Abstract

Several megatrends of our society such as the electrification of transportation, the digitalization of our daily life, and the shift from fossil fuels to renewable energies are to a great extent fueled by the advancement of semiconductors. In order to appropriately address the needs of the different applications, the semiconductor portfolio of Infineon has expanded and diversified over the years with different technologies like Silicon and the wide-bandgap materials SiC and GaN in multiple different package options. As a result, selecting the right combination of power electronic topology, semiconductor technology, and control scheme has become more complex.
This talk focuses on how Infineon leverages the capabilities of multi-objective Pareto optimizations for exploring the trade-offs between efficiency, power-density and cost, in order to provide the best system solution for its customers. This enables Infineon to grow from a semiconductor supplier to a solution partner. Additionally, this talk also showcases how Pareto fronts can aid the development process of future semiconductors generations by identifying the impact of technological variations on the system performance.

Speaker

Matthias Kasper received the M.sc. and Dr.sc. degree in electrical engineering from the Swiss Federal Institute of Technology (ETH) Zurich, Switzerland, in 2011 and 2016, respectively. During his studies he focused on power electronics, mechatronics and energy systems. In his PhD at the Power Electronic Systems Laboratory at ETH Zurich he dealt with multi-cell converter systems for different applications, e.g. PV module integrated dc-dc converters, auxiliary supplies for high step-down ratios, and a highly efficient telecom rectifier module. Since January 2017 he is part of the Systems Innovations Lab of Infineon Technologies Austria AG where he works on novel topologies, control schemes, and multi-objective optimization routines. He has authored and co-authored more than 20 scientific publications and holds 12 international patents.