Multi-objective optimization of dimpled sheets for high temperature loading conditions

This thesis investigates and optimizes aerothermal turbulator features which are designed with the added functionality of reducing stress under thermal loading conditions.The study is motivated by gas turbine components which are exposed to extreme loads and temperature leading to low service life or a reduction in engine efficiency.Previous work has shown that optimized porous void shapes can improve component life by orders of magnitude; however, non-porous components are also needed for applications which do not allow porosity.Recent findings showed a dimpled sheet structure which provides tunable mechanical behavior but which has not been optimized for peak stress or for aerothermal performance.This thesis investigates the mechanical and aerothermal behavior of a dimpled sheet, defined as a sheet with alternating indentations and protrusions of spherical caps in a square array, in high temperature loading conditions, such as those seen occurring in gas turbines.The dimpled sheet is first optimized independently for stress reduction in a single-objective problem formulation and then optimized for stress and thermal performance concurrently in a multi-objective optimization.Results of the single-objective stress optimization are presented and optimal regions in the design space are identified.The design space of thermal performance is also mapped.As expected, the optimal design regions for the thermal and stress objectives were not equivalent, highlighting the conflict between the two objectives and motivating the multi-objective study.The multi-objective study was carried out using surrogate models due to excessive costs to run physical simulations.The low cost of evaluation of the surrogate models enabled implementation of alternative schemes for multi-objective optimization.The results of each method are presented individually and collectively to assess the performance of each method and to construct the Pareto front showing the trade-off between stress reduction and improved heat transfer.While none of the selected schemes showed superiority over the others, the genetic and hybrid algorithms performed relatively better, while the gradient-based methods performed notably worst.The Pareto front representing the trade-off between stress and thermal performance was constructed by combining results from all algorithms and eliminating any dominated solutions.The Pareto front provides guidelines to select an optimal dimple geometry for a given gas turbine component based on the component requirements for stress and thermal performance.This thesis is dedicated to my sister, Brianna Delagran.She was my first and best teacher in the way only an older sister can be.In addition, I would like to acknowledge and thank the people who taught, mentored, and supported me leading up to, and throughout the creation of this thesis.To Professor Pasini, for teaching and guiding me in the research and academic worlds

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.