied in the context of the metal recycling industry. The aim of this thesis is the modeling and the development of an architecture of energy conversion allowing the realization of this ejection. The generated variable magnetic field is used to separate non-ferromagnetic metallic materials such as aluminum or copper from a waste stream by means of a Laplace force generated by the conjunction between the magnetic field created and the magnetic field induced by the eddy currents in the conductive materials.The developed architecture is composed of several elements: a rectifier, an inverter and an inductor. The rectifier part with a wide operating range connected to the three-phase grid network allows to obtain an adjustable DC voltage and ensures a sinusoidal current in phase with the voltage. The inverter part allows to control the transferred power, by adjusting the amplitude and frequency of the current flowing through the last part of the system represented by the inductor, responsible for the generation of the variable magnetic field.The rectifier is based on the classical Buck rectifier structure because of the low impedance of the inductor used. Although this structure allows to lower the three-phase grid voltage, its operating range can be easily increased without the addition of passive components. The classical control of this rectifier is based only on its output variables which can lead to uncontrolled oscillations caused by the resonance of the lightly damped input LC filter excited by the harmonics generated by the switching of transistors. In this thesis, we proposed a new control method that deals with both its input and output variables and that allows both to control the oscillations of the input LC filter while obtaining a better dynamic response when the system is subjected to a load step. This control method is based on the flatness properties of differential systems, so it does not depend on the operating point and guarantees the large signal stability of the system.The single-phase inverter is based on a full bridge structure allowing the application of three voltage levels and a wide choice of control of the amplitude, shape and frequency of the current flowing through the inductor. Different controls of this converter have been studied and compared. These allow to vary the power injected in the inductor, have an impact on the harmonic content of the current flowing through it and on the constraints of the different components of the system.A modeling of the inductor as well as an estimation of the value of the magnetic field necessary for the ejection is carried out. The different methods proposed are verified by numerical simulations but also by experimental tests performed on the whole system

The present papers deals with three variable analogue of Boas and Buck [14] type generating functions forpolynomials of two variables and then the same has been extended for m-variable analogue. The results obtained are extensions of those obtained by us in our earlier paper [14].
El presente artículo trata el anólogo de tres variables de la función generatriz de Boas and Buck [14] para polinomios de dos variables y lo mismo se puede extender para el análogo de m variables. Los resultados obtenidos son extensiones de un artículo previo [14].

Le sujet de la thèse traite de la modélisation d’un système d’énergie avec piles à combustibles (PàC) de type PEMFC destiné à être intégré dans les véhicules électriques. Le but recherché par la réalisation de cet émulateur (objet matériel qui a le même comportement qu’un système réel) est de développer des lois de commande assurant un bon fonctionnement du système PàC. Après une étude bibliographique sur les différentes modélisations de cœur de Pile à Combustible (PàC), le choix d’un convertisseur de type Buck a été retenu et mis en œuvre pour l’émulateur. L’originalité de ce travail réside dans la prise en compte de toutes les auxiliaires du système PàC. Les auxiliaires ont été introduites sous forme Hardware In the Loop (HIL). Pour cela l’environnement DSPACE a été utilisé. Plusieurs lois de commande et de contrôle ont été étudiées et mises en œuvre. L’émulateur a été conçu avec une large bande passante. Le contrôle de la tension est réalisé avec la technique « State Variable Feedback », qui est capable de fixer correctement les pôles du système à boucle-fermé afin d’assurer la bande passante souhaité et le comportement à régime permanent. Le cas particulier de la gestion de l’air a été largement développé et a permis de valider le principe d’émulation retenu. En effet, différentes stratégies de commande (static feedforward, PI) ont été étudiées et comparées en utilisant une nouvelle stratégie de commande basée sur les réseaux neuronaux. Cette dernière repose sur le principe d’inversion de la relation entre la vitesse de rotation du compresseur et le coefficient de « oxygen excess ratio », dont la régulation est d’importance capitale pour éviter l’appauvrissement en oxygène de la membrane
This thesis deals with modelling of a PEM-Fuel Cell System (FCS) for power generation in an electrical vehicle. The goal of the research is the construction of an emulator of the PEM Fuel Cell stack, that of a device having the same behaviour as the real system, and the development of command strategies for the FCS. After a bibliographical study of the models of Fuel Cell stack, a buck converter structure has been chosen and then implemented to build the emulator. The novelty of this thesis is that all the auxiliary components of a FCS have been considered in a Hardware In the Loop (HIL) fashion by using a DSPACE development platform. Several command strategies have been implemented and assessed by using the emulator, which has been designed with a high bandwidth. The voltage control of the emulator has been accomplished by using the “State Variable Feedback”, which is a pole-placement technique for achieving the desired bandwidth and dynamical and steady-state performance. The particular case of control of the air-management system has been considered and used to asses the emulator. Actually several control strategies (static feedforward, PI) have been studied and their results compared also by using a novel neural network based command strategy. This neural network implements the inversion of the relationship between the compressor speed and the “oxygen excess ratio”, whose regulation is a key issue for preventing the oxygen starvation of the membrane

碩士
國立臺北科技大學
電力電子產業研發碩士專班
96
In this thesis, a non-isolated buck-type converter, together with half resonance and zero voltage transition to achieve zero current switching and zero voltage switching, is presented, in which the source terminal of the MOSFET switch is connected to the ground so as to make this switch easy to drive. In addition, the pulse width modulation (PWM) control technique is utilized in this converter so as to render the required output filter easy to design. By doing so, the problems in the electromagnetic interference (EMI) and power dissipation are reduced and hence the efficiency of this converter is upgraded. In this thesis, the theoretical derivation is firstly introduced, and secondly the simulated and experimental results are provided to demonstrate the effectiveness of the proposed topology.

碩士
國立臺北科技大學
電機工程系所
102
The thesis will present a converter with multiple sources for energy harvest of internal combustion engine vehicle. The energy sources include solar panel, regeneration energy during braking and thermal electrical generator. Since these three energy sources have different power rating, a power distribution control method is proposed in this thesis. The proposed power distribution control method can effectively implement power distribution while three input sources operate in the MICs at the same time. The second function is mode switching method. MICs is required to use Thermo-Electrical Generator for main energy source. As output power need is more than TEG standard, mode switching method can operate MICs back to breaking recharge system, and posses power distribution function. The specifications of the converter include: input voltage ranging extends from 16 V to 60 V, output power is around 1.32 kW and output voltage is 12 V. The three-phase control signals are with interleaved control to reduce the output voltage ripple. According to the simulation and the result of experiment, MICs can implement power distribution base on different load needs. When it is full load, output voltage ripple is smaller than 1%, and full load efficiency can achieve 90.08%. Besides, when MICs implement switching-mode technology, the system can effectively operate extra input source according to load needs, or part the input source from system.

碩士
國立中山大學
電機工程學系研究所
101
Battery power modules (BPMs) with bidirectional buck-boost converters, which are connected in series, are operated interactively but substantially can be controlled individually for either charging or discharging. During the charging process, the battery currents can be scheduled by adjusting the duty-ratios of the associated buck-boost converters. To fully utilize the charger’s capacity, several charging scenarios are proposed according to the state of charges (SOCs) of the batteries under the limitations of the charger’s power and current. To demonstrate the feasibility and capability of the charging scenarios, a battery power system formed by 4 serial buck-boost type BPMs is built and tested. A microcontroller is used for estimating the battery SOCs, and then scheduling the battery currents accordingly. Experimental results show the effectiveness of the charging scenarios.

碩士
國立臺北科技大學
電腦與通訊研究所
101
The first part of this thesis is a Split-Type III (ST3) compensation technique for CMOS DC-DC buck converter with 91.2% efficiency. The proposed compensation not only provides the fast transient response as the conventional Type III compensator, but also reduces area and power consumption of passive components. The buck converter combines current signals of bandpass filter and ramp generator for the input of pulse width modulation (PWM). The proposed buck converter has been fabricated with a TSMC 0.35-μm CMOS 2P4M process. This design is based on the 1MHz operating frequency with the inductor of 4.7 μH and the capacitor of 10 μF to reduce switching loss. Measurement results shows that the settling time of the converter is less than 5 μs for a load current step of 200 mA. Peak efficiency of 91.2% is obtained at 150 mA load current. The next part of this thesis introduces a Split-Type III (ST3) compensation technique and a charge-pump circuit method for CMOS DC-DC buck converter with fast transient response and high efficiency. The proposed buck converter has been fabricated with a TSMC 0.35-μm CMOS 2P4M process. The operating frequency of proposed converter is 1 MHz. The buck converter with ST3 compensator and the ramp generator using Schmitt-trigger circuit can reduce chip area, power consumption and circuit complexity. Measurement results shows that the settling time of the converter output is less than 2 μs for a load current step of 200 mA. The buck converter can offer a stable output voltage between 1.0V to 2.5V. Finally, the efficiency of the buck converter can stay above 80% in 50 to 400 mA load; peak efficiency of 90.8% is obtained at 100 mA load current.

碩士
明新科技大學
電機工程研究所
94
In general, the switching power supplies of electronic products have constantoutput, but in order to satisfy all-around power requirements, research and design of the converter with variable output is necessary. In this thesis, the author uses the familiar converter Buck converter as the research sample. The content includes analysis of the Buck converter; derivation of small signal model, and design of closed-loop controller. From the small signal model, the open-loop frequency response of converter system by using Bode plots can be obtained. Furthermore, the theory of control system was applied to design a closed-loop controller for increasing stability of the whole system. Finally, one high power Buck converter has been made. The whole circuit design procedures and the entity's data measure have been described. The variable output voltage function has been accomplished by using a linear adder. Simulation and experimental results are shown and compared to verify the feasibility of the converter.

碩士
國立成功大學
電機工程學系
102
This thesis presents a multi-phase buck-type pulse-current converter. For the purposes of high power and high efficiency for laser diodes, the large current and the high current slew rate are required to drive the laser diodes. Furthermore, the output current ripple must be reduced for the great fiber-optical transmission performance. However, the constant-current operation for the laser diodes causes the temperature rise, which further results in the decrease of the light output power. In order to maintain the same light output power, the operation current must be increased, and then the temperature rises again, which results in the thermal run away and the damage for the laser diodes. Therefore, the pulse-current operation for the laser diodes is employed in this thesis to improve the thermal run away. In order to generate the pulse current with the characteristics of the shorter rising time and the low current ripple, the synchronous control and the interleaving control are employed in the proposed circuit during the rising stage and the constant-current stage, respectively. Furthermore, the low-frequency switch, the center-tapped transformer, and the feed-forward diode are added to further improve the falling time during the falling stage. Finally, a prototype circuit of the 30A four-phase buck-type pulse-current converter is built to verify the performances, such as the rising time, the falling time, and the current ripple.

碩士
國立中山大學
電機工程學系研究所
101
The operation of a battery power bank with buck-boost type battery power modules (BPMs) connected in series and parallel is studied. All BPMs in the power bank are collaboratively to cope with the load requirements under the restraints of a same output current for serial configuration and a same output voltage for parallel configuration. However, the BPMs can substantially be operated individually to schedule the discharged currents from batteries can be in accordance with their state-of-charges (SOCs) and the operating mode of the converters. The operations of BPMs with series and series-parallel configurations are analyzed to derive the current- distribution equations and then to figure out the discharging strategy accordingly. Experimental results demonstrate that excellent performance on discharge equalization can be achieved during the discharging processes. In addition, a fault-tolerance mechanism can be included to isolate those completely exhausted or damaged batteries. These features are helpful to maintenance and management of a battery power system.

碩士
國立臺灣海洋大學
電機工程學系
107
Along with the development of technology, in order to improve the efficiency of power and energy, the design of power converter is particularly important. In recent years, power management integrated circuits have also been widely used to reduce the area cost. Besides, the processor continues to reduce the operating voltage and increase the operating current. When the operating current changes instantaneously, it is easy to cause a large surge of the operating voltage, which causes the buck converter applied to the processor to face more severe challenges. A common solution is to implement adaptive voltage position techniques through output impedance design. In addition, general electronic products are used for light loads for a long time. In order to improve light load efficiency, the constant on-time control can be used to enable the product to automatically reduce frequency at light loads to improve switching loss. In this study, a buck converter chip with an output impedance design is implemented. The control architecture uses the constant on-time (COT) control with the enhanced V2 loop. The chip is implemented using the TSMC 0.35um Mixed-Signal 2P4M Polycide 5V process with an input voltage range of 3.3V to 4.2V, an output voltage of 1.2V, a switching frequency of 1.2MHz, a load current range of 25mA to 600mA, and a conversion efficiency of up to 91.1%. The inductance value is 3.3uH and the capacitance value is 16uF. Keywords: Constant output impedance, adaptive voltage position technique, V2 loop, constant on-time control.

碩士
國立臺灣科技大學
電子工程系
107
In the applications of full bridge, phase shift is a most popular architecture. It has an excellent performance on efficiency thanks to soft-switching, especially in high power high voltage applications. However, whether the secondary side uses synchronous rectification or non-synchronous rectification, it will produce the voltage spike on secondary rectifier switches. The voltage spike increases the voltage stress of rectifier switches. This thesis proposes a new type of buck snubber circuit not only to clamp the voltage spike but also to replace the of clamping diode on the primary side because its temperature will be high in light load and increase the power loss. The buck snubber stores the energy of voltage spike into the capacitor and transfers the energy to output which makes the spike energy recovery. By the way, the voltage stress on rectifier switches was reduced, the rectifier switches can be changed as lower voltage stress with lower turn-on resistance. The efficiency of the system will also be increased. In addition, the thesis will compare the difference of clamping point between the rectifier switch and secondary transformer. Finally, the simulation waveform and measure waveform will be used to prove the feasibility of buck snubber and its improvement of efficiency.

Power consumption is the primary concern in battery-operated portable applications. Buck converters have gained popularity in powering portable devices due to their compact size, good current delivery capability and high efficiency. However, portable devices are operating under light load condition for the most of the time. Conventional buck converters suffer from low light-load efficiency which severely limits battery lifetime. In this project, a novel technique for buck converter is proposed to reduce the switching loss by reducing the effective input supply voltage at light load. This is achieved by switching between two different input voltages (3.3V and 1.65V) depending on the output current value. Experimental results show that this technique improves the efficiency at light loads by 18.07%. The buck voltage possesses an output voltage of 0.9V and provides a maximum output current of 400mA. The buck converter operates at a switching frequency of 1MHz. The prototype was fabricated using 0.18µm CMOS technology, and occupies a total active area of 0.6039mm^2.

碩士
國立臺北科技大學
電機工程系
107
With the progress of science and technology, the demand for rechargeable batteries is increasing, and lithium-ion battery is one of popular rechargeable batteries. Since the maximum voltage of a lithium-ion battery is 4.2 V, the lithium-ion battery cannot meet the requirement of high-power products such as electric motors, wheelchairs, and robots. A conventional method is to connect batteries in series to increase output voltage. But the series-connected batteries suffer the issue of voltage imbalance under charging and discharging situations. This thesis proposes a buck charger with active balancing architecture. A TI DSP TMS320F28035 is adopted to monitor the information of load voltage and current and then determine the charger operates in trickle current, constant-current, or constant-voltage mode. The proposed active balancing circuits are composed of transistors, comparators, current sources, and voltage divider resistors. The active balancing circuits charge or discharge the series-connected batteries to realize voltage balance of the series-connected batteries. The input voltage of the system is 19.5V, and the maximum output voltage is 16.8V. The experimental results show that the active balancing circuits can fulfill voltage balance of the series-connected batteries.

碩士
國立臺灣科技大學
電機工程系
91
The purpose of this thesis is to design an AC/DC converter which uses single-switch structure to achieve high efficiency, high power factor and low harmonics. The main switch operates in zero-current-switching mode and the output diode operates in zero-voltage-switching mode to reduce switching losses and increase power density. By operating with soft-switching, the single-switch structure has higher switch utilization than three-switch structure and six-switch structure and substantially reduces the complexity of control circuit. A simple control circuit has been proposed to make sure the main switch operating with soft-switching and to help the power circuit achieve high efficiency, high power factor and low harmonics. The thesis not only describes the system circuit principle but also verifies the theory by IsSpice simulation. A 2kW prototype of the AC/DC buck type converter has been implemented. The main switch and output diode of the converter can operate with soft-switching in different loads by changing switching frequency.

碩士
國立交通大學
電機工程學系
106
A non-inverting buck-boost DC-DC converter for USB-C charger is proposed for compact charger of the portable devices. In order to achieve the general solution for portable devices and the functional update, the pulse skipping modulation and the combined control mode used on the converter are implemented on the controller. The high voltage input, wide range output power and optional output mechanism are realized for the application. For achievement wide range output power and compact size, the proposed buck-boost converter utilizes the 0.18 μm BCD technology. The converter is consisted by power stage and control stage. The power stage is composed of four power switches of on-chip MOSFET, a low ESR off-chip inductor, and output capacitors. The controller circuit is implemented in low voltage supply to reduce the power consumption for higher efficiency.

碩士
國立交通大學
電機工程學系
105
Recently, the latest Universal Serial Bus type-C connecter not only can transfer data but also deliver power. As mentioned in USB Power Delivery Specification Release 2.0 Version 1.2, USB type-C can deliver power ranged from 0.5W (100mA at 5V) to 100W (5A at 20V). In order to support such input range, the converter must be able to endure up 20V of input voltage. Also, the converter is required to support the latest Intel® turbo boost technology. Turbo boost technology, discussed in an Intel® November 2008 Whit paper, can enable the CPU to run above its base operation frequency in order to draw out more performance when some cores aren’t operating at maximum speed. In the meantime, the processor requires more power to perform at higher speed. Therefore, the DC-DC converter must be able to deliver high power to CPU cores as well. Combining the two condition, the DC-DC converter we need should be able to receive a wide variety of input voltage (5V to 20V) as well as deliver a large power when the CPU need it. Therefore, a buck-boost converter that can step-up and step-down is most suitable. The thesis proposed a current mode buck-boost converter in order to simplify the compensation circuit and have good load regulation at buck/boost mode. However, the current sensing design as well as slope compensation is limited in high voltage condition. In this thesis, we will further discuss high voltage circuit design. The converter this thesis proposed can withstand input/output voltage ranged from 5V to 20V and can transfer 60W maximum power. The load current is up to 6A at buck mode. Only buck mode and boost mode are implemented in order to reduce loss while input and output are at the same voltage. A high voltage current sensing circuit as well as a dynamic slope compensation circuit is implemented. The chip is fabricated with Magna 0.18um process and operated at 1MHz switching frequency.