Three-level inverters suffer from higher parasitic inductance due to the increased number of series-connected power switches in a single current commutation loop (CCL) and geometrically larger size of CCL with compared to their 2 level counterparts. As such semiconductors are subjected to higher voltage stress and severe ringing at the switching transient. To solve this problem, an optimized 3-level T-type NPC power module has been proposed with a hybrid combination of the switch (SiC MOSFET + Si IGBT) rated for 1200V/180A. A step by step design methodology has been presented in this work. A stacked DBC with via has been utilized to have a vertical power loop as such, lower loop inductance. All the commutation loop has been identified and corresponding loop inductances have been optimized as low as 4.5 nH and dynamic current sharing between paralleled devices has been ensured.

A 1200V/650V/160A SiC+Si IGBT 3-Level T-type NPC Power Module with Optimized Loop Inductance

More-electric aircraft (MEA) concept is getting popular as it increases system efficiency with less costs. The concept leads to numerous challenges including power density and volume. This paper presents a novel application of model based optimization (MBO) on 3L-TNPC inverter to determine the optimal semi-conductor selection, modulation strategy, and passive components. The algorithm extensively relies on double fourier integral analysis (DFIA) for selection of the modulation strategy and filter design. The algorithm helps optimize the design of a 450 kVA three-phase full SiC-MOSFET based inverter with peak efficiency of 99.47%. And the estimated converter efficiency at full-load is 99.10%. The algorithm developed is scalable and can be extended for future high-performance inverter design applications.

Model Based Optimization of Propulsion Inverter for More-Electric Aircraft Applications Using Double Fourier Integral Analysis

More electric aircrafts (MEA) recently attracts increasing attention due to improvements to efficiency, reduce weight, fuel cost, and carbon emissions. High-specific-power machines, and corresponding integrated motor drives have been identified as the crucial enabling technology for the realization of more electric aircraft propulsion. Such a concept requires converter working under depressurized environments, which poses challenges to the electric insulation due to partial discharge (PD). This threatens the reliability of the drive system. This paper proposes an insulation design and optimization methodology for a laminated busbar in a 450-kVA electric-aircraft motor drive. The design aims to avoid any partial discharge in the insulation and also achieves optimized busbar stray inductance. To achieve the targets, partial discharge inception voltage (PDIV) with respect to air pressure was obtained by the experiment, and the experimental results were used as PD -free design criteria in electric field simulation of the laminated busbar. Then, the insulation structure regarding insulation material selection and thickness selection was optimized by taking both the PD-free criteria and parasitic inductance into consideration. The proposed design procedure provides valuable references for future laminated busbar design, which is used for the MEA system.

Insulation Design and Optimization of Laminated Busbar for More Electric Aircraft Motor Drive under High Altitude and Depressurized Environments

http://manuscript.elsevier.com/S2542435122001957/pdf/S2542435122001957.pdf

Heat transfer innovations and their application in thermal desalination processes

This paper presents a prototype of a 450 kVA inverter system by using 3-level T-type neutral-point-clamped converter (3L-TNPC). The prototype features high-power density and high-efficiency design. The design highlights an improved busbar structure, which achieves lower stray inductance than published literature of 3-level converters with limited temperature rise. The improved power-loop design provides sufficient room for accelerating switching speed. Over 67.7% of switching loss reduction is achieved by choosing optimal gate resistance (1 Ω). The paper also focuses on compact integration of signal sensing, conditioning circuits, and controller circuits to improve power density. The design strategy for the sensing system is discussed to optimize the trade-off between compact sensing and noise distortion. As a result, the converter reaches the power density of 26.6 kW/L. The peak efficiency of 99.47% is measured at 266 kVA under 70 kHz switching frequency.

A Three-phase 450 kVA SiC-MOSFET Based Inverter With High Efficiency and High Power Density By Using 3L-TNPC

\n With the increase of electronic device power density, thermal management and reliability are increasingly critical in the design of power electronic systems. First, increased density challenges the capability of conventional heat sinks to adequately dissipate heat. Second, higher frequency switching in high voltage, high current, wide bandgap power modules is creating intensified electromagnetic interference (EMI) challenges, in which metallic heat removal systems will couple and create damaging current ringing. Furthermore, mobile power systems require lightweight heat removal methods that satisfy the heat loads dissipated during operation. In this effort, we introduce an additive manufacturing (AM) pathway to produce custom heat removal systems using nonmetallic materials, which take advantage of impinging fluid heat transfer to enable efficient thermal management. Herein, we leverage the precision of additive manufacturing techniques in the development of three-dimensional optimized flow channels for achieving enhanced effective convective heat transfer coefficients. The experimental performance of convective heat removal due to liquid impingement is compared with conventional heat sinks, with the requirement of simulating the heat transfer needed by a high voltage inverter. The implementation of nonmetallic materials manufacturing is aimed to reduce electromagnetic interference in a low weight and reduced cost package, making it useful for mobile power electronics.

Additive Manufactured Impinging Coolant, Low Electromagnetic Interference, and Nonmetallic Heat Spreader: Design and Optimization


 With the increase of electronic device power density, thermal management and reliability are becoming increasingly important. First, increased density challenges the capability of conventional heat sinks to adequately dissipate heat. Secondly, higher frequency switching in wide bandgap power modules is introducing new issues in electromagnetic interference (EMI) in which metallic heat removal systems will couple and create damaging current ringing. Lastly, lightweight heat removal is required to meet the increasing needs of mobile power systems. In this effort we introduce an additive manufacturing pathway to produce custom-tailored heat removal systems using non-metallic materials, which take advantage of convective heat transfer to enable efficient thermal management. Herein, we leverage the precision of AM techniques in the development of 3D optimized flow channels for achieving enhanced effective convective heat transfer coefficients. The experimental performance of convective heat removal due to liquid impingement is compared with conventional heat sinks, with the requirement of simulating the heat transfer needed by a high voltage inverter. The implementation of non-metallic materials manufacturing is aimed to reduce EMI in a low weight and reduced cost package, making it useful for mobile power electronics.

Additive Manufactured, Low EMI, Non-Metallic Convective Heat Spreader Design and Optimization

The spectral decomposition of graph adjacency matrices is an essential ingredient in the design of graph signal processing (GSP) techniques. When the adjacency matrix has multi-dimensional eigenspaces, it is desirable to base GSP constructions on a particular eigenbasis (the `preferred basis'). In this paper, we provide an explicit and detailed representation-theoretic account for the spectral decomposition of the adjacency matrix of a Cayley graph, which results in a preferred basis. Our method applies to all (not necessarily quasi-Abelian) Cayley graphs, and provides descriptions of eigenvalues and eigenvectors based on the coefficient functions of the representations of the underlying group. Next, we use such bases to build frames that are suitable for developing signal processing on Cayley graphs. These are the Frobenius--Schur frames and Cayley frames, for which we provide a characterization and a practical recipe for their construction.

Skyler Hudson

Papers

Additive Manufactured Impinging Coolant, Low Electromagnetic Interference, and Nonmetallic Heat Spreader: Design and Optimization

Additive Manufactured, Low EMI, Non-Metallic Convective Heat Spreader Design and Optimization

Frames for signal processing on Cayley graphs