Back to Search Start Over

Resolution Improvement in a High-Power Magnetic Resonance Imaging Gradient Power Amplifier.

Authors :
Zeng, Keqiu
Popovic, Jelena
Rietveld, Gert
Mao, Saijun
Yu, Hui
Wang, Liguo
Liu, Kun
Zhou, Zhiding
Source :
IEEE Transactions on Power Electronics. Sep2022, Vol. 37 Issue 9, p10594-10608. 15p.
Publication Year :
2022

Abstract

High-resolution gradient power amplifiers are required to generate high-fidelity gradient fields in magnetic resonance imaging (MRI) systems. Various aspects of gradient power amplifiers have been the subject of research in the past years, however, systematic analysis and design methods for its resolution improvement have not been sufficiently addressed. This article addresses this research gap with comprehensive analysis and design methods for resolution improvement. First, a method for systematic resolution characterization of the MRI gradient power amplifier with gradient coil is proposed. Second, noise modeling with respect to the critical aspects in the system resolution degradation is developed and discussed. Third, the resolution improvement methods of bandwidth optimization, pseudorandom code modulation, and utilizing silicon carbide (SiC) switching power devices are proposed and analyzed, achieving a resolution at the one part in ${10^6}$ level. Finally, the resolution improvement methods are experimentally validated in a 1000 V/400 A SiC gradient power amplifier. The experimental results show a 3.6 times reduction in the peak spurious signals and resolution improvement at all current levels and pulse lengths, compared to a conventional insulated gate bipolar transistor GPA. These results prove the validity of the proposed methods for resolution improvement in a high-power MRI gradient power amplifier, enabling future high-power and high-resolution amplifier designs. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
08858993
Volume :
37
Issue :
9
Database :
Academic Search Index
Journal :
IEEE Transactions on Power Electronics
Publication Type :
Academic Journal
Accession number :
157073258
Full Text :
https://doi.org/10.1109/TPEL.2022.3162385