University of Illinois Urbana-Champaign

Optimal design of thermal systems using geometric projection and topology optimization

Bello, Waheed B.

Content Files
BELLO-THESIS-2022.pdf

Permalink

https://hdl.handle.net/2142/116291

Description

Title
Optimal design of thermal systems using geometric projection and topology optimization
Author(s)
Bello, Waheed B.
Issue Date
2022-07-22
Director of Research (if dissertation) or Advisor (if thesis)
James, Kai
Department of Study
Aerospace Engineering
Discipline
Aerospace Engineering
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
M.S.
Degree Level
Thesis
Keyword(s)
  • Geometric Projection
  • Topology Optimization
Abstract
The design of thermal systems in component and system level often relies on the expertise of engineers, which leads to sub-optimal designs. Topology optimization promisingly provides a systematic method for the optimal design of these systems. With an objective of extending the geometric projection method to thermal applications, we present an automated framework for the optimal design of thermal systems using a combination of topology optimization and geometric projection. On a system level, the layout of an Automotive Fuel Cell (AFC) system was optimized. This involved representing the system components as simple geometric shapes, which were then projected onto a fixed mesh using geometric projection. A finite element model was developed to model the thermal interaction between devices while a lumped parameter model was developed for the fluid flow within the routing bars. The framework was demonstrated on case studies in both 2D and 3D to reduce the overall bounding box volume of the system subject to performance constraints on temperature and pressure losses in the system. The packaging density was significantly improved as a 55% and 93% reduction in the bounding box volume was obtained for the 2D and 3D problems respectively. The component level study involves the optimal design of a cross-flow air-to-air heat exchanger. The configuration of air channels in the heat exchanger was optimized to improve the system performance. The channels are projected as circles onto a finite element mesh using geometric projection. The physics interaction of the heat exchanger was modelled using a 2D thermoelastic model. Three frameworks were explored for this problem. First, an accurate heat convection model, which requires information of the location and diameters of those channels, was combined with the geometric projection optimizer. Secondly, an approximate heat source model, which does not require any information outside of the density distribution, was combined with a density-based topology optimizer. Finally, a hybrid combination of the approximate heat source model and geometric projection method was developed. The system performance of the heat exchanger was improved by increasing the heat transfer rate by up to 13% while ensuring the von Mises stresses in the system remain below a specified limit.
Graduation Semester
2022-08
Type of Resource
Thesis
Handle URL
https://hdl.handle.net/2142/116291
Copyright and License Information
Copyright 2022 Waheed Bello

Owning Collections

Graduate Dissertations and Theses at Illinois PRIMARY
Graduate Theses and Dissertations at Illinois