YWYulei Wang

Yulei Wang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yulei Wang has a total of 42 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • DC EMI Filter With Characteristic Matching to Enhance Insertion Loss and Mitigate Mid-Frequency Degradation for High-Power Motor Drive Systems

    by Yulei Wang, Jiaming An, Linke Zhou, Di Wang, Mohamed Abdalmagid, Piranavan Suntharalingam, Mikhail Goykhman, Armen Baronian, Ali Emadi

    EMI filters are a primary solution for mitigating conducted EMI emissions in high-power variable-speed motor drive (VSD) systems. However, significant mid-frequency (MF) degradation has been observed when a DC EMI filter is integrated into a VSD system. This paper investigates and identifies the mechanism behind this degradation, revealing that cross-coupling between the non-ideal characteristics of the filter output and the inverter input initiates the issue, with a pronounced mismatch in their MF characteristics accelerating the degradation process, eventually resulting in near-complete filter failure. To address this challenge, a two-stage y-cap design with characteristic matching is proposed. Through matching with the inverter input, the first stage achieves effective decoupling and mitigates MF degradation, while the second stage, in combination with a common-mode (CM) inductor, provides sufficient attenuation. Extensive experiments validate the effectiveness of the proposed characteristic-matching DC EMI filter, demonstrating an additional improvement of over 20 dB in CM insertion loss and 30 dB in DM insertion loss within the MF degradation region. This improvement ensures that the conducted EMI emissions of the high-power VSD system comply with the DO-160G standard limits.

  • Design of a High-Power-Density 1-MVA Three-Level Inverter for Electric Aircraft Propulsion

    by Di Wang, Linke Zhou, Jiaming An, Yulei Wang, Samuel Hemming, Mohamed Hefny, Giorgio Pietrini, Piranavan Suntharalingam, Mikhail Goykhman, Armen Baronian, et al.

    The development of megavolt-ampere (MVA)-class three-level inverters has become the focus in advancing electric aircraft propulsion (EAP) systems. In this article, a highly power-dense three-level inverter design is demonstrated, targeting 1-MVA output power at a 1.5-kV dc bus voltage. The design focuses on optimizing the inverter’s power density at both the system and component levels, including topology and component selection, power stage layout, lightweight bus bar design, as well as cooling system and electromagnetic interference (EMI) filter designs. Through comprehensive optimization, the designed inverter achieves a total weight of only 22.57 kg and a gravimetric power density of over 44.3 kVA/kg, setting a new benchmark for MVA-class EAP. A prototype has been fabricated and tested under laboratory conditions, with experimental results validating both the electrical and thermal performance of the proposed inverter, confirming its capability for high-efficiency operation in demanding aerospace environments.


  • Near-Magnetic Interference in Rogowski Coil for Power Electronic Applications: Mechanism and Mitigation

    by Yulei Wang, Rachit Pradhan, Di Wang, Shreyas B. Shah, Linke Zhou, Jiaming An, Mohamed Abdalmagid, Giorgio Pietrini, Ali Emadi

    Rogowski coil (RC), valued for its high bandwidth, galvanic isolation, and ease of integration, is widely employed in high-speed integrated power electronics. However, near-magnetic interference (NMI) arising from inductive coupling with adjacent current sources is often underestimated, which undermines the RC’s robustness in crucial applications such as characterization, protection, and control. This article first reveals the inevitability of NMI in RCs. Building upon this, the mutual inductance equation between the RC and external interference sources is derived in detail, establishing a solid theoretical foundation for efficient and quantitative interference evaluation. The derivation methodology is further generalized to arbitrary points in space, allowing for the quantitative analysis of the effect when the circuit under test (CUT) is offset from the RC’s center. Theoretical analysis highlights the number of turns as a key factor, demonstrating that appropriately increasing it can effectively mitigate the influence of both external NMI and internal CUT eccentricity. The inherent design tradeoff between measurement bandwidth and interference immunity is further analyzed. Design guidelines for establishing the upper and lower limits of the number of turns are derived, and the differential mode rejection ratio is introduced as an intuitive metric to quantify the interference suppression capability. The accuracy of the proposed NMI model and the effectiveness of mitigation strategies are corroborated by Ansys Q3D simulations and extensive frequency- and time-domain experiments.


  • Planar Rogowski Coil Based Short-Circuit Detection and Partial Open-Circuit Recovery of the Resonant Capacitor in Series-Resonant DC Transformers

    by Rachit Pradhan, Shreyas B. Shah, Yulei Wang, Ahmed Elezab, Samuel Hemming, Giorgio Pietrini, Mohamed I. Hassan, Piranavan Suntharalingam, Mario F. Cruz, Ali Emadi

    The 10 kW Series-Resonant DC Transformer (SR-DCX) enables both switch and resonant element count reduction and is a building block of a bidirectional two-stage dc–dc converter for 600–28 V MIL-PRF-GCS600 A and MIL-STD-704F compliant applications. The dc–dc converter is constructed with its first stage as a buck converter and the second stage as the SR-DCX. Failure mode assessment of the SR-DCX indicates a single-point failure on the resonant capacitor that will either prohibit or degrade the operation of the dc–dc converter, depending upon the failure mode. This article discusses the design of a planar Rogowski coil in accordance with MIL-PRF-31032 and its signal chain optimization for the measurement of ac current on the high-frequency link. We introduce a novel failure detection method to simultaneously identify both partial open-circuit (OC) and short-circuit failures in the resonant capacitor network by re-purposing the ac link current measurement, conventionally used for over-current protection. The method also enables recovery from partial OC failures to restore normal operation. The proposed scheme aids in raising system-level warnings for redundancy activation prior to catastrophic failures in the dc–dc converter for high-reliability applications.


  • Systematic Efficiency-Density Co-Optimization of 100 kW GaN Traction Inverter: Methodology and Integration

    by Mingrui Zou, Peng Sun, Zheng Zeng, Yulei Wang, Jiakun Gong, Yuxi Liang, Liang Wang

    The increasing pursuit of high efficiency and power density in EV traction inverters is accelerating the deployment of the wide bandgap device, including silicon carbide (SiC) and gallium nitride (GaN). As the efficiency-density boundary of the SiC traction inverter has approached its limitation, the traction inverter utilizing GaN devices, with lower power loss and higher switching speed, is considered as a promising solution. Focusing on the GaN traction inverter for EV applications, this article proposes a design methodology considering the co-optimization of the pursued efficiency and power density objectives. Based on the created mathematical models of design concerns, including power loss, dc-link ripple, and thermal resistance, the efficiency-density design domain of the GaN inverter is established, upon which the optimal design space of the efficiency-density is determined through Pareto frontier analysis. Based on the co-optimization results, a 100 kW / 400 V GaN traction inverter prototype is compactly integrated, incorporating meticulously designed key components, including the GaN power module, dc-link capacitor, and heat sink. Experimental validation confirms the effectiveness of the proposed co-optimization methodology and demonstrates the feasibility of the integrated GaN inverter, achieving the high power density 62.1 kW/L and peak efficiency 99.3%.


  • Practical Thermal Solution for 1-MVA Three-Level ANPC Inverter in Aerospace Applications

    by Linke Zhou, Mohamed Hefny, Samuel Hemming, Di Wang, Jiaming An, Yulei Wang, Giorgio Pietrini, Piranavan Suntharalingam, Mikhail Goykhman, Ali Emadi

    In aerospace applications, high-performance cooling systems are essential to ensure the reliability of power inverters under extreme operating conditions. This paper presents the design and analysis of a practical pin-fin liquid cooling system for a 1 MVA silicon carbide (SiC)-based three-level active neutral point clamped (ANPC) inverter. The thermal and hydraulic performance was evaluated through finite element analysis (FEA) models of SiC power modules and computational fluid dynamics (CFD) simulations, confirming the heat transfer efficiency required for aerospace environments at 70°C coolant temperature. The cooling solution leverages a staggered pin-fin arrangement to improve heat transfer efficiency while maintaining manufacturability and cost-effectiveness. Two different cold plate configurations—series and parallel—were compared for uniform fluid distribution and cost-effectiveness. Special attention was also given to sealing design, ensuring reliability in preventing coolant leakage. The system effectively manages over 7.2 kW power loss, maintaining junction temperature Tj safely below the 150°C limit. Experimental tests validated the cooling system’s hydraulic and thermal performance, demonstrating its ability to handle high power losses with low pressure drop, making it a robust and cost-effective solution for high-power aerospace inverters.


  • Optically Isolated Rogowski Coil: Non-Invasive Current Measurement With High d v /d t Immunity for WBG Device

    by Jiakun Gong, Yulei Wang, Yuxi Liang, Mingrui Zou, Liang Wang, Peng Sun, Zheng Zeng

    Rogowski coil (RC) has been extensively employed for current measurement of power devices owning to its non- invasiveness, flexibility and accuracy. However, the increasing blocking voltage and accelerated switching speed of wide-bandgap (WBG) devices impose significant challenges on RC-based measurement systems, primarily manifested as more severe electric field interference. In this letter, a novel optically isolated Rogowski coil (Opt-RC) is proposed to address the electric field interference issues. By incorporating optical isolation, the original common-mode (CM) interference pathways are effectively interrupted, substantially mitigating CM noise without modifying the coil structure and compromising bandwidth. Furthermore, the excellent insulating properties provided by optical fiber imparts the Opt-RC with significant potential for accurate current measurement in high-voltage scenarios. Comprehensive experiments confirm the electric field interference immunity and excellent insulation of the proposed Opt-RC.


  • Eddy-Current Effect in Shielded Rogowski Coil for Power Electronic Applications: Mechanism, Essence, and Countermeasure

    by Yulei Wang, Jiakun Gong, Zheng Zeng, Mingrui Zou, Liang Wang

    With next-generation wide-bandgap power devices achieving faster switching speeds and higher dv/dt, high-bandwidth shielded Rogowski coil (RC) is being actively explored to accurately capture rapid current transients while effectively suppressing external common-mode interference induced by dv/dt. However, the eddy-current effect inevitably parasitizes the external shielding of the shielded RC, resulting in significant variations in its forward gain characteristics. This article first investigates the pivotal role of external shielding in achieving high bandwidth and interference immunity in RC design while also highlighting the inevitability of eddy-current effect within it. Following this, an in-depth analysis of the mechanism underlying eddy-current effect is presented, providing theoretical insights into the characteristic degradation of the shielded RCs forward gain, which follows a characteristic inductive–resistive–inductive response. Building on this foundation, the study further elucidates the physical essence of these behaviors: in the midfrequency range, resistive behavior is mainly driven by the eddy-current loss induced by the parasitic resistance of the external shielding, whereas in the high-frequency range, inductive characteristics are predominantly influenced by the field distortion resulting from external shielding degradation. Finally, a countermeasure is proposed to establish a corresponding capacitive–resistive–capacitive response by leveraging the inherent dual-transition characteristics of a “passive + active” integrator. This method enables the shielded RC to maintain a flat system gain across a broad frequency range (not less than 100 MHz) without altering circuit setup, increasing system complexity, or requiring cumbersome implementation. Extensive comparative experiments in both the frequency and time domains validate the accuracy of the established eddy-current model and confirm the effectiveness of the proposed compensation methodology.