Litcius/Paper detail

Synergetic Control of Three-Phase AC-AC Current-Source Converter Employing Monolithic Bidirectional 600 V GaN Transistors

Neha Nain, Daifei Zhang, Jonas Huber, Johann W. Kolar, Kennith Kin Leong, Bhargav Pandya

20212021 IEEE 22nd Workshop on Control and Modelling of Power Electronics (COMPEL)26 citationsDOI

Abstract

AC-AC current-source converters (CSCs) are an interesting alternative to the widely-used voltage-source converters for next-generation motor drive applications, as they inherently feature continuous output voltages and minimum filter effort. This mitigates issues such as reflections on long motor cables, high-frequency motor losses, the need for expensive shielded motor cables, and common-mode currents that damage motor bearings. In this paper, we first propose a new synergetic control for an AC-AC CSC that always operates either the front-end rectifier or the output-side inverter stage with one phase clamped, reducing switching losses compared to conventional modulation that continuously operates all three phases. We then provide calorimetric switching loss measurements of a novel 600 V, 140 mW monolithic bidirectional GaN transistor, and model the impact of the third, passive switch present in CSC commutation cells on switching losses. This facilitates a subsequent quantitative performance evaluation of an exemplary AC-AC CSC motor drive that employs the GaN M-BDS and operates from a 200 V three-phase mains. For a nominal motor power of 1.5 kW and a required AC-AC semiconductor efficiency of 98 %, a conventionally controlled CSC can operate with a switching frequency of 48 kHz, whereas the proposed synergetic control enables twice the switching frequency, i.e., 96 kHz and accordingly facilitates a significant volume reduction of the DC-link inductor and the input and output side filter capacitors.

Topics & Concepts

CommutationRectifier (neural networks)TransistorCapacitorMotor driveInverterInductorElectrical engineeringVoltageConvertersThree-phasePower semiconductor deviceCapacitanceMaterials scienceComputer scienceElectronic engineeringEngineeringPhysicsArtificial neural networkMachine learningMechanical engineeringStochastic neural networkQuantum mechanicsRecurrent neural networkElectrodeAdvanced DC-DC ConvertersMultilevel Inverters and ConvertersSilicon Carbide Semiconductor Technologies