ABSTRACT. We propose a method to evaluate impedance-based stability of grid-connected inverters in real-time. An existing inverter is used to inject a broadband perturbation on top of the nominal grid current, and Fourier methods are applied to extract the frequency components in both the grid voltage and current. The ratio of inverter and grid impedance is computed online and plotted in real-time enabling online monitoring of stability margins. The method allows post-fault monitoring and self-diagnostics during turn-on or after a fault that caused inverter to disconnect. Online monitoring of stability margin is a necessary step toward more advanced adaptive control algorithms.

ABSTRACT. This paper proposes a new load impedance specification for multi-stage dc power conversion systems. The proposed specification avoids drawbacks of the existing specifications, such as the lack of explicit connections to the dc link dynamics and unnecessary conservativeness. The proposed specification offers a direct command/supervision of the frequency- and time-domain dynamics of the intermediate dc link, while being less conservative than the existing specifications. This paper also presents procedures of redesigning ill-conditioned load impedances to comply with the specification. The validity and utility of the proposed specification are demonstrated using an illustrative example.

ABSTRACT. Virtual oscillator control (VOC) and droop control are distinct methods to ensure synchronization and power sharing of parallel inverter systems. Since VOC is a time-domain controller that reacts to instantaneous measurements without additional filters or computations, it provides a rapid response during transients. In contrast, droop controllers typically use low-pass filters in the closed-loop control structure to suppress ac harmonics. The low cutoff frequencies of such filters impedes bandwidth. We present an analytical framework to compare the dynamic response of VOC and droop control. Our analysis is experimentally validated and we show that VOC exhibits superior speed.

ABSTRACT. Unlike the high short-circuit current withstanding capability of conventional transformer, an SST can fail in the event of any fault due to limited over current rating of active devices. A new control approach is proposed to improve the fault tolerance of SST by identifying a short circuit fault and limit the current before a healthy module is brought in to operation. In the proposed method, the control is designed to switch from the conventional control to a fault tolerant control in the event of fault. The proposed method is validated using PLECS real time hardware-in-loop platform and results are presented.

ABSTRACT. This paper presents the modeling and the stability analysis of a current controller for multi-paralleled grid-connected converters with LCL-filter. A discrete-time state-space model with N converters is developed. Initially, a discrete-time robust current controller is designed for only one converter connected to the point of common coupling. Then, the same controller is also applied to a second converter connected to the same point of common coupling and the stability of the whole systems is analyzed. Simulation results are shown, when two converters are considered, and hardware-in-the-loop results will be shown in the final version to validate the proposal.

ABSTRACT. A new implementation and performance comparison of the recently proposed two-sample (2S) Phase Locked Loop, PLL, applicable to single-phase sensorless Bridgeless Power Factor Correction (PFC) is proposed. Its characteristics are accuracy and low computational burden. The proposed PLL is based on a fixed-frequency 2S quadrature generation subsystem (QSG) and here it includes a frequency feedback loop to improve the synchronization under line frequency variations or steps. Its performance embedded in the digital controller of a current sensorless bridgeless PFC is evaluated through simulations and experimentally. The results are compared to previously published PLLs suitable for this application.

ABSTRACT. This paper presents a compact, lightweight, highly efficient, 6.6 kW isolated three-port converter for onboard charger (OBC) applications. The converter was designed and fabricated using normally-off GaN transistors; a 3-D printed coldplate; high-voltage printed circuit board (PCB) power planes; low-voltage and high-current PCB power planes; and a planar transformer. The prototype has a power density of 10.5 kW/L and specific power of 9.6 kW/kg. Test results show greater efficiency than a Si-based counterpart, even at 2.5 times higher switching frequency. The GaN converter was then integrated with a SiC traction inverter and successfully tested as a level-2 OBC.

ABSTRACT. This digest introduces a new approach to compensate for coupling variations in wireless power transfer (WPT) systems using an active variable reactance (AVR) rectifier. The AVR rectifier incorporates lossless resonant networks and actively-controlled bridge rectifiers, interfaced with the system output using dc-dc converters. The input reactance of the AVR rectifier can be continuously varied to compensate for coupler misalignments while maintaining constant output power at high efficiency. A prototype AVR rectifier is designed, built and tested with a 6.78-MHz, 65-W capacitive WPT system for laptop charging applications, and coupler misalignments up to 50% are fully compensated.

ABSTRACT. Active twice-line-frequency power pulsation buffering is an area of active research, owing to the need for highly efficient and compact single-phase grid-connected power converters. In this digest, an equivalent impedance model is developed for dc side series-resonant buffer architectures. As an example, the series-stacked buffer (SSB) architecture is analyzed using the proposed impedance model, and a loss compensation scheme is developed and generalized to both voltage and current control architectures. Experimental results of a SSB prototype verify operational effectiveness of the model for a PWM-based voltage-controlled operation.

ABSTRACT. The performance of power electronic converters is sensitive to main design variables such as topology, control, and modulation schemes as well as details such as manufacturing variability. In this paper, sensitivity index is introduced as an indicator of design quality to measure converter robustness in presence of manufacturing uncertainty. This paper presents the design and optimization of a Vienna rectifier that converts variable frequency 115 V AC voltage into 340 V DC voltage to illustrate the multi-objective optimization with parametric uncertainty quantification. The primary design target is to maximize converter efficiency and robustness within given size limitations and operation requirements.

ABSTRACT. Behavioral models of semiconductor switching devices have been widely implemented . However, the IGBT behavioral model is more complicated than that of the MOSFET because the BJT structure exhibits special carrier behaviors such as the tail current. In this paper, a simple behavioral model with easy parameter extraction procedure and wide temperature range is proposed to estimate hard switching loss at different operation conditions. A tail current model is implemented so that the estimation of turn-off loss becomes more accurate. The switching performance of the proposed model is validated at different voltages, currents, gate resistance and temperatures.

ABSTRACT. Diode rectifier models which accurately capture the i-v characteristics of diodes are currently not available in Simulink. This paper will present the development of system-type rectifier models, with both voltage and current AC inputs, which allow accurate diode i-v characteristics. The models are structured in such a way that the resulting nonlinear equations are functions of only a single variable and strictly monotonic, and are therefore easily solved. Example simulations using rectifier models with both AC voltage and current inputs are provided.

ABSTRACT. As advanced functionality is being demanded from distributed generation (DG) systems, smart inverters are being developed to provide reactive compensation to provide system support during disturbances. This compensation, however, leads to increased semiconductor device stress in the inverter. This work investigates the electro-thermal impacts of reactive compensation upon the longevity of SiC power MOSFET and Si IGBT products.

ABSTRACT. This paper investigates a method for quantifying the additional losses in high-voltage GaN enhancement mode HEMTs (eHEMT) employed in converter applications. The additional losses stem from the phenomenon known as current collapse or dynamic on-state resistance. The goal of this work is to investigate how these losses contribute to the total power loss in converter application. Measurement and modelling methods in the literature are reviewed. Changes to the measurement circuit are made to improve measurement accuracy. Measurements of dynamic on-state resistance are made on a commercial GaN eHEMT. The experimental results shows that the resistance depends on both dc-link voltage and blocking time. The resistance waveform and initial attempts at modelling the dynamic on-state resistance indicate that the suggested model is suitable for modelling the losses in LTSPICE software.

ABSTRACT. A physics-based compact device model for an enhancement-mode gallium nitride gate injection transistor (GIT) is presented in this work. The model is implemented in Saber® using MAST hardware descriptive language (HDL). The physics-based compact model allows the user to extract the model parameters from the dc-IV and C-V characteristics.The high voltage GaN transistor also shows bias-dependent drift resistance. The dynamic validation of the model is demonstrated against the characteristics of a 600 V commercial Panasonic GaN device using a double-pulse test (DPT) circuit. The simulated vs. measured device characteristics show satisfactory agreement and validate the model for power electronics applications.

ABSTRACT. In power electronics converters, the insulated gate bipolar transistors (IGBTs) are easy to fail during possible large disturbance on the source or load side. This paper redefines the safety operating area (SOA) by proposing a relationship between the junction temperature rising time and the disturbing current on the device. An improved temperature-dependent Cauer-type thermal model is used to uncover the junction temperature rising mechanism of IGBT under transient current steps. The SOA results can be used in the protection design for the IGBT modules.

ABSTRACT. In this work we propose a PI passivity-based (PI-PBC) decentralized controller that ensures global asymptotic stability of a multi-terminal (MT) HVDC system based on Modular Multilevel Converters (MMCs). An equivalent model of the MMC with fixed equilibrium point is expressed using a port-Hamiltonian representation, upon which a PI-PBC is next derived for the case of a single MMC as well as for the case of an MMC-based multi-terminal HVDC system with radial topology is finally presented. The results are validated on a three-terminal benchmark.

ABSTRACT. This paper studies sucient conditions for the existence of an equilibrium point of multi-terminal HVDC (MT-HVDC) network for oshore applications. The behavior of the wind power plant is non-linear since it must be modeled as a constant power load with drop control. The aim is to show asimple but generalized dynamic model for MT-HVDC and to use the Banach xed point theorem to set the requirements for the existence of the equilibrium. Finding this equilibrium is equivalent to the power flow in power systems applications. Computational results demonstrate the requirements for equilibrium of the system in a realistic MT-HVDC grid.

ABSTRACT. In multi-infeed HVDC systems, AC faults at the terminal will lead to concurrent commutation failures of multiple HVDC converters. Therefore, it is important to explore the interaction of voltages at different inverter side AC nodes and identify boundary conditions of concurrent commutation failure. The expression of multi-infeed interaction factor was derived using steady-state power flow equations at the receiving-end bus. Then, concurrent commutation failure interaction factor is derived using the minimum extinction angle of the inverters and the change in the DC current after the fault. The boundary conditions of concurrent commutation failures are investigated and parameterized by this factor.

ABSTRACT. The dc-dc conversion will play an important role in multi-terminal dc networks and dc grids. This paper presents two isolated resonant modular multilevel converters (IRMMCs) to fulfill the large step-ratio conversion for medium voltage dc (MVDC) networks. The conventional resonant modular multilevel converters (RMMCs) suffer the common problems of non-isolation and high current stress, which are solved in the proposed IRMMCs. They not only inherit the beneficial features of inherent sub-module (SM) voltage-balancing and soft-switching operation from RMMCs, but also develop multi-module configurations to neutralize the current ripples on both sides of the dc-links. The theoretical analysis is verified by a set of full-scaled simulations for different application examples in MVDC collection and distribution. The results demonstrate the proposed IRMMCs and its derived configurations have good potential for operation as large step-ratio MVDC transformers.

ABSTRACT. The modular structure and the many other advantages of the Modular Multilevel Converter makes it an attractive converter for Battery Energy Storage Systems, allowing the battery units to be distributed throughout the convert, connected in each submodule. However, such an arrangement results in large oscillating components in the battery current, which is harmful to battery performance and lifetime. In most literature this is solved using DC-DC converters as active interfaces between battery and submodule. This paper investigates a passive filter arrangement as an alternative solution to the DC-DC converters. Where the batteries are interfaced with the submodules using a passive filter that suppresses the fundamental component and control technique that injects circulating current to suppress second harmonic component. It is concluded that this passive technique could be an attractive solution especially where high reliability is of concern.

ABSTRACT. This paper classifies non-isolated DC/DC converters utilizing modular multilevel converters (MMC) into two broad classes based on the two possible AC power transfer mechanisms in the absence of an intermediate transformer. A unified DC MMC (UDCMMC) is proposed to demonstrate that existing MMC-based DC/DC converters of both classes can be realized through a general structure and a common control strategy. The UDCMMC also realizes a new topology that is potentially more cost-efficient than existing topologies within a range of DC/DC voltage conversion ratios. The operation of the UDCMMC with both mechanisms of AC power transfer is verified in PLECS simulation.

ABSTRACT. This paper presents a comprehensive harmonic analysis of the interleaved voltage source converters based on different carrier configurations. The impact of carrier configurations on the interleaved three-level (3L) converters is also investigated. The relationship among the harmonics in line-to-line voltage, common mode voltage (CMV), and circulating current is identified for interleaved two-level converters. By extending the analysis and generalizing the tri-carrier (TC) configuration to 3L converters, two new carrier configurations are obtained, which can be applied in single or interleaved 3L converters. The new methods provide extra tradeoffs for system designs.

ABSTRACT. This paper presents a control scheme that utilizes energy storage of the Modular Multilevel Converter (MMC) for Power Oscillation Damping (POD) service. Such a service is used to enhance grid stability by providing damping power in the electromechanical (0.2-2Hz) dynamics range. The scheme uses compensated modulation with average energy control. The aforementioned service comes at the cost of additional energy storage requirement for the MMC. A worst case estimate of this additional capacity, together with potential options to obtain it, is also addressed in this paper. The scheme is validated by simulation studies on a four terminal HVDC test grid.

ABSTRACT. Due to problems caused by variable switching frequency a modulated model based predictive control was proposed for different topologies and its applications. However, for cascade H-bridge multilevel converters, there is another problem, the imbalance in the DC-link voltages. This paper proposes extend the modulated model based predictive control to a cascade H-bridge 7-level STATCOM, further including a switcher of redundant states to solve the imbalance voltage issue. Simulations results shows improvements in terms of total harmonic distortion and balance voltages in the DC-links.

ABSTRACT. This paper describes a model predictive strategy to reduce the sub-module voltage oscillation in a single-phase modular multilevel converter. This task is accomplished by a predictive controller which solves an optimal control problem sequentially. By choosing the objective function weights appropriately, this controller can naturally trade off sub-modules ripple and recirculating current. It is shown that the sub-module voltage oscillation can be reduced without degrading the efficiency excessively, enhancing the performance of the overall device. Additionally, it is guaranteed that the recirculating current can be regulated without exceeding the physical limitations of the device.

ABSTRACT. The presence of Director Switches (DS) in its arms enables the Alternate Arm Converter (AAC) to generate smooth current waveforms while using a reduced number of Sub-Modules (SMs) when compared to the Modular Multilevel Converter. To maximise the power efficiency of the AAC and simplify the DS design, it is highly beneficial to soft-switch the DS. This paper presents a method which reliably ensures their soft-switching operation using the blocked state of SMs during the non-conducting part of the cycle of an arm. Experimental results demonstrate the effectiveness of the proposed method over a wide range of operating conditions.

ABSTRACT. This paper presents multi carrier Phase Disposition PWM technique with elliptical modulating wave for cascaded multilevel inverter (CMLI). The control strategy has been implemented using Xilinx System Generator to reduce the complexity of the implementation. To reduce the complexity of hardware implementation and output measurements, a novel FPGA Wavect controller, which is a real time digital controller, is used. In this paper, performance of single phase CMLI with RL-load is verified by simulation and hardware implementation using proposed modulation technique which results in improved performance of inverter in terms of output voltage, power and power factor for a given load.

ABSTRACT. When it comes to the computation / estimation of the switching losses in a multilevel converter, it is crucial to know the instantaneous value of the output current. If the average output current value is used instead, there is a non negligible error when the quality of the output current waveform is low. This paper addresses this issue with a generic and versatile fast numerical method for phase-disposition PWM that produces the exact switching pattern, hence the exact inductor flux, without having to run time-domain switched simulation(s), as the calculations are performed for the minimal set of points.

ABSTRACT. A well-discussed strategy to miniaturize magnetic components is to increase frequency. We consider the impact of frequency on the size of the transformer in a flyback converter. We have built a flyback transformer optimization routine that employs accurate and computationally efficient loss models to evaluate practical designs. An 18/11 pot core was picked and optimized over a range of frequencies. The best performing designs fell in the 500 kHz - 3 MHz range suggesting that the biggest opportunities for miniaturization lie in this frequency range.

ABSTRACT. The short lifetime electrolytic capacitor brought reliability issues to the inverter system. However, state-of-the-art design of inverter DC bank concentrates on voltage ripple and stability without reliability consideration. This digest proposes a design method of paralleling film capacitors with electrolytic capacitors to construct a hybrid DC bank to prolong the lifetime of inverter. It establishes the design parameter related thermal-electrical lifetime model based on the calculation of DC bank current. The influences of the ratio of film capacitor on DC bank lifetime have been investigated. More theoretical analysis and experimental results will be represented in the final paper.

ABSTRACT. This paper reviews performance factors for comparing soft magnetic materials, including components subject to high frequency ac winding resistance effects. The effect of permeability on component performance factors is investigated, showing that even low permeabilites can be useful for MHz level power converters. These results provide a high-level means for component designers to choose magnetic core materials and more realistic targets for magnetic materials development.

ABSTRACT. This research presents a Maximum Power Point Tracking (MPPT) algorithm for a solar photovoltaic (PV) power station which is constructed based on modified circuit topology of the Modular Multilevel Converter (MMC). The proposed MPPT is featured with Optimized Finite-Control-Set Model Predictive Control (FCS-MPC) to provide continuous extraction of maximum solar energy. The estimation-based approach allows for elimination of the current sensors in PV panels unlike the well-proven MPPT methods, e.g., perturb and observe (P&O) and ripple-correlation control (RCC), where costly sensing devices are widely deployed. With developed control strategy, fast tracking response for the PV plant along with acceptable control bandwidth under solar irradiation transients and PV panels mismatch are achieved. The introduced circuit of power conversion has led to decreased number of solar cells and a 50% reduction of the energy storage sizing compared to the basic MMC configuration. The effectiveness of the discussed MPPT solution over the solar irradiance variations is compared with the both P&O and RCC methods. The behavior of the studied PV system is also evaluated under different operational scenarios through simulations to confirm the designed concept.

ABSTRACT. Solar Panel Companion Inverter (SPCI) is an H – bridge inverter placed on the back of each PV module to convert the DC voltage output of that module to a quasi-square wave voltage with variable pulse width. When these voltages are aggregated across multiple panels connected in a series string, they synthesize an AC waveform that can be interfaced with the power grid. In this paper, a new closed loop Maximum Power Point Tracking algorithm has been proposed, which can implement system level maximum power point tracking.

ABSTRACT. Power system is seeing higher penetration of Utility PV at distributed level as well as transmis-sion level. As negative incremental resistance caused by constant power behavior of power con-verters may bring stability problem, stability of PV integration to distributed system attracts more and more attention. Impedance of utility PV farm is derived and verified by MATLAB. Applica-tion of Generalized Nyquist Criterion (GNC) based on PV and grid impedances is able to predict the stability of PV integration. The results show that instability problem may occur when multiple PV inverters are under reactive power control mode of Q = f (V).

ABSTRACT. Due to shadows caused by elevation angle variation of the sun according to season, solar panels should be installed with certain distance between each other in succession, which causes wide unused space. To solve this problem, plane-type reflectors are installed between solar panels. In this paper, the curved reflector type is newly proposed, which can increase the efficiency at high elevation angle of the sun. Calculation results showed that the average efficiency ratio with respect to elevation angle of the sun for the solar panels with no reflector, plane-type reflector, and the proposed curved reflector is 1:1.5:1.6, respectively.

ABSTRACT. High voltage conversion ratio dc-dc converters are needed in a wide range of applications. Hybrid switched-capacitor-magnetics power converters can leverage the advantages of capacitor-based circuits and inductor-based circuits to achieve high efficiency and high power density. This paper investigates the design method for an emerging family of tapped-series-capacitor (TSC) power converters that can benefit high voltage conversion ratio applications. By resonating switched-capacitor circuits with coupled-magnetics and applying frequency modulation, ZVS can be achieved across a wide input voltage range and load range. Methodologies for designing high-frequency printed-circuit- board (PCB) magnetics with sophisticated winding patterns are also presented and experimentally verified.

ABSTRACT. This paper presents the design and implementation of a Very-High-Frequency (VHF) multi-resonant gate drive circuit. The design procedure is greatly simplified compared with some VHF self-oscillating multi-resonant gate drivers presented in previous works. The proposed circuit has the potential to reduce long start-up time required in a self-oscillating resonant gate drive circuit and to better utilize fast transient capability of VHF converters. A prototype resonant gate driver is demonstrated and able to reduce 60% of the gate driving loss of a 20 MHz 32 W Class-E power amplifier with Si MOSFET.

ABSTRACT. Stability for dc current distribution systems with long transmission cables are not well studied. In this paper, impedance expressions for long cables are derived and applied to analyze stability of dc current distribution systems. The relationship between converter closed-loop and open-loop input impedance is discussed. Impedance-based stability analysis are proposed by employing the Nyquist plot of the system minor loop gain. Analysis presented are validated through simulations in the digest, while experimental verification will be provided in the full paper based on a system with 1A current source, 100km cable emulator and 1kW series resonant converter that has been built.

ABSTRACT. This digest introduces a single-stage isolated 48V-to-1.8V point-of-load converter based on the impedance control network (ICN) converter architecture. This converter achieves large step-down using a novel immittance network transformer and a resonant current-doubler rectifier, and achieves high efficiency by maintaining soft-switching across wide operating ranges. The converter's power density is enhanced by combining its magnetic components into two integrated magnetic structures. A prototype 90-W ICN converter designed to operate over an input voltage range of 36 V to 60 V and an output voltage of 1.8 V is built and tested, and experimental results demonstrating its performance are provided.

ABSTRACT. Matching networks have useful applications in transforming voltages and impedances in resonant inverters and dc-dc converters. Stacking multiple stages of matching networks can, in some cases, increase the efficiency because each stage is responsible for smaller transformation, but it also reduces the available inductor volume for each stage which can increase the loss. We present in this paper optimization of matching networks with volume constraints to determine the optimum number of stages and other design choices for various transformation ratios, volumes and impedances. Scaling models of inductor performance with size are presented and their effect on the efficiency of single-stage and multistage matching networks is analyzed. The analytical results are verified by an experiment using 1- and 2-stage matching networks with a total volume constraint and a voltage transformation ratio of 4. Simple design rules for designing matching networks are presented for voltage transformation ratios lower than 20.

ABSTRACT. This paper discusses the derivation of an accurate zero-voltage switching (ZVS) boundary expression for bidirectional dual-bridge series resonant converter (DBSRC). The closed form solution under hard- and soft-switching boundary condition is derived by employing state plane analysis, which shows the soft-switching range depends on two factors: voltage conversion ratio and resonant/switching frequency ratio. Compared with the fundamental harmonic approximation (FHA) method, the boundary obtained in this digest shows higher accuracy. Results are verifed on a 800W experimental DBSRC prototype.

ABSTRACT. High-frequency, low-power DC-DC converters operated in the hundreds of kHz to multi-MHz range present a number of benefits such as fast transient response and reduced passives. Drawbacks associated with switching and capacitive losses due to the increased switching rate can usually be mitigated by the use of resonant topologies with soft-switching operation. This paper presents a quasi-resonant buck converter with a low-complexity digital controller which minimizes the active switch's voltage across the active switch at turn-on and extend input voltage operating range on a switching cycle basis. Simulations and preliminary experimental results are reported for a 500kHz, 5V-to-3.3V, 500mA prototype.

ABSTRACT. We present a dc-dc resonant converter structure that has a pair of loosely coupled inductors in place of one filter inductor and two resonant inductors. Not only the number of inductors is reduced by one, but in this arrangment  the transformer provides galvanic isolation and requires a sub-unity coupling coefficient which eases the implementation, especially at high switching frequencies. Various topologies, including class-E, DE, and Phi2, benefit from this structure. Since the new converter structure is equivalent to the conventional structure, existing design methodologies are directly applicable. We demonstrate three proof-of-concept converters consisting of a class-E inverter and a class-E rectifier with the efficiency of 74%, 90%, and 92% and the input-to-output voltage ratio ranging from 5 to 12.

ABSTRACT. This paper introduces a new two-phase inter­leaved flying-capacitor LLC converter topology with high output current applications. Compared to a conventional two-phase LLC converter, the new converter adds a single capacitor that contributes to lower voltage stress on the pri­mary side’s switches, automatically balances the current dis­tribution between the phases and enhances the power pro­cessing capabilities. All the attractive features of LLC con­verters are preserved, such as zero-voltage switching on the primary side’s MOSFETs, zero-current switching on the secondary side’s power devices, narrow switching frequency range and simple design. Full principle of operation and analysis of the converter is described, as well as the con­verter’s primary characteristics and the impact of non-ideal components on the current sharing behavior.

ABSTRACT. This paper proposes data transmission method in flyback converter battery charger without additional communication circuit. A battery management system (BMS) and a charger constantly transmit and receive data for the safe operation of battery and implementation of fast charging. The most conventional system uses an additional line or wireless communication modules for this data transmission. The proposed system, in comparison, does not use additional signal transceiver but instead exchanges data by alternating operation mode of the flyback converter, thus reducing the volume of the terminal and overall system. The waveform of transformer voltage is used to count the number of resonant pulses, which is used for decoding and encoding the message. Bidirectional communication between input and output is possible using an appropriate communication protocol. The experimental results are presented to verify the performance of the proposed communication method.

ABSTRACT. A topology using two dual-bridge series resonant converters and a three-phase unfolder has been introduced for three-phase power conversion. This topology synthesizes high quality waveforms, but at the expense of high number of switches. This paper proposes a three-port resonant dc-dc converter topology with reduced number of switches. Two switches and associated gate-drive circuits are removed by using a common leg of a converter. A control method is presented to reduce rms switch currents in the common leg. Comparison with conventional topology shows 18% reduction in primary-side conduction loss and 40% reduction in rms input capacitor current in proposed topology.

ABSTRACT. Presenting a control scheme for standalone-microgrid having variable-speed wind-generator(WG) and diesel-generator(DSG) directly interfaced at PCC, and battery energy storage(BES) interfaced at PCC via bidirectional DC-AC converter. BES not only regulates PCC voltage-frequency during wind/load power variations, but also ensures proper synchronization/disconnection of WG/DSG based on availability of energy sources, for seamless transitions between various operating modes, while providing uninterrupted supply to critical load. Using only one interfacing converter greatly simplifies microgrid structure. Performance under various operating modes and transitions is tested through modeling and implementation of control algorithm in Simulink to demonstrate applicability of scheme.

ABSTRACT. Aggregation of behind-the-meter distributed energy resources holds promise to provide valuable grid services, enabling greater penetration of intermittent renewable energy sources. We propose the Smart Dim Fuse system as an approach to utilize loads for aggregation services. The Smart Dim Fuse, installed on individual circuits at the breaker panel, enables load power control without service interruption via direct AC to AC voltage modulation, while reporting load measurement, identification, and resource availability forecasting to the aggregator. Using data from 131 homes, we evaluate the aggregate resource availability, and demonstrate a proof-of-concept prototype of the power electronics needed for this functionality.

ABSTRACT. DC power systems increase the flexibility and efficiency of electrical power networks by eliminating unnecessary conversion stages. However, loads interconnected to the network through a power electronics converter exhibit nonlinear behavior which can lead to stability problems. In this paper, Sum of Squares (SOS) programming methods are utilized to compute stability metrics for dc networks. While the nonlinearity appearing in the dc constant power loads is non-polynomial, SOS techniques provide flexibility in dealing with them. It is shown that the SOS method is simple and efficient to compute an estimate of the region of attraction of dc microgrids.

ABSTRACT. Droop–based control methods are used for operation of different inverters in converter-interfaced AC microgrids. To ensure accurate and fast response to disturbances in a stand-alone AC microgrid with constant impedance and constant power loads, this paper proposes a droop-based controller for power sharing among multiple inverters. The performance of the proposed controller is verified against two common state-of-the-art controllers, and the advantages of the new controller in terms of speed, accuracy, and stability are demonstrated. Results from experimental prototypes and analysis of their performance will be presented in the full paper.

ABSTRACT. Many electronic loads nowadays are equipped with power factor correction (PFC) and meanwhile modern smart meters have the capability of measuring higher-order harmonics in addition to power. This paper investigates using the smart electronic load with PFC for compensation of harmonic within a distributed power system. Such smart loads may communicate with smart meter for control of power quality in the system. This paper analyzes the operation of PFC converter and proposes a new approach to broaden the range of harmonic compensation. Simulation and experimental results will be presented based on a commercial PFC prototype to demonstrate the proposed approach.

ABSTRACT. This work proposes an output voltage regulation scheme for a fuel-cell based power source. The proposed methodology relies on the design of an average current mode controller considering an exponential proportional-integral (PI) term. Also, an external loop for voltage regulation using a PI action is added. Finally, based on a 460 W testbench and a DSPACE platform, the closed-loop system performance is evaluated through experimental results. An appropriate output voltage regulation and robust behavior concerning to load variation and fuel-cell stack unregulated DC output voltage are exhibited.

ABSTRACT. Wireless power transfer is emerging as the pre-eminent powering technology for implantable medical devices. Efficiency, simplicity, and reliability are the key goals for receivers in vivo. We use piecewise resonant wireless power transfer (PR-WPT) to achieve these goals. A high peak-to-average power ratio (PAPR) waveform is generated by a current-mode class D amplifier operating at 6.78 MHz. A 4^th-order passive filter is matched to the fundamental and third harmonic voltages of the transmitter, using harmonic elimination for the waveform and closed-form impedance analysis. A full-bridge Schottky rectifier converts the matched voltage into dc. Experiment demonstrates the proof of principle and simulation results show that the piecewise resonant methods can increase the dc output voltage by up to 30%, hence improving the rectifier efficiency. Potential applications for PR-WPT systems are discussed.

ABSTRACT. In this paper we analyze and demonstrate switched receivers for direct-sequence spread spectrum wireless power transfer (DSSS-WPT). With the limitations on single frequency operation, there is a strong impetus towards spread spectrum techniques. DSSS-WPT offers stronger security, simultaneous power transfer among multiple actors, lower electromagnetic interference, and higher rate communications embedding. These receivers designs offer a proof-of-principle for wideband power transfer that is spread-spectrum yet resonant.

ABSTRACT. A GaN-based synchronous rectifier is examined for 6.78 MHz WPT applications. Compared to a traditional diode rectifier, the proposed circuit has advantages in terms of efficiency, THD, and impedance adjustment through phase control. This work develops a modeling and design approach for the proposed circuit. Device selection and resonant tank design are linked to maximum system power, efficiency, range of input phase control, and THD through an analytical model. Experimental results of a GaN-based 20W rectifier prototype are used to verify the modeling approach.

ABSTRACT. This paper presents the design and implementation of an impedance compression network (ICN) to correct distance variation or misalignment between coils in wireless power transfer (WPT) systems. Magnetic resonant coupling coils provide high efficiency for charging mid-range WPT applications. However, a distance variation or misalignment between coils causes a coil-impedance change and significantly affects performances of WPT systems. In order to mitigate a coil-impedance variation, we propose an ICN compressing changes in coil-impedance. While a resistance compression network effectively compresses a magnitude variation, a phase compression network is designed using a Smith chart to reduce a phase variation.

ABSTRACT. Operation of two-coil inductive power transfer (IPT) is affected by bifurcation phenomenon. So far, its dependence on primary and secondary quality factors was examined. However, similar dependencies exist for coupling coefficient and resonance frequency. In this paper, these dependencies and mutual influences are examined. Analysis results are verified by simulations in ANSYS designed to correlate with actual situations during operation. 5 kW model comprise a FEM model of transfer coils and electronics based on power GaN transistors. Bifurcation is examined under change of load, distance between coils, coil displacement and resonance frequency.

ABSTRACT. The demand for increased range and efficiency of resonant wireless power transfer has spurred research into the improved performance of resonant coils; however, these improvements are difficult to measure without a metric for comparison. The conventional figure of merit (FoM) does not include important system performance parameters such as coil size and range of power transfer. In this digest, we propose a new FoM that is derived in terms of coil-to-coil efficiency, distance between the coils, and diameter of the coils. Using this FoM for comparison we are better able to assess resonant coil technologies.

ABSTRACT. A "hollow" litz wire structure is proposed, where strands are confined to the outer perimeter of a dielectric core. Power losses of this structure are compared to two traditional litz wire designs and a solid wire. Hollow litz can attain lower power loss than traditional litz designs when the strands are constrained to a thickness greater than two skin depths.

ABSTRACT. Traditional winding loss estimation models lead to unacceptable estimation errors when applied to low-permeability inductors. This is due to the highly two-dimensional character of the electromagnetic problem. In this paper, a new analytical model is proposed which makes it possible to carry out a straightforward and accurate calculation of the winding loss in high-frequency low-permeability inductors.

ABSTRACT. In high-frequency and very-high-frequency power converters, which operate with switching frequencies above 10 MHz, efficient power inductors cannot be constructed with commercially available core materials. In this regime, air-core inductors are preferred, and the toroidal geometry best constrains the magnetic field. These inductors have been implemented as printed circuit board (PCB) traces. Here, we present the first unified electrical model for high-power, air-core toroidal inductors implemented as PCB traces, which includes closed-form calculations for the lumped capacitance, resistance, and inductance of the structure. In this regime, the self-resonant frequency (SRF) of the inductor may be of the same order of magnitude as the circuit operating frequency, so the capacitance must be carefully modeled and tightly controlled. The lumped capacitance is calculated by assuming that the impedance between two turns is well approximated by the characteristic impedance of a microstrip geoemetry. We demonstrate prediction of the SRF within a 10% error for fabricated inductors with inductances ranging from 50 nH to 10 uH and PCB heights of 1 mm, 1.6 mm, 2 mm, 3.2 mm, and 6.4 mm. Modeled capacitance is typically approximated within 1 pF of the measured value.

ABSTRACT. High-Frequency (HF) litz wires are extensively used for the windings of Medium-Frequency (MF) magnetic components in order to reduce the impact of eddy current losses that originate from skin and proximity effects. Literature documents different methods to calculate eddy current losses in HF litz wires, however, most of the computation methods rely on perfect twisting of the strands, which is often not present in practice. This paper analyzes the implications of imperfect twisting on the current distribution among the different strands of HF litz wires and the corresponding losses by means of a fast 2.5D PEEC (Partial Element Equivalent Circuit) method. The effects of different types of twisting imperfections (at the bundle-, sub-bundle-, or strand-level) are examined. It is found that imperfect twisting can lead to increased losses (more than 100%). However, perfect twisting of the strands, which is difficult to achieve, is often not required, i.e. suboptimal twisting is sufficient. Analytical expressions are given for distinguishing between critical and uncritical imperfections. The experimental results, conducted with a 7.5kHz\65kW transformer, reveal a reduction of the error on the predicted losses from 52% (ideal HF litz wire model) to 8% (presented model) and, thus, confirm the accuracy improvement achieved with the proposed approach.

ABSTRACT. Compound magnetic materials based on parallel continuous ferromagnetic micro-wires can potentially offer significant mechanical advantages over well-known fragile ferrite in some applications where the magnetic material has to withstand mechanical vibrations. Since the diameter of the magnetic micro-wires is very small (10’s of μm) compared to the rest of the dimensions (100’s mm to 10’s cm) brute force simulation by Finite Element Method is not possible. In this work, a method to overcome this limitation is proposed based on the use of a homogenous material that will behave, in a macroscopic sense, as the compound magnetic material.

ABSTRACT. Substrate termination of lateral 600 V-class GaN-on-Si HEMTs and half-bridges other than source-connection is investigated: Using conductive Si-substrates as current-return path and the GaN-buffer for high-side transistor-to-heatsink isolation eliminates parasitic inductances and thermal interfaces. Floating-substrate reduced 440 V switching times to 60% (reduced output capacitance) and fixed-to-ground substrate thermal resistance to 54%. However, substrate-to-gate capacitive-coupling significantly increases gate-charge which is analyzed using four-terminal measurements and simulations. Measured high-side current-collapse with fixed-to-ground high-side substrate termination (switching over 300 V, 2 MHz) is more pronounced in first-quadrant compared to third-quadrant conduction, and explained by dynamic source-access-region voltage feedback to the source-referenced gate-driver.

ABSTRACT. The RCD (Resistor-Capacitor-Diode) snubber is usually designed without considering which wiring inductance seriously affects circuit behaviors, although the influence appears significantly at high-frequency operation. We investigate an optimal design of the RCD snubber for MHz-switching of SiC-MOSFETs considering location of parasitic wiring inductance. The mechanism of ringing induced by wiring inductance, and ringing suppression by the RCD snubber are also discussed. The wiring inductance near the gate and source terminals should be minimized. The wiring inductance near the gate and drain terminals must be considered to design an optimal RCD snubber.

ABSTRACT. This paper presents an experimental investigation of the switching energy loss and peak overvoltage occurring in a half-bridge Insulated Gate Bipolar Transistor (IGBT) converter which is switched using an Active Gate Drive. A range of voltage profiles is systematically applied to the IGBT gate and the resulting switching behaviour is measured to obtain the switching energy loss and peak overvoltage. The experimental apparatus which allows different gate waveforms to be tested is described, and the minimum achievable switching loss for a set overvoltage limit is found for turn-on using gate voltage waveforms of increasing complexity. A reduction in turn-on switching energy loss of 24% is achieved with the most sophisticated gate voltage waveform tested compared to a simple voltage ramp waveform. Turn-off overvoltage is controlled when using more complex gate waveforms whereas simpler voltage ramping fails to influence the turn-off voltage overshoot.

ABSTRACT. Using virtual prototyping (VP) design tool to evaluate power converter electro-thermal performance can help designers to validate prototype in a quick way. However, different system time-scale requires efficient electro-thermal simulation techniques. Thus, an approach, by using average power loss of a switching cycle, is presented in the paper to decouple electrical and thermal simulation and it is validated by comparing with a traditional method to obtain device junction temperature. By implementing it in a previously developed VP design tool, alongside a SiC-MOSFET behavioural model, designers can obtain power converter electro-thermal waveforms more quickly than using traditional simulation tools.

ABSTRACT. GaN technology promises exceptional performance for converters in high switching frequency; nonetheless, reduction in device performance when used in real circuits have been observed. This paper evaluates power losses in GaN transistors when used in class Phi 2 resonant soft-switching inverters at MHz frequencies. Our experiments show that tested GaN transistors can display excessive loss of up to three times their predicted values under certain test conditions. The following paper presents the loss separation method, and the breakdown of these losses for tested devices operated across the frequency range from 6.78 MHz to 27.12 MHz.