Table of contents monday, September 9, 1: 30pm-4: 00pm modular Multi-Level Converters, hvdc, and dc grids I 3
Yüklə 2,78 Mb.
|
- Bu səhifədəki naviqasiya:
- Poster Session: Control, Modelling and Optimization of Power Converters
- Poster Session: Electrical Machines
| P4517 Cuk-Based Universal Converters in Discontinuous Conduction Mode of Operation [#1471]
Mahshid Amirabadi, Northeastern University, United States This paper introduces an open-loop control scheme for Cuk-based universal converters that have recently been proposed by the author. Cuk-based universal converters extend the principles of the operation a dc-dc Cuk converter to systems with single- phase or multi-phase ac sources and/or loads. These power converters can provide galvanic isolation by adding a single-phase high frequency transformer to the link. This eliminates the need for low frequency transformers that have large volume and weight. Being single-stage, the Cuk-based universal converters do not require decoupling components, and a small capacitor is used for transferring the power from input towards output. This eliminates the need for large electrolytic capacitors that have high failure rates. In prior works, a closed-loop control system was used for the Cuk-based universal converters. These converters were controlled such that they always operate at the boundary of continuous and discontinuous conduction modes, which leads to a variable switching frequency and a large voltage ripple across the link capacitor. The closed-loop operation along with variable switching frequency result in a complex control algorithm, which can limit the maximum switching frequency of the converter. This paper proposes an open-loop control scheme that allows the converter operate in discontinuous conduction mode and with a fixed switching frequency. Implementation of the proposed control scheme is simpler than the closed-loop control method. This paper presents details of the proposed method and evaluates its effectiveness. P4518 Neutral Points Voltage Balancing Control of a Four-level pi-type Converter [#1084]
In this paper, a carrier based modulation method with optimal zero sequence signal injection has been introduced to modulate a four level pi type converter as well as regulate its DC link neutral points voltages. The two neutral points voltages can be well controlled with a back to back configuration even under high modulation index and high power factor. A back to back experimental system has been built and tests under 300V have validated this control strategy. P4519 A Novel Three-Phase Multilevel Diode-Clamped Inverter Topology with Reduced Device Count [#1369]
This paper presents a novel configuration of three-phase multilevel diode-clamped inverter (DCI) with reduced part count. In the proposed topology, three-phase multilevel output voltage is generated by employing only two inverter phase legs and the third phase is sharing the unused switches of these two legs depending on the line current flow direction. The structure allows synthesizing multilevel voltage waveform with reduced number of power electronic components: diodes, power semi-conductor switches and gate driver circuits compared to classical topologies. Total savings on the power device count increases proportionally for higher output voltage level DCIs. Reducing the counts on the power electronic devices result in decreased complexity and overall system cost for the inverter. The compatibility of developed general space vector modulation (SVM) technique with new multilevel DCI arrangement; make the control scheme simpler and easily implementable. The simulation results confirm the feasibility of the proposed configuration in comparison with the conventional multilevel DCI topology. P4520 Maximum Boost Space Vector Modulated Three-Phase Three-Level Neutral-Point-Clamped Quasi-Z-Source Inverter [#1040]
Z-source inverter is a power electronic converter with potential applications in areas like renewable energy, electric vehicles, etc. because unlike traditional inverters, it can provide a boost in output voltage without any additional DC-DC boost converter stage in between the DC source and inverter. This feature is particularly useful if the input DC source is of low voltage like battery or of fluctuating nature as in case of PV panels. The elimination of need for DC-DC converter stage in an inverter system increases the overall reliability of the system and also increases the overall efficiency by the elimination of DC-DC converter stage losses. In this paper, a maximum boost space vector modulation strategy which gives maximum gain for three-phase three-level quasi-Z-source inverter (QZSI) is proposed. The proposed algorithm has been verified by both simulation and experimental results. Poster Session: Control, Modelling and Optimization of Power Converters Tuesday, September 20, 3:00PM-4:30PM, Room: Exhibit Hall, Chair: Pericle Zanchetta, Luca Solero P4701 High Dynamic and Static Performance FCS-MPC Strategy for Static Power Converters [#1156] Rodrigo Mendez, Daniel Sbarbaro and Jose Espinoza, Concepcion University, Chile This paper proposes an improved Finite Control Set Model Predictive Control (FCS-MPC) strategy to control static power converters that overcomes the parameter sensitivity of the conventional strategy. The parameters error can be due to both a poor estimation and their time variant behavior, as the grid parameters in grid connected static power converters. Contrary to the standard FCS-MPC approach that tracks a system output reference, the proposed scheme is based on a system input reference tracking that can be obtained from a state-feedback. In the proposed scheme, output system integrators are added to modify the dynamic performance and ensure zero steady-state error under parameters errors. This approach overcomes one important disadvantage that presents the conventional FCS-MPC strategy which is the dependence of a model with accurate parameters to avoid performance degradation. The results verify the correct performance in both with or without parameters errors P4702 New Logic-Form-Equation Based Active Voltage Control for Four-Level Flying Capacitor Multicell (FCM) Converter [#1225]
This paper presents a new active capacitor voltage balancing method for flying capacitor multicell (FCM) converter which is implemented using logic-form equations. The proposed active capacitor voltage balancing technique, measures output current and FC voltages to generate switching states to produce the required output voltage level as well as balance FC voltages at their reference values. Output voltage of the FCM converter with the proposed active voltage balancing method can be modulated with any kind of the pulse-width-modulation method such as phase-shifted-carrier PWM (PSC-PWM) or level-shifted-carrier PWM (LSC-PWM). Advantage of the proposed active voltage balancing method is its simplicity without any complex computation burden. Simulation results of a three-cell four-level FCM converter are presented to verify performance of the proposed active capacitor voltage balancing method. P4703 Experimental Evaluations of Thinned-Out and PDM Controlled Class-E Rectifier [#1307]
This paper presents the design of the closed-loop thinned-out controlled class-E rectifier. It is possible to realize both thinned-out control and the PDM one at the same control-circuit topology. In addition, fine output-voltage controls are possible because the control circuit can make a thinned-out signal at any thinned-out ratio. For evaluating the proposed class-E rectifier, the resonant dc/dc converter with the class-DE inverter and the proposed class-E rectifier is designed and implemented. The control characteristics of the resonant converter were measured for load and input- voltage variations. It is seen from the experimental results that the thinned-out control provides higher efficiency than the PDM control. In addition, the power spectrum of the switch voltages suggested that the cutoff frequency of the low-pass filter for thinned-out control can be higher than those for the PDM control, which contributes circuit-volume reduction. P4704 Variable Slope External Ramp to Improve the Transient Performance in Constant On-Time Current Mode Control [#1632]
These days, constant on-time current mode (COTCM) control schemes are widely used in the industry VR controllers for its light load efficiency and higher BW design with simpler compensation requirement. In COTCM control, external ramp is required to be added for jittering reduction to improve the noise performance of the controller. But unfortunately, with the increment of external ramp value, although the jittering in the system gets better, transient performance becomes worse. For this reasons, from transient point of view, it is very challenging to use large external ramp in the system to improve the noise performance. In this paper, a novel method is proposed to modify the external ramp at the transient instant to improve the transient response by increasing the slope of the ramp, and thus allows the control to use the large external ramp for noise performance improvement and enjoys the fast transient response at the same time. P4705 PWM Methods for High Frequency Voltage Link Inverter Commutation [#1641]
This paper presents two new pulse-width modulation (PWM) methods for a high frequency ac-link (HFlink) inverter commutation which uses a form of four- step switching scheme for safe commutation. A HF-link inverter topology enables galvanic isolation and voltage step-up, with a reduced-size transformer, but eliminates the secondary-side rectification so it has less power conversion stages than other inverter topologies. However, converting the HF-link voltage to the desirable inverter output voltage adds complexity to the control of the circuits because it is a form of ac-ac converter with both the inverter bridge input and output voltages timevarying. The voltage- reference four-step switching technique was previously presented as a safe commutation technique for the ac-ac converter with bidirectional switches. The method provided safe commutation without detailed information of the exact output current zero-cross and without needing lossy snubber circuits. However, it assumed that the input voltage polarity remained constant during the current commutation interval. This paper proposes methods to overcome the limitation of the previous method to ensure so call safecommutation even when the HF-link voltage is changing polarity and that can be implemented by modification of conventional triangular carrier PWM methods. Therefore, the four-step switching scheme can be safely applied to the HF-link inverter. Simulations of the proposed new PWM methods are conducted using Simulink and Plexim software to demonstrate the concepts. The proposed PWM methods were experimentally verified using the dSpace 1007 HIL platform. P4706 Switching Pattern of a Modular Voltage Balancing Circuit for Battery Cells [#1681]
Switching patterns are described for balancing the voltages of series connected Li-ion polymer batteries using non-dissipative cascaded modular power electronics. A balancing circuit is connected to two battery cells and consists of an asymmetrical half-bridge and a high frequency transformer. This reduces the number of switches and diodes to an average of one per battery. A balancing circuit or module can balance the voltage of two batteries by using two magnetically coupled windings connected to the centre tap of the two batteries: intra-bridge voltage balancing. Each module is coupled to adjacent modules by connecting transformer windings. Two fundamental control methods are described for transferring electrical charge between the batteries in adjacent modules: inter-bridge voltage balancing. Both control methods, "2-switch flyback control" and "phase-shifted control", assume that each module determines its own switching pattern by monitoring the dc voltages in each of its adjacent modules. The relative merits of both controllers are described. Simulated and experimental results are presented for both controllers to validate the charge transfer methods used. P4707 Steady State Impedance Estimation of a Weak Grid to Assist Optimal Current Injection for Minimal Power Losses [#413]
Nowadays, latest advancements in power electronics have devised new ways to optimise power systems by reducing power losses. These developments provide convenience for private power producers to harness renewable energy and allow injection of optimized reactive currents to compensate for the reactive power drawn from the grid. This paper presents the development of an efficient laboratory based system that can monitor a transmission line, sense changes in the load and redistribute the optimal currents in a manner that ensures minimized power (I2R) losses at the point of common coupling (PCC). A method is discussed and implemented to calculate and feed the required optimal currents into the grid. This method depends on the measurement of dynamic parameters of the electrical grid which are calculated by applying a steady state impedance estimation technique. Once parameters are obtained, the aforementioned method is applied which allows the flow of optimal currents in the system. National Instruments (NI) PXI controller with built-in FPGA module is used for experimental investigation which demonstrates capability of the discussed technique to produce the required results. P4708 A Single-phase Unified Power Quality Conditioner with An Enhanced Repetitive Controller [#916]
This paper proposes a robust and flexible control strategy for a single-phase unified power quality conditioner (S-UPQC) to solve the power quality problem due to the distorted input supply voltage and the nonlinear load. In order to enhance the performance of the series and shunt active power filters (APFs) in S-UPQC, an advanced current and voltage control schemes are introduced based on the modified repetitive controller (RCs) connected with a proportional-integral (PI) controller. Unlike the conventional control scheme for the series and shunt APFs, the proposed scheme is implemented without any harmonic extractor. Furthermore, the dynamic performance is improved significantly due to the reduced delay time of the RC, and the computation time is also reduced. The effectiveness of the proposed control schemes for the S-UPQC is validated through simulation. P4709 Single-Phase Universal Active Power Filter Based on AC/AC Converters [#1176]
This paper presents four single-phase universal active power filters (UAPF) topologies with 4, 5, 6 and 7 legs. The proposed systems are based on AC/AC converters with two transformers. They are suitable to improve the harmonic distortion in the grid currents and a voltage compensation. The filter uses a combination of two series filters and one shunt filter sharing a single DC-link. The proposed active filters are capable of generating input and output voltages with low harmonic content and consists of low-power switches. The complete control system, including the PWM techniques based on vector approaches, is developed. Simulation and experimental results are shown for validation purposes. P4710 Circulating Resonant Current Between Integrated Half-Bridge Modules with Capacitor for Inverter Circuit Using SiC-MOSFET [#874]
A design for the inverter circuit configuration of integrated half-bridge modules with a focus on the circulating resonant current is clarified in this paper. Although this configuration has the advantages of small DC-side stray inductance, an analysis of the equivalent circuit suggests that a DC-side capacitor current increases depending on the relationship between the resonant and switching frequencies. This fact will lead to an increase in the capacitor volume. The capacitor current was experimentally investigated at switching frequencies up to 100\,kHz using a SiC-MOSFET. The experimental results demonstrated that the resonant current was remarkable when the resonant frequency was close to either the fundamental or the third harmonics of the switching frequency. By the proposed design, the increase in the capacitor current can be avoided. P4711 Computationally Efficient Event-Based Simulation of Switched Power Systems and AC Machinery [#979]
Switched power systems are inherently difficult to simulate, due to the significant sensitivity of the simulation algorithm to the timing of the switching events as well as the potentially complex dynamics of the components. Numerical simulation techniques, while flexible, are not well suited for large time scale simulation of such systems. Methods for increasing the accuracy of numerical simulation techniques increase the computational complexity of the solution, resulting in an obstacle for adoption into time-sensitive applications such as model-based design optimization and hardware-in-the-loop simulation. This paper presents a method and development framework for a computationally efficient method for analyzing the behavior of switched power systems via piece-wise analytical solutions to differential equations. Enhancements of the method to expand the utility are also presented. P4712 Design Optimisation and Trade-offs in Multi-kW DC-DC Converters [#1047]
This work investigates design optimisation and design trade-offs for multi-kW DC-DC Interleaved Boost Converters (IBC). A general optimisation procedure for weight minimisation is presented, and the trade-offs between the key design variables (e.g. switching frequency, topology) and performance metrics (e.g. power density, efficiency) are explored. It is shown that the optimal selection of components, switching frequency, and topology are heavily dependent on operating specifications such as voltage ratio, output voltage, and output power. With the device and component technologies considered, the single-phase boost converter is shown to be superior to the interleaved topologies in terms of power density for lower power, lower voltage specifications, whilst for higher-power specifications, interleaved designs are preferable. Comparison between an optimised design and an existing prototype for a 220 V - 600 V, 40 kW specification, further illustrates the potential weight reduction that is afforded through design optimisation, with the optimised design predicting a reduction in component weight of around 33%. P4713 Switching frequency optimization for a Solid State Transformer with Energy Storage Capabilities [#1079]
This paper is focused on establishing a procedure for measuring the efficiency dependency on the switching frequency for a solid state transformer, being one of the ports connected to an energy storage device (Lithium-Ion battery). Multiple contributions for measuring the efficiency/losses for different power converter structures for energy storage applications can be found in the literature. However, there are few references which consider the effects of the high frequency model of the battery in the complete system performance. This research will obtain a parametric high frequency model of the battery cells, based on a vector fitting method in frequency domain. This model will be used for the estimation of the overall system losses. It will be demonstrated that the contribution of the battery losses, as well as its behavior as a function of the switching frequency, can significantly affect the selection of the converter's switching frequency. P4714 Lag-Free Terminal Voltage Sensing in Low-Pass Filtered PWM Converters [#1085]
Several high-performance control algorithms for ac machine drives and other applications require accurate knowledge of the instantaneous converter output voltages. The use of a simple low-pass filter to remove the PWM switching frequency harmonics from the AC voltage measurements introduces undesirable amplitude and phase lag errors into the voltage measurements. This work presents compensation functions that make it possible to compensate these measured filter output voltages for both amplitude and phase lag errors introduced by zero-order-hold samplers and two different types of PWM algorithms. The impact of the selected filter time constant and the required accuracy of the compensating equations relative to the ADC accuracy are also addressed, leading to identification of the optimal filter break frequency for maximum accuracy. Poster Session: Electrical Machines Tuesday, September 20, 3:00PM-4:30PM, Room: Exhibit Hall, Chair: Bruno Lequesne, Mohammad Islam P4901 Cogging Torque Minimization with Rotor Tooth Shaping in Axial Flux-Switching Permanent Magnet Machine [#1346] Ju Hyung Kim, Yingjie Li, Emrah Cetin and Bulent Sarlioglu, University of Wisconsin-Madison, United States This paper proposes a rotor tooth shaping method to reduce the cogging torque of an axial flux-switching permanent magnet machine. The rotor tooth is shaped in different configurations, including constant, trapezoidal, asymmetrical, extended, and square shapes. 3-D finite element analysis is performed to compare the proposed shapes of rotor tooth. Peak-to-peak cogging torque, fundamental flux linkage, loaded torque, and the total weight of teeth are compared using various rotor tooth angles. The simulation results show that overall minimum cogging torque is achieved by the extended shape of rotor tooth. P4902 A 3D Printed Fluid Filled Variable Elastance Electrostatic Machine Optimized with Conformal Mapping [#1525]
Recently, fluid filled electrostatic machines have demonstrated specific and volumetric torque density that hold promise to be competitive with electromagnetic machines in niche applications. These demonstrations of variable elastance (dual of reluctance) machines were non-optimized from an electrostatics perspective as their geometry was heavily constrained due to manufacturability. Higher performance electromechanical power conversion for electrostatics requires optimal geometric design. This paper proposes a semi-analytical method incorporating conformal mapping techniques with finite element (FE) analysis to optimize a variable elastance electrostatic machine in a low speed direct drive applications. As a proof of concept, an optimized geometry was built using additive manufacturing, specifically stereolithographic 3D printing, to circumvent geometry constraints. By building the machine from plastic plated with conductor, it is lightweight with improved torque density. This manufacturing approach suggests that a machine can be injected molded or cast in a single step. Experimental results support both the design and manufacturing approaches and the resulting machine is benchmarked against previous work.
Yüklə 2,78 Mb. Dostları ilə paylaş:
|