ABSTRACT. Single-stage Power Factor Correction (PFC) ac-dc rectifiers potentially open an pathway to achieve high power density and efficiency in grid-connected rectifier applications where the target dc voltage is lower than the peak ac voltage (e.g. data center power delivery). Typically in data center and similar applications, a two-stage solution employing a step-up ac-dc stage followed by a step-down dc-dc stage is employed to achieve grid to 48 V conversion. This approach suffers from the efficiency penalty of a cascade of power converters and potentially lower power density due to the design of two separate power conversion stages. A single-stage, buck-type PFC rectifier where the output dc voltage is lower than the peak ac voltage circumvents these issues. This work analyzes and develops a single-stage buck-type PFC rectifier utilizing a six-level Flying Capacitor Multilevel (FCML) converter with an active voltage balancing and current controller to achieve high power density rectification in a single-stage solution.

ABSTRACT. This digest presents an improved adaptive feedback control strategy for the inversion stage of the recently proposed high-power-density transformer-less single-phase online uninterruptible power supply (UPS). The proposed inversion stage operates in buck/buck-boost mode depending on the sinusoidal output voltage polarity. This dual-mode operation of the inversion stage results in poor output voltage regulation and hence high output voltage total harmonic distortion (THD) with standard feedback control. Previously proposed adaptive feedback control strategy provides some improvement in output voltage THD compared to standard feedback control but the reduction in output voltage THD is limited and hence unacceptable for practical applications. To achieve better output voltage regulation and hence lower output voltage THD, an improved adaptive feedback control strategy is proposed. Analytical models capturing the dynamics of the inversion stage with adaptive feedback and improved adaptive feedback control strategies are also provided to aid with the affiliated controller design and performance comparison. A GaN-based 1-kW prototype high-power-density online UPS is designed, built, and tested to validate the effectiveness of the adaptive feedback and the proposed improved adaptive feedback control strategies. The improved adaptive feedback control strategy results in a 50% improvement in the output voltage THD compared to the previously proposed adaptive feedback control strategy.

ABSTRACT. This work develops a general analytical framework to understand and quantify state estimation and flying capacitor voltage balance dynamics in hybrid switched-capacitor converter topologies. We generalize a state-space modelling framework previously only applied to flying-capacitor multilevel (FCML) converters to treat other topology classes including Dickson-based, Series Parallel (SP), and others. The model uses the linear algebraic concept of condition number to better quantify controllability and observability versus conversion ratio. This helps to illustrate the relative difficulty in achieving balance and state estimation across converter classes and is verified using Series-Parallel and FCML hardware prototypes.

ABSTRACT. Hybrid switched capacitor converter (HSCC) topologies have found use in high-density power delivery applications, such as 48\,V to point-of-load. Of this family, Dickson-type converters are recognized as having the lowest volt-amp switch stress, indicative of a smaller semiconductor footprint for equivalent performance. However, some of these topologies require a non-conventional clocking scheme --- termed ``split-phase'' switching --- to ensure that capacitor-induced hard-charging losses are avoided. A small number of recent works have analyzed split-phase switching with varying degrees of rigor, however, to date no closed-loop active control scheme has been demonstrated. This work presents the first hardware implementation of closed-loop split-phase control, eliminating both hard-charging losses and reliance on modelled converter operating points, while providing an increased degree of immunity to component mismatch. Capitalizing on periodic low-noise switching states in which flying capacitors are inactive, a low-cost and high accuracy analog front-end is used to converge on optimal split-phase timing durations while maintaining output regulation, irrespective of load.

ABSTRACT. The digest presents a composite converter-based LED driver fed by an automotive battery and capable of delivering constant current to an LED string. A two-stage architecture consisting of a composite boost converter in the first stage connected in an IPOS configuration, cascaded with a buck converter in the second stage which acts like a constant current source for the LED string is proposed. The composite boost converter is capable of processing indirect power efficiently, thereby reducing the device voltage and current stresses, and resulting in a higher overall efficiency compared to a conventional boost+buck topology. The architecture is capable of operating over wide input (8V–18V) and output voltages (3V–50V) driving a string of 1 to 16 LEDs. Analytical loss models are formulated which results in peak efficiency of 95.8 %. Preliminary experiments are shown validating the operation of a composite boost converter from 12V to 36V at 1 A.

ABSTRACT. This paper proposes an inter-stage feedback-based switched-capacitor (IFSC) converter that can be configured to realize any rational voltage-conversion ratio (VCR) using a minimum number of 2:1 switched-capacitor (SC) stages. The converter enables efficient dynamic voltage scaling (DVS) operation over a larger voltage range by significantly increasing the resolution of voltage levels. Additionally, a restructuring algorithm has been proposed to achieve all VCRs without requiring any additional dedicated voltage stages. The converter extends the idea of the recursive switched-capacitor (RSC) topology, which spans 2^N -1 ratios with N 2:1 SC cells. Comparative analysis shows that the proposed converter outperforms the negative-output feedback-based converter in terms of bottom-plate parasitic loss for most VCRs. To validate the concept, a fully-reconfigurable hardware prototype has been developed.

ABSTRACT. In this paper, the averaged switch modeling method is utilized to derive the small-signal model of multi-inductor hybrid (MIH) converters. While modeling complementary switches with DC transformers is enough for conventional converters, the models for the hybrid converters need additional modifications to accurately reflect the actual converter behavior. To demonstrate the modeling and analysis technique, the generated models of four example MIH converters successfully capture the higher-order transfer function characteristics by their higher number of inductors and capacitors. The result provides corroborated analytical proof that MIH converters behave similarly to the multi-phase Buck converters, while the flying capacitors contribute to pole-zero doublets that create minimal impact on the converter operation.

ABSTRACT. This paper presents discrete frequency and phase-shift control for a bidirectional class-E² converter, enabling a wide range of output power in energy storage applications. Class-E² converters typically operate within a narrow power range to achieve zero voltage switching (ZVS) across the switching devices and ensure high efficiency. To address this limitation, we propose an adaptive frequency control algorithm that utilizes multiple frequency steps to reduce output ripple and minimize output capacitor requirements. Our adaptive frequency control algorithm ensures high efficiency across a wide output power range by using 16 discrete switching frequency steps ranging from 800 kHz to 1.6 MHz. Duty ratios are pre-computed and stored in a lookup table to provide ZVS at each frequency. We also apply a variable phase shift between the switching devices to maintain ZVS and control the power flow direction. By directly selecting and applying the appropriate frequency from the lookup table instead of sequential searching, our proposed algorithm enables faster dynamic response with minimal undershoot and overshoot. Through simulation and implementation of closed-loop control of the bidirectional class-E² converter prototype using an MCU, we achieved bidirectional output power level variation from 13 W to 350 W with a maximum efficiency of 93.5%.

ABSTRACT. This paper introduces a scalable power hardware-in-the-loop (PHiL) battery emulation system. The battery emulator enables the simulation of real battery voltage profiles with full power rating sink and sourcing capabilities using off-the-shelf components. The battery emulator tracks battery voltage, temperature, and current to provide real-time monitoring of the emulated battery's state of charge (SOC) and remaining useful life (RUL). The emulator operates in a continuous battery emulation mode or a cyclic mode for repetitive battery testing. The new battery emulator can replace end-product batteries during system development and is realized in two parts, PC Graphical User Interface (GUI) and battery emulator (Hardware). A battery profile-generating algorithm is introduced to accurately reflect the behavior of an actual battery during emulation. All measured data and battery voltage profiles are transferred via Wi-Fi to enable maximal freedom in system deployment. An experimental prototype has been built and tested to verify the battery emulation operation. The prototype handles a maximum input voltage of 150V and an input current of 60A.

ABSTRACT. The discrete state event driven (DSED) framework has been proposed to simulate linear power electronics circuits with fast speed. To extend the capability of DSED to solving nonlinear systems, this paper proposes an enhanced DSED framework with the decomposition of nonlinear circuits’ equations and suitable events processor. With the proposed framework, we successfully simulate a railway wireless power transfer (WPT) system in both stable and coil switching process. Compared with Simulink, the enhanced DSED framework takes less execution time with the same accuracy.

ABSTRACT. MARS is an integrated development of photovoltaic (PV), and energy storage systems (ESS), connecting to alternating current (ac) transmission grids and high-voltage direct current (HVdc) links. This three-phase plant is a host to several thousands of components with a complex hardware and hierarchical control architecture. In this paper, the design process for the hierarchical control architecture for MARS is presented. Furthermore, the controller is validated and evaluated using control hardware-in-the-loop (cHIL) for numerous operating conditions and thus showcasing the stability of MARS.

ABSTRACT. Core loss is often reported using power-law fits on a limited amount of data collected on a single core shape. The development of automated testers which output a full `loss map' is one approach towards improving this reporting. Conventional measurement techniques are automatable but only to 500 kHz-1MHz. In this paper, we develop an automated core loss tester suitable for the > 1MHz regime and discuss its implementation and trade-offs. The tester produces data that is within 12.6%-57% of manufacturer-reported data, following similar trends for flux-density versus power loss density but tending to overestimate core loss. Sources of error in the automation procedure are discussed as well as strategies for future improvement of the system. 

ABSTRACT. The paper introduces a novel technique for linearizing the switching function of diode-based rectifiers. This approach establishes a linear correlation between the switching perturbation and the current perturbation using the harmonic balance method, and then builds a small-signal model using the harmonic state space (HSS) method. By considering the internal dynamics of the switching process, the proposed model provides a more precise representation of the rectifier's low-frequency AC impedance. The effectiveness of the proposed method is demonstrated through frequency scanning in the electromagnetic transient (EMT) time-domain simulation, which confirms the enhanced accuracy of the small-signal model. In summary, the paper presents a new approach for improving the accuracy of small-signal models of diode-based rectifiers by linearizing the switching function, which has been validated through simulation.

ABSTRACT. We propose a dual grid forming controller for power converters, and in particular for modular multilevel converters (MMC), that establishes voltage in the DC grid and voltage and frequency on the AC grid. Nowadays, the parameter tuning controller of MMC heavily relies on rich engineering experience, which usually leads to a suboptimal power system operation. Instead, we propose a $H_\infty$ controller, which considers AC and DC grid dynamics. We impose grid-forming conditions among other control objectives by using weighting functions. Finally, we provide simulations that allow us to validate that the proposed double grid forming algorithm is suitable for hybrid grid applications.

ABSTRACT. Wave Energy Converters (WECs) are a form of renewable energy that harvests energy from the waves in a large body of water. The mechanical power take-off (PTO) of these systems has been optimized to absorb the maximum amount of energy available in the wave environment. Still, the electrical PTO of WECs has not yet been optimized to maximize power conversion. The final version of this paper will use Typhoon HIL real-time software to model the mechanical and electrical systems of a WEC operating in an irregular wave environment. The switching signals of the inverter will be studied to maximize the efficiency of the WEC.

ABSTRACT. A high altitude electromagnetic pulse (HEMP) or other similar geomagnetic disturbance (GMD) has the potential to severely impact the operation of large-scale electric power grids by introducing low-frequency common-mode (CM) currents which deteriorate the performance of key grid components, such as large power transformers. In this work a solid-state transformer (SST) that can replace susceptible equipment and improve grid resiliency by safely redirecting, absorbing, and/or converting these CM insults is described. An overview of the proposed SST power electronics and controls architecture is provided, a system model is developed, and the performance of the SST in response to a simulated CM insult is evaluated.

ABSTRACT. The conventional trench termination for the 4H-SiC sidewall-implanted Super Junction device has relatively low avalanche capability. In this paper, a novel high-k dielectric assisted trench termination is proposed. The high-k dielectric TiO2 is incorporated into the SiO2-filled trench. TCAD simulations are conducted to validate the efficacy of the proposed termination. The results show that with the proposed termination, the leakage current flows uniformly from the active region, thus improving the avalanche current. Moreover, the dimensions of the TiO2 are optimized in terms of the maximum current density. By taking the dielectric strength into consideration, the optimum depth and length of the TiO2 region are found to be 0.6μm and 2μm, respectively.

ABSTRACT. Voltage-power relationship in the diode rectifier unit(DRU)- high-voltage direct current(HVDC) connected offshore wind farm (OWF) is analyzed, with controlled hydrogen electrolysis and small grid connection onshore. Chain rule is implemented to bridge the voltage at point of common connection (PCC) and the OWF, reflecting the effect of AC transmission, DRU filters and reactive power compensator, so as to obtain both steady-state and small-signal relationships between the power transferred to HVDC link and the voltage generated by the OWF, which are fundamental to the further design of grid-forming wind turbines (WT) operating in such scenario. The rigorous analytical results are verified by PSCAD/EMT simulation.

ABSTRACT. Voltage source converter (VSC) based DC systems are an effective way to connect offshore wind farms to the power grid. To better solve the excess power generation of wind farms during a power grid fault, a novel modular DC chopper, which may contain hundreds of submodules, has been proposed to absorb excess power in the VSC-DC systems. The real-time simulation test setup is a viable solution to facilitate validating the DC chopper and its controllers. In this paper, a high-fidelity real-time model of the modular DC chopper is developed. The DC chopper model and its low-level control are implemented on FPGA to provide sub-microsecond time resolution. Experimental results from the real-time tests are provided. More detailed discussions of the development and real-time simulation results are provided in the final paper.

ABSTRACT. In this paper a method to detect a man in the middle attack (MiTM) on a grid following PV inverter is discussed. The control objective of the grid following inverter is to utilize the measurement data from the smart meter to supply maximum available solar power at any given point to a residential load, while simultaneously preventing any reverse power flow to the grid. It is envisioned that false data is injected (FDI) in the smart meter data (P and Q) communicated to the inverter by malicious actors. In such cases, the FDI can result in inverter malfunction / reverse power flow resulting in system disconnect. The proposed method superimposes a small randomly varying voltage with a unique signature termed as ”watermark” into the input inverter DC link voltage and then monitors the smart meter data to detect any possible acts of FDI. A unique feature of the proposed approach is to inject a watermark signal in the the dc-link (i.e. external to the PV inverter), therefore can be applied to any commercial inverter installation. It is shown via extensive simulation and test results that the proposed method is capable of detecting un-observable FDIs. Communication protocols such as Modbus that could serve as attack points are reviewed. The proposed system is implemented on 3kW laboratory prototype grid following inverter system. Fewer results are included in the paper summary. Final paper will have a complete discussion on grid connection standards and test results.

ABSTRACT. This paper presents a study of a fault-tolerant magnetic coupling topology for a network parallel multilevel inverter in aircraft applications. The modeling, the sizing, and the design of the Secondary Loop topology are developed for 15 kW/phase. Then, the experimental results were compared to the simulations in an open-loop study. For the closed-loop study, a linear dynamic model of the inverter is discussed to ensure the balancing of the leg currents. Finally, this topology is tested in terms of fault tolerance due to the loss of a leg, showing promising results.

ABSTRACT. Accurate flux estimation is crucial for control of the Permanent Magnet Synchronous Machines (PMSMs). Parameter uncertainty reduces this accuracy and degrades the controller's performance. This paper presents a novel vector control of PMSMs based on robust techniques. Two different flux estimation techniques are combined to ensure the best performance at all speeds. It is demonstrated that the proposed regulator is robust to parameter uncertainty compared to traditional current and flux regulators.

ABSTRACT. With the proliferation of power electronic (PE)-based resources in the electric grid, there have been numerous different kinds of simulation models developed and studied to understand their dynamic behavior. Therefore, it is vital to carefully select the appropriate model for the evaluation of the PE-based resources. In this paper, one such effort is made to analyze the different types of models in the case of a multi-port autonomous reconfigurable solar power plant (MARS). The simulation models are evaluated for different test cases and control algorithms.

ABSTRACT. Loss estimation in electric machines is useful for performance evaluation, condition monitoring, and optimal control. The parameters determining losses can vary with operating conditions; therefore, accurate identification of these parameters is challenging. Online loss estimation is one solution where parameters are identified and updated as they vary with operating conditions over time. For applications requiring good accuracy, robust loss estimators which consider model uncertainty are needed. This paper presents a comparative robustness analysis for online loss estimators of AC permanent magnet (PM) machines based on electrical and thermal models. It is found that loss estimators based on thermal models are more robust than those based on electrical models over a wide range of operating conditions.

ABSTRACT. This paper presents a robust modelling technique for power electronic circuits that accurately captures device losses. The temperature-dependent i-v characteristics of the power electronic devices are used to develop a scalar nonlinear equation for a power-pole configuration that has a unique solution. A switching loss model is proposed that power at each switching transition. Experimental results are presented confirming the validity of the above approach.

ABSTRACT. In this paper, interpolation methods to enable fast simulation of photovoltaic (PV) inverters are proposed. The interpolation methods are based on the event-driven interpolating method that can be introduced in the simulation model to run the model at a higher time step with higher accuracy. The proposed method is tested on a neutral point clamp (NPC) converter typically used in large-scale PV plant applications.

ABSTRACT. This paper introduces a dynamic cosimulation approach to evaluate the effect of the selection of magnetic core material in toroidal inductors for DC-DC converters under varying load conditions. This cosimulation approach is based on the combination of transient analysis and finite element analysis to investigate how different high-frequency magnetic materials perform as potential core components for the converter's inductor. The study considers a DC-DC buck converter modeled in Simulink and a detailed toroidal core inductor modeled through COMSOL Multiphysics. The LiveLink for Simulink tool available in COMSOL Multiphysics is utilized for accurate inclusion of the nonlinear inductor model and its integration into the dynamic buck converter model. The study provides insights into the behavior of different magnetic materials under high current exposure, and their suitability for use in DC-DC converters. The results of this investigation can provide practical guidance for designing and optimizing DC-DC converters in various electrical systems, with a focus on selecting appropriate magnetic materials for toroidal inductors.

ABSTRACT. The modern power system has been challenged by the conspicuous transition from centralized power facilities to distributed generations (DGs), especially in the perspective of the power grid stability. The hazards of the power system instabilities and vulnerabilities have been increasing in recent years due to the following reasons: (1) The intermittency and unpredictability of renewable energy power generations creates more power flow fluctuations and frequency/voltage deviations. (2) DG inverters are usually PLL-oriented and controlled in peak power mode, which extracts the maximum power from the power source and does not leave any margin to handle variations on the grid side. With the vigorous increase of inverter-based DGs’ penetration, the energy share of traditional synchronous generators significantly shrinks, the total power system inertia can be reduced as well as the corresponding stabilizing effects. (3) On the demand side, many different types of loads, such as motors, electric vehicles, and DC microgrids, are interfaced with the grid through front-end power electronic rectifiers, which makes the system level behaviors more complex. Most of existing grid-interfaced rectifiers adopt direct power control or voltage-oriented control, which gives the converter direct axis input impedance a ‘negative incremental resistor’ property and raise the chances of small signal instabilities. To mitigate instability risks, it’s necessary to design grid-interfaced converters with grid-support features. on the generation side, grid-forming inverters (GFIs) are proposed to mimic the power balance mechanism of synchronous generators, therefore the power injection from DGs can be adjusted for grid side regulation purposes. At the same time, on the demand side, there have been few control strategies proposed to allow rectifiers participate in grid regulation. This paper adopts a grid-supporting rectifier (GSR) featuring grid-side frequency and voltage regulation capabilities. It can provide a more benign terminal impedance characteristics compared to traditional PLL-based rectifiers, thus can assist the power system operation for stability enhancement. The abovementioned features of the GSR makes it suitable for DC microgrids. Therefore, opens an opportunity to design DC microgrid based manufacturing plants as opposed to conventional AC system-based design. Thus, in this paper we propose a DC microgrid based plant topology for the flexible cold-rolling steel plant (F-SMP) to demonstrate application of the GSR as well as to highlight some of the advantages of the DC microgrid based plant topology.

ABSTRACT. A huge variety of different concepts for solid-state transformers has been proposed in recent years. DC/DC converters based on a multi-cell structure, implementing the required insulation stage represent a decisive building block of a solid-state transformer. While the majority of proposed topologies is based on magnetic coupling, capacitive coupled cell structures are an attractive and rarely discussed alternative. In this paper, a capacitive coupled 400 kW multi-cell DC/DC converter concept connected to a 13.5 kV DC-link is analyzed in detail. Requirements and implementation of the coupling capacitors are discussed, relevant design constraints are given and proper design rules are derived. The concept is finally comparatively evaluated with an existing magnetically coupled structure, where it is shown that capacitor coupled structures are a serious alternative to existing magnetically coupled approaches.

ABSTRACT. It is urgent to ensure small signal stable operation of wind farm under variable wind speed. However, the sub-synchronous resonant (SSR) stable wind speed range varies with the ratio of Type-III and Type-IV wind turbines in the hybrid wind farm grid-connected system (HWF-GCS) with series compensation lines. Firstly, the wind speed-frequency two-variable equivalent admittance model is established. Then, the influence of the ratio of Type-III and Type-IV wind turbines on the SSR stable wind speed range is analyzed by means of the amplitude phase contours plot criterion. The results show that the less the Type-III wind turbines in the HWF, the larger the stable wind speed range and the higher the stability margin, which can provide reference for the stability design in full wind speed range of actual hybrid wind farms. Simulation plat based on Matlab/Simulink was built to verify the correctness of the model and analysis results.

ABSTRACT. Pole-to-pole faults of MMC results in a significantly large fault current due to the discharge of submodule capacitors. The fault current not only risks damaging the MMC but also demands a considerable breaking capacity from the DC circuit breaker. This paper introduces two novel active fault current limiting methods (AFCLs), namely virtual impedance-based and energy control-based AFCL. The first method, utilizes circulating current feedforward to introduce a virtual arm impedance to suppress the rate of rise of the fault current. On the other hand, the second method relies on control of the  internally stored energy of the MMC to automatically minimize the number of submodules that get discharged during a pole-to-pole fault. Therefore, both the dc-side fault current and arm current can be effectively suppressed. The proposed methods does not require fault detection, and its response is proportional to the rate of rise of the fault current. Case studies are presented to verify the proposed designs.

ABSTRACT. This work investigates the ripple content on DC-link capacitor for reduced common mode voltage PWM schemes which directly impacts its thermal stress and resultingly its lifetime. The piece-wise sinusoidal form of DC-link current has been utilized for computing a closed-form RMS current expression that is further analyzed for varying load and mod-indexes. The RMS DC-link current of conventional space vector PWM has also been analyzed as base case. It would be seen that these modified PWMs mostly put higher ripple stresses on capacitor. However, there are specific regions where one PWM scheme can be employed for reducing DC-link stress.

ABSTRACT. This paper proposes the use of an adaptive voltage controller for the inner loop of grid-forming inverters. The proposed controller is suitable for grid-connected and islanded microgrid applications. The grid impedance uncertainty at the point of common coupling is considered in the design of the adaptive voltage controller.

ABSTRACT. On-vehicle integration of photovolatics can extend the range of electric vehicles by a useful amount each day. However, partial shading can significantly limit PV power production even in stationary installations, and this is expected to be more severe in vehicles. Differential power processing (DPP) approaches can maximize PV output power despite partial shading. This work presents a PV-to-isolated-bus DPP architecture specifically for electric vehicle integration and a converter module that is designed to be extensible and inexpensive. The proposed architecture uses the vehicle's existing low voltage battery as the common bus for the DPP modules and reuses the existing on-board charger to interface the solar string to the high-voltage battery. The proposed converter module achieves maximum power point tracking (MPPT) for the cell(s) it is connected to without requiring any communication or power transfer across the isolation barrier while allowing bidirectional power with synchronous rectification. The proposed architecture offers an inexpensive solution with high system efficiency and simple control that scales easily to large numbers of DPP units. The paper will include modeling of the advantages of the architecture, experimental characterization of the proposed DPP module, and experimental demonstration in a multi-cell, multi-DPP system.

ABSTRACT. GaN devices offer ultra-fast switching, superior electrical performance, and radiation hardness – making them a favorable choice for pulsed power applications. One potential drawback is the high dynamic on-resistance of GaN devices under switching conditions, caused by electrons occupying trap states when exposed to high blocking voltages. Previous investigations have observed high dynamic Ron in continuous switching conditions, sometimes with accurate measurements only after long delays from the switching instant – making their data of limited value for low-duty, high-speed pulsed power systems. This work proposes a fast measurement approach with minimal auxiliary circuitry, designed specifically for pulsed conditions.

ABSTRACT. This digest discusses the weight optimization of a receiver for wireless drone charging applications. A system-level design and optimization method is presented to maximize the gravimetric power density of the onboard receiver components. A 200 W, GaN-based prototype is implemented to validate the modeling. The current gravimetric power density of the entire receiver is 1.71 W/g. The prototype achieves a peak 90.7 % DC-DC power transfer efficiency from inverter input to rectifier output, and a peak 97 % buck efficiency.

ABSTRACT. This paper presents the design of an LLC DC transformer (DCX) capable of directly converting power from a regulated 48 V DC bus to the point of loads (PoLs). An LLC DCX architecture is proposed, which can achieve a high voltage conversion ratio through the series connection of the transformer windings and enable high output current through the parallel connection of multiple rectifiers. The optimization of the high-frequency transformer is introduced. Finally, a 48-to-1 V DCX prototype with 300 A output current is constructed, exhibiting a power density of 1kW/in3 and an estimated full load efficiency of 90%.

ABSTRACT. Bridgeless Power Factor Correctors (PFC) with a controller utilizing rectified ac variables can benefit from well-established strategies and circuits employed in PFCs with diode-bridge front-end. The grid voltage polarity is detected to compute the rms value of the grid voltage, and also used to generate and route the gate signals for the power devices. However, depending on the implementation, grid voltage disturbances may propagate through the polarity detection and RMS calculation stages, leading to a degradation of the input current and output voltage. This issue is addressed in this manuscript by investigating a single-phase bridgeless totem-pole (TP) PFC through simulation and proposing the replacement of the conventional implementation with a frequency-locked loop (FLL) to enhance the converter dynamics.

ABSTRACT. This paper presents a four-level flying capacitor multi-level (FCML) converter operating as a drain supply modulator (DSM) for radio frequency power amplifiers (RFPAs). The proposed FCML-based DSM excludes the standard output LC filter and employs a direct multi-level DSM approach, changing the output according to the RFPA level demand. A simple state-machine-based controller ensures that the flying capacitor voltages remain balanced for arbitrary sequences of requested output voltage levels. A GaN-based hardware prototype is currently being manufactured and the experimental results will be provided in the full paper.

ABSTRACT. High voltage drivers are needed for electrostatic and piezoelectric actuators in small-scale electro-mechanical systems such as microrobotics and haptics. This paper presents a miniaturized platform for a modular switched-capacitor-based driver that uses a low voltage auxiliary boost circuit to boost from 3.7V to 1.5kV. The platform uses a small PCB interposer to house the 180nm HV-SOI CMOS integrated circuit and all associated passives and connectors. Providing stacking modularity, the platform achieves 70x volume reduction while delivering up to 4W of reactive power to high voltage dominantly capacitive loads with over 96% reactive power efficiency.

ABSTRACT. Broad-scale modeling and optimization play a vital role in the design of advanced power converters. Optimization is normally implemented via brute force iterations of design variables or utilizing metaheuristic techniques which are time consuming for a wide range of potential topologies, device implementations, and operating points. Recently, discrete time state-space modeling has shown merits in rapid analysis and generality to arbitrary circuit topologies but has not yet been utilized under rapid optimization techniques across multiple converter parameters. In this work, we investigate methods to incorporate a rapid projection-based optimization technique to leverage discrete time state-space modeling and showcase the approach in the power converter design process. The method is validated on a 48-to-1V converter designed using the proposed techniques.

ABSTRACT. Many modular multilevel converters (MMCs) and their variations exist today for high-modulation applications, also called high-AC/low-DC voltage applications. These include conventional full-bridge-MMCs, alternate-arm-converters, and hybrid-MMCs, among others. Each with unique advantages and limitations. While these converters are well understood, no research has yet quantifiably compared these topologies. This absence of proper quantification and selection guidelines is hampering industrial adoption of these excellent modular converters. This research fills this gap, comparing major MMC variants in terms of devices utilized, submodule capacitance and losses. Clear guidelines on optimal operation region for each variant are explored and discussed. 20kW testbench is also developed.

ABSTRACT. This paper proposes an improved burst for dual active bridge (DAB) converters to achieve zero voltage switching (ZVS) in full-load-range. The proposed method introduces an oscillation of device voltages which can be resonant to zero during the non-switching interval of burst mode. By discretely adjusting the oscillation period, the switching devices can achieve ZVS even at light-load conditions. The effectiveness of the proposed method was verified by experiments with a fabricated DAB converter, which consisted of SiC-MOSFETs and was operated at 100-kHz.

ABSTRACT. The digest presents the design of a resonant gate drive (RGD), which can be employed in various high-frequency power converters where active bridges operate at ∼ 50% duty cycle. The RGD design is based on a typical LCLC resonant tank, which provides the flexibility to utilize the magnetizing and leakage inductances of a small gate-drive isolation transformer. The proposed RGD is simple to implement, does not require auxiliary supplies or high-side gate drivers, has low losses, and provides isolation and intrinsic dead time. Simulation results are shown to validate the RGD when applied to a 1 kW, 6.78 MHz GaN-based wireless power transfer (WPT) system. The results show that the dead time generated by the resonant gate driver is sufficient to achieve ZVS of the power-stage GaN-FETs. Experimental validation of the RGD and the complete WPT system architecture will be provided in the final paper.

ABSTRACT. The series-capacitor buck (SCB) converter has been demonstrated to be a compact and highly-efficient alternative to the multi-phase buck converter in data center power delivery applications. For highest possible power density, there is a desire to minimize the total flying capacitance in this topology. However, for insufficient flying capacitance, a discontinuous capacitor voltage mode (DCVM) manifests, leading to an imbalance in inductor currents. This work characterizes this imbalance by developing a clamped steady-state model and proposes recovery of balancing by driving the branches with modified duty cycles in a constant power regime. Validation of the model and recovery of balanced inductor currents are demonstrated on a 4-branch SCB prototype.

ABSTRACT. In this work, we combine two popular power conversion topologies: the flying capacitor multilevel (FCML) converter and the LLC converter, to arrive at the proposed flying capacitor LLC (FCLLC) solution with advantages of each. The FCLLC comprises a step-down resonant switched-capacitor (ReSC) stage followed by a resonant transformer-based stage which provides soft charging of the flying capacitors, galvanic isolation, and an additional step-down through the transformer. These conversion ratios are multiplied, making the FCLLC especially suited for extreme conversion ratios. An FCLLC prototype capable of direct 400 V to 1 V conversion is fabricated, which demonstrates the potential this topology has for future datacenter applications.

ABSTRACT. In addition to increasing the output current, an interleaved buck converter can significantly reduce the current ripple at the output. However, the bottleneck of the interleaved buck converter application is the unbalanced inductor current. In order to achieve the current balance in the entire section of the duty cycle, this paper proposes a control architecture with interleaved alignment, using the adjustment of the duty cycle and phase offset, to analyze and realize a high output current and current balanced in three-arm series capacitor interleaved buck converters (SCIBCs) in this work. By analyzing and modeling the SCIBC and using a single-loop control architecture that feedbacks the output voltage, the proposed current-balancing method balances all inductors current within the full duty cycle without any current sensor. In addition, the small-signal transfer function of the control signal to the output voltage and design the controller parameters are derived according to the specification, the SCIBC with the proposed current-balancing control with PI controller of the alignment of interleaved phase angle are constructed. The high output current and current-balancing characteristics of the proposed circuit control architecture are validated in the simulation tool Simplis.

ABSTRACT. This digest introduces stacked inverter architecture for high-frequency capacitive wireless power transfer (WPT) systems suitable for electric vehicle (EV) charging. The proposed architecture combines the output voltages of two or more high-frequency inverters using parallel-in series-out air-core transformers, thereby increasing the power transfer capability of a capacitive WPT system. A comprehensive methodology to design a stacked inverter-based capacitive WPT system is presented by leveraging magnetizing and leakage inductances of the air-core transformers. Furthermore, performance of the stacked inverter-based capacitive WPT system is analyzed in terms of number of inverters stacked for given system specifications and also, in the presence of mismatches between stacked inverters. Finally, a 6.78-MHz 12-cm air-gap 1-kW stacked inverter-based capacitive WPT prototype is designed, built and tested to validate the theoretical predictions.

ABSTRACT. This digest presents a synchronous resistance compression network (RCN) based resonant dc-dc converter utilizing L-section matching network in its transformation stage. Synchronous RCN comprises differential reactances connected to two phase-shifted rectifiers at its output controlled to enable soft-switching across wide load ranges. An L-section matching network is proposed to reduce the required number of turns in the transformer for large step-up designs. The digest also discusses the optimization of the matching network and compares the efficiency performance of the new design with that of a conventional design that uses a large turns ratio transformer. A 300 W, 12- V input, and 200-400 V output prototype converter is designed, built, and tested. The converter achieves a peak efficiency of 94.86% in the conventional design.

ABSTRACT. This paper presents an inspection method for analyzing the soft-charging capabilities of switched-capacitor converters (SCC). The method utilizes a multi-step approach to derive the voltage changes (∆V) across each flying capacitor leading to an intuitive way of understanding the behavior of hybrid-SCC topologies on circuit level. The method is presented by examples and can also be used with topologies using multi-phase and multiple inductors. It provides insights and improvements in the design of various types of hybrid converters including double step-down (DSD) and dual-inductor hybrid converters (DIHC).

ABSTRACT. An efficient first stage interleaving technique for Boost Extender topology is presented. A unique single conversion operation of the boost extender topology, and current stress distribution between the modules pose a challenge on creating a successful and efficient interleaving scheme with this converter. A mechanism is developed, where a supporting first stage in a multilevel high voltage gain structure is added. The supporting stage shares the high current stress of the first boosting stage, compatible with interleaving technique, which reduces the ripples of each inductor along with the input and first stage output capacitor ripples. In addition, the voltage multiplication modules are shared between the interleaved stages providing significant component reduction comparing to traditional interleaving schemes. The concept was validated on a 280W experimental laboratory prototype. Theoretical predictions well agree with simulation and experimental results.

ABSTRACT.  Planar magnetics are being actively explored in a wide range of power electronics applications, with argumentation of their low cost being a key aspect of their popularity. However, unlike conventional wire wound magnetic components, their costs do not scale convincingly with volume owing to a strong dependence on manufacturing complexity. This paper assesses the cost factors impacting the design of planar magnetic windings. This is done qualitatively, through direct manufacturer survey, and quantitatively by presenting a cost modelling approach derived from cost quotations of exemplar planar transformers. The developed model fits excellently to the data it was trained on (6.5% error). It is shown to have good predictive power on designs that are well-represented by the training data (though not in the training set, 13% error), and reasonable performance (20-58% error) on designs which have limited representation in the training data.

ABSTRACT. This paper proposed a piecewise analytical transient (PAT) model of SiC MOSFET and SiC Schottky diode pair that can simulate the switching transient in a complex power electronics system with a large number of power switches. The proposed model eliminates the stiffness caused by the parasitic parameters and has a fast simulation speed. The nonlinearity of the junction capacitances is considered. Double pulse tests are conducted to verify the accuracy of the model.

ABSTRACT. Hybrid switched-capacitor converters (HSCC) offergreat potential for high efficiency and power density comparedto purely capacitor- or inductor-based converters. However, therecent proliferation of HSCC topologies has made it difficultto choose the best one for a particular application. This paperpresents a benchmarking framework that allows for direct comparisonof popular HSCC topologies by analyzing various performancemetrics such as passive component volume and bandwidth.By comparing all topologies at the same efficiency, same inductorripple, and same output voltage ripple, this approach generatesguidelines for topology selection and optimization, which can aidin wider industrial adoption and exploration of new topologies.

ABSTRACT. This paper investigates the utilization of a planar Rogowski coil located between the dc-bus traces to measure the half-bridge's switching transient current. The senor fundamentals are explained, and design considerations are derived. A compact current sensor prototype is designed and fabricated for a SiC MOSFET half-bridge application. Simulations show a minimal increase in the current communication loop stray inductance of 399 pH. The current sensor's performance is experimentally validated with double pulse tests.

ABSTRACT. A 48V-1V point of load converter presents a case with significant switching and conduction losses. One effective solution to reduce these losses is to achieve a modularized design, where each module only shares a small fraction of the input voltage and load current. In this digest, we investigate a modular structure based on series capacitor Buck topology. In addition, we also proposed a solution to free duty cycle limitation associated with series capacitor Buck converter so that excellent transient response is achievable. An experimental prototype has been built and tested to validate the proposed method.

ABSTRACT. Non-inverting buck-boost (NIBB) converters used PV in applications where input and output voltages vary over a wide range, typically require both step-up and step-down conversion ratios. Conventional approaches include designing complex mode switching PI/PID compensators for different modes of operation. This digest proposes a unified sliding mode control approach in order to achieve smooth mode transitions, while at the same time ensuring excellent dynamic performance over a wide operating range. The proposed approach is verified using simulation case studies on a NIBB converter for an input voltage of 8 V to 60 V, output voltage of 40 V for solar photovoltaic (PV) systems.