Reza Mirzadarani

This paper presents a scalable transformer-coupled open-loop gate-driving technique that enables voltage balancing across series-connected GaN devices. High-voltage pulse generation with short rise times at kilovolt levels is a significant challenge. Conventional solid-state devices such as IGBTs and SiC MOSFETs, though capable of high blocking voltages, are limited in switching speed and cannot reach the nanosecond regime. Gallium Nitride (GaN) High-Electron-Mobility Transistors (HEMTs) due to their lateral structure offer low gate charge and ability to switch faster than 50 V/ns, are an attractive potential candidate. However, due to their lateral architecture, GaN HEMTs also have limited voltage-blocking capability, with most commercially available GaN power devices rated up to approximately 650 V, limiting their direct use in fast high voltage waveform generation at kilo-volts level. Series-connecting GaN devices can overcome this limitation but introduces severe voltage balancing challenges, as even minor gate-signal mismatch at nanosecond timescale can cause destructive imbalance. Conventional closed-loop balancing methods, are difficult to implement at GaN switching speeds due to feedback latency. This paper presents a simple open loop technique for driving GaN devices in series with a transformer-coupled gate driver. The proposed gate drive ensures simultaneous turn-on/off with identical gate signals across all devices. An ultrafast full-bridge GaN based inverter excites wideband gate-drive transformers designed to preserve the ultrafast transition speeds of individual devices, while providing high voltage isolation and near-equal voltage sharing establishing a scalable solution. Experimental results with two series-connected GaN HEMTs confirm nearly balanced voltage sharing at 1 kV across varying loads and currents.

This work presents a down-scaled validation of a medium-voltage, medium-frequency transformer (MFT) concept designed for high-current operation on the secondary side using multiple parallel paths. The design is based on a modular winding approach, which simplifies the construction process and conductor placement on the bobbin. A systematic design and optimization procedure is developed, combining analytical calculations and finite-element simulations to explore the mass-efficiency trade-off and to select a candidate design that meets specified leakage inductance and loss targets. The developed prototype serves as a proof of concept, demonstrating that the electrical, magnetic, and insulation requirements of the full-scale MFT can be effectively verified at reduced power levels. The fabricated prototype is tested under short-circuit and partial discharge conditions. The impedance measurements confirmed the expected resonance behavior, and the partial discharge test results verified sufficient insulation performance under high-voltage stress. The results provide experimental evidence for the scalability and feasibility of the proposed transformer design and offer guidelines for the use of 3D-printed supports, grain-oriented electrical steel cores, and windings in medium-voltage, medium-frequency transformer systems for hydrogen production applications.

This paper introduces a novel control strategy for Modular Multilevel Resonant converters (MMR) in Solid-State Transformer (SST) applications, with a focus on medium-voltage conversion for hydrogen electrolyzers. The article first reviews voltage control methods in MMR, analyzing their operational principles and regulation capabilities. A continuous modulation index control method with double-step staircase waveform modulation is then proposed, simplifying the control scheme to a single control variable while maintaining robust controllability. Meanwhile, the proposed approach maintains comparable power loss and harmonic performance to existing methods under the investigated operating conditions. Simulations and experiments are conducted to verify the feasibility and practical implementation of the proposed approach.

This study presents a current balancing technique for high-current windings in medium-frequency transformers (MFTs), particularly relevant to solid-state transformer (SST) applications. Handling high currents on the low-voltage high-current winding of MFTs is challenging due to skin and proximity effects. Conventional techniques, such as continuously transposed conductors (CTCs) and parallel winding paths, are applicable but have limitations in medium-and high-frequency applications such as SSTs due to skin and proximity effects. To address these issues, a modular and tunable compensation method is proposed, based on adding small, series-connected inductive elements (compensation toroids) to each parallel winding path. Experimental results from a prototype validate the proposed compensation technique, highlighting its effectiveness in mitigating unbalanced current distribution. Finite element analysis (FEA) and experimental validation across a wide frequency range (1–10 kHz) confirm the effectiveness of the method. The results demonstrate a significant reduction in current imbalance with minimal added losses or system impact.

To enhance the voltage handling capability of a switch, the series connection of switching devices is a cost-effective method that preserves many advantages of mature low-voltage devices. Dynamic voltage imbalance and electrical isolation for the devices at the HV side are two important challenges associated with series connection topology. Transformer-coupled gate drivers are excellent for providing both dynamic voltage balance and high galvanic isolation. However, they can only provide the switching function at the transformer pulse frequency. To generate complex waveforms of future power electronics-dominated grids, a switch with user-defined turn-on/off timing is required for testing grid assets under high-voltage conditions. This paper presents a simple, cost-effective open-loop gate driver that overcomes this limitation by introducing two sets of complementary pulse transformers to initialize programmable frequency and duty cycle. Successful experimental verification of the series-connected SiC MOSFETs prototype is performed at 3.2 kV at various frequencies and duty cycles. The paper also demonstrates that the measurement probes placed across series-connected MOSFETs significantly affect the voltage distribution and validates a compensation mechanism.

This paper summarizes the main results and contributions of the MagNet Challenge 2023, an open-source research initiative for data-driven modeling of power magnetic materials. The MagNet Challenge has (1) advanced the stateof-the-art in power magnetics modeling; (2) set up examples for fostering an open-source and transparent research community; (3) developed useful guidelines and practical rules for conducting data-driven research in power electronics; and (4) provided a fair performance benchmark leading to insights on the most promising future research directions. The competition yielded a collection of publicly disclosed software algorithms and tools designed to capture the distinct loss characteristics of power magnetic materials, which are mostly open-sourced. We have attempted to bridge power electronics domain knowledge with state-of-the-art advancements in artificial intelligence, machine learning, pattern recognition, and signal processing. The MagNet Challenge has greatly improved the accuracy and reduced the size of data-driven power magnetic material models. The models and tools created for various materials were meticulously documented and shared within the broader power electronics community.

In the production of green hydrogen, electrolyzers draw power from renewable energy sources. In this paper, the design of SST (Solid State Transformer) for large-scale H ₂ electrolyzers is benchmarked. The three most promising topologies are chosen for design and comparison, including Modular Multi-level Converter (MMC) based SST, Modular Multi-level Resonant (MMR) based SST, and Input-Series-Output-Parallel (ISOP) based SST. The distance between converter towers for insulation and maintenance, the insulation system of the transformer, and the cooling system are designed with practical considerations in order to have an accurate estimation of the volume and weight of the SST. Losses in the switches are calculated based on equations, and losses in passive components are calculated based on FEM simulation. The operating frequency for each topology is optimized to minimize loss, weight, and volume. The best of each topology is then compared with each other to identify the most suitable one for large-scale H ₂ electrolyzers.

Traditional methods such as Steinmetz's equation (SE) and its improved variant (iGSE) have demonstrated limited precision in estimating power loss for magnetic materials. The introduction of Neural Network technology for assessing magnetic component power loss has significantly enhanced accuracy. Yet, an efficient method to incorporate detailed flux density information—which critically impacts accuracy—remains elusive. Our study introduces an innovative approach that merges Fast Fourier Transform (FFT) with a Feedforward Neural Network (FNN), aiming to overcome this challenge. To optimize the model further and strike a refined balance between complexity and accuracy, Multi-Objective Optimization (MOO) is employed to identify the ideal combination of hyperparameters, such as layer count, neuron number, activation functions, optimizers, and batch size. This optimized Neural Network outperforms traditionally intuitive models in both accuracy and size. Leveraging the optimized base model for known materials, transfer learning is applied to new materials with limited data, effectively addressing data scarcity. The proposed approach substantially enhances model training efficiency, achieves remarkable accuracy, and sets an example for Artificial Intelligence applications in loss and electrical characteristic predictions with challenges of model size, accuracy goals, and limited data.