Novel Method to Speed-Up Power Module Optimisation using Geometrical Rule Checking of Electrical and Thermal Parameter

Full automatic power module design optimisation requires novel methods to reduce the time and resources required for highly intensive computational tasks. Tasks such as simulating electrical parasitics and temperatures limit the number of experiments in a given timeframe. With this, the design space cannot be fully explored to find an optimal design. The paper describes how new rules can be developed and applied based on geometrical relations to electrical and thermal parameters. These rules are used as pre-filters for the full automatic optimisation or as pre-checks to determine if a design template can be optimised. A power module with parallel Silicon Carbide metal oxide semiconductor field-effect transistors sharing a common source path was used for this. The geometrical loop matrix was extracted and analysed as an example of such a rule. This matrix enables the analysis of shared copper percentage and the design symmetry of the parallel switching devices. The paper's results significantly improve the optimisation speed by filtering infeasible designs from eight to four hours. The novel method described in this paper is adaptable to other designs, enhancing the overall efficiency of the power module design and verification process. (Abstract)

Authors

Institutions

Publication Details

Published
2024-11-18
DOI
https://doi.org/10.1109/dmc62632.2024.10812153
Primary Topic
Multilevel Inverters and Converters
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Novel Method to Speed-Up Power Module Optimisation using Geometrical Rule Checking of Electrical and Thermal Parameter

Andreas Schwarzbacher, Alperen Tunç, Gergő Juhász, P. Jakob et al.
Multilevel Inverters and Converters
article

Novel Method to Speed-Up Power Module Optimisation using Geometrical Rule Checking of Electrical and Thermal Parameter

Andreas Schwarzbacher, Alperen Tunç, Gergő Juhász, P. Jakob, Florian Bauer, Wilfried Wessel
article en

Abstract

Full automatic power module design optimisation requires novel methods to reduce the time and resources required for highly intensive computational tasks. Tasks such as simulating electrical parasitics and temperatures limit the number of experiments in a given timeframe. With this, the design space cannot be fully explored to find an optimal design. The paper describes how new rules can be developed and applied based on geometrical relations to electrical and thermal parameters. These rules are used as pre-filters for the full automatic optimisation or as pre-checks to determine if a design template can be optimised. A power module with parallel Silicon Carbide metal oxide semiconductor field-effect transistors sharing a common source path was used for this. The geometrical loop matrix was extracted and analysed as an example of such a rule. This matrix enables the analysis of shared copper percentage and the design symmetry of the parallel switching devices. The paper's results significantly improve the optimisation speed by filtering infeasible designs from eight to four hours. The novel method described in this paper is adaptable to other designs, enhancing the overall efficiency of the power module design and verification process. (Abstract)

Siemens (Germany) (DE), Technische Universität Darmstadt (DE), Technological University Dublin (IE)
Affordable and clean energy
Openalex Percentile: Top 77%
Multilevel Inverters and Converters
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.