CAREER: A Multi-dimensional Study of Electromagnetic Interference in Wide Bandgap Power Electronics: Modeling, Estimation, and Mitigation

在宽禁带半导体技术发展的前沿领域，电磁干扰(EMI)是需要重点研究解决的问题之一。这项研究的目的是对EMI进行多物理建模和表征，并开发新的有源EMI缓解方法，以促进噪声免疫和高密度WBG功率转换。作为这项研究的产品，新型滤波解决方案适用于广泛的功率转换器拓扑，应用范围广泛，包括电机驱动、航空电子、军事、空间和数据中心。我们研究开发的多物理EMI模型和解决方案可以扩展到专用集成电路(ASIC)级和电源管理IC级高速微电子产品。该项目的研究部分融合了电学、计算机科学和数学优化，以推动实现具有物理信息的EMI模型的下一代WBG电力电子。该项目的高度跨学科性质将使具有系统工程、电气和电子工程背景的下一代学生受益，他们将获得科学知识和工程技能，以开发抗噪音的电力电子电路。总体而言，长期目标是(A)将先进的EMI建模理论应用于下一代电力电子应用，以及(B)培训不同的工程师群体，使他们意识到电力电子领域的主要设计挑战，并为满足美国未来的能源需求做好充分准备。在这项研究中，我们提出了新的方法，通过识别高频功率转换器中的谐振电路路径，对EMI源、传播路径和寄生噪声的耦合动力学进行解析建模，从而综合出新的有源-混合滤波补偿方案。提出的基本突破是：(A)建立共模(CM)噪声源的数学模型等价物，然后估计高频功率变流器中的寄生分量；(B)研究EMI对高频功率变换中控制回路稳定性和动态性能的耦合影响；(C)针对高密度功率变换，提出基于多约束体积优化的EMI滤波器设计的统一方法；以及(D)将共模和差模(DM)滤波网络耦合拓扑集成，以实现体积优化和元件数量减少。据估计，采用拓扑集成的有源-混合EMI滤波器可以达到50 kW/L的功率密度，与最先进的完全无源解决方案相比，体积缩小了70%，同时保持了99.8%的效率和卓越的电能质量(1度相移)和改进的CM性能。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

In the advancing domain of wide bandgap (WBG) semiconductor technology, electromagnetic interference (EMI) is one of the major concerns, resolving which needs significant research focus. The objective of this research is to perform multi-physics modeling and characterization of EMI and develop new active EMI mitigation methodologies to facilitate noise immune and high-density WBG power conversion. The novel filtering solutions as a product of this research are applicable to a broad family of power converter topologies with a variety of applications including motor drive, avionics, military, space, and data centers. Multi-physics EMI modeling and solutions developed in our research can be extended to application specific integrated circuit (ASIC)-level and power management IC-level high-speed micro-electronics. The research components of this project integrate electrical science, computer science, and mathematical optimization to advance the realization of next-generation WBG power electronics with physics-informed EMI models. The highly interdisciplinary nature of the project will benefit next-generation students with systems engineering, electrical, and electronics engineering backgrounds who will gain scientific knowledge and acquire engineering skills to develop noise-immune power electronic circuits. Overall, the long-term goals are to (a) employ the advanced EMI modeling theories into next-generation power electronics applications, and (b) train the diverse group of engineers to make them aware of the major design challenges in the power electronics field and well-prepared for addressing the future energy needs of the United States. In this research, we propose new methodologies for analytical modeling of the EMI sources, propagation paths, and the coupling dynamics of parasitic noises by identifying the resonating circuit paths in a high-frequency power converter, and thereby synthesize new active-hybrid filtering compensation schemes. The fundamental breakthroughs proposed are: (a) development of mathematical model equivalents for common mode (CM) noise sources followed by estimation of parasitic components in high-frequency power converters, (b) studying the coupling effect of EMI on the control loop stability and dynamic performance in high-frequency power conversion, (c) formulation of a unified methodology for multi-constraint volumetric optimization-based EMI filter design for high-density power conversion, and (d) coupled topological integration of CM and differential mode (DM) filter networks for volumetric optimization and component count reduction. It is estimated that the topologically integrated active-hybrid EMI filters can achieve a power density of 50 kW/L, with a shrinkage of 70% volume compared to state-of-the-art fully passive solutions, while maintaining an efficiency of 99.8% and superior power quality (1 degree phase displacement) and improved CM performance.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.