Affiliations: [a] Department of Power Electrical Engineering, University
of Mons, Mons, Belgium
Correspondence:
[*]
Corresponding author: Zacharie De Grève, Department of Power Electrical Engineering, University of Mons, Bd Dolez, 31,
B-7000 Mons, Belgium. Tel.: +32 0 6537 4125; Fax: +32 0 6537 4120; E-mail: [email protected].
Abstract: This paper compares the power loss and weight of multiobjective
optimal (MO) designed isolated DC-DC power converters based on Si and SiC
technologies. To that end, a computer-aided design (CAD) tool previously
published by the authors and dedicated to the MO design of isolated DC-DC power
converters with genetic algorithms is used. The design problem requires the
minimization of the power loss and the weight of the converter while ensuring
the satisfaction of several constraints. The results show clearly that the wide
band gap (WBG) semiconductor devices, typically the SiC technology for the
power and voltage levels considered in this paper, lead to lighter converter
with less power loss than classical Si devices. The use of our tool allows to
precisely quantify that gain within a multiobjective optimization framework.
Keywords: Power electronics, computer-aided design tool, multiobjective optimization, genetic algorithms, SiC devices
