Topology optimization methods for heat sink design applied to power electronics

Topology is defined here to relate the geometric properties and spatial relations unaffected by the continuous change of shape or size of physical engineering artifacts. Researchers have used topology optimization methods to design next-generation structural components for a wide variety of applications (e.g., automotive, aerospace, industrial machines, etc.). This has enabled improved part performance, reduced component volume, and the elimination or reduction of subsystem assembly steps through the use of a free-form structural design representation for commercial products. The free-form capability of topology optimization methods has enabled realization of novel parts otherwise difficult to obtain through the optimization of parametric models. In this dissertation, topology optimization methods are applied to design heat spreading structures for next-generation power electronic systems. Wide band-gap semiconductor devices provide an opportunity to operate next-generation power electronics at higher temperature. The resulting increase in temperature capability is not shared by all components within a power electronic circuit, and careful consideration must be taken when routing heat throughout the system. Topology optimization methods provide a flexible and powerful solution to address challenges associated with these next generation electronic systems. First, a framework to classify general topology optimization problems is presented. This framework draws upon research from multiple application domains and defines a unifying language. Then, some practical heat conduction formulations for topology optimization are reviewed. This includes an analysis of common objective functions, in addition to some practical formulations for power electronics. Next, topology optimization methods are applied to design a heat spreader for a next-generation power inverter. The topologically-optimized design is fabricated and benchmarked against a baseline heat sink design. These studies motivated an investigation of reduced-order convection models for topology optimization, where experimental evaluation was used to assess heat spreader performance. The topology of the best-performing structure motivated a critical look at the structural representations used in topology optimization. For heat spreader design, it is common to use maximum approximation functions within a topology optimization. Several maximum approximation functions are analyzed in detail for use as optimization objectives and constraints. Advancing these topology optimization methods extends the envelope of design capabilities and therein lies the future of next generation power systems.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Systems & Entrepreneurial Engr

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2020

Author dc:creator

• Lohan, Danny John

Contributors dc:contributor

• Allison, James T.
• Tortorelli, Daniel
• Miljkovic, Nenad
• Wang, Pingfeng

• Topology Optimization
• Heat Spreader Design
• Power Electronics

Statement dc:rights

• Copyright 2019 Danny John Lohan

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/106231

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F106231