Reviewer, AIAA Journal of Thermophysics and Heat Transfer (2026–present).
D. A. Stewart, "Investigation of Surface Heat Exchangers for Jet Engines," DPhil Thesis, University of Oxford, to be submitted February 2027.
D. A. Stewart, J. D. Coull, and P. Ireland, "Transfer Learning for Sample-Efficient Bayesian Optimization Across a Real Distributional Shift in Experimentally-Validated CFD," in preparation for arXiv.
Tests whether a Bayesian-optimization surrogate warm-started on one CFD-validated flow regime needs fewer expensive simulations to adapt to a physically distinct one — comparing classical Gaussian-process and neural deep-kernel transfer mechanisms on a real experimental distributional shift, not a synthetic benchmark.
The AIAA AVIATION 2026 conference paper below has been submitted to the AIAA Journal of Thermophysics and Heat Transfer and is currently under review.
D. A. Stewart, J. D. Coull, and P. Ireland, "The Effect of Turbofan-Representative Distortion on Surface Heat Exchanger Performance," AIAA 2026-4433, presented at the AIAA AVIATION Forum, San Diego, CA, 2026.
Compares a finned and a plain surface heat exchanger installed in an annular sector rig under two inlet conditions — a simple uniform flow and an engine-representative distorted flow generated by an upstream outlet guide vane row — to determine how installation-representative inlet distortion affects aerodynamic loss and heat transfer.
D. A. Stewart et al., "Flow Mechanisms Governing Inlet-Distortion Effects on Surface Heat Exchanger Performance," planned for presentation at ASME Turbo Expo 2027, Calgary, Canada.
Validates the experimental dataset underpinning the AIAA AVIATION 2026 paper above and further analyzes the flow mechanisms responsible for the aerodynamic and thermal performance shift between the uniform and engine-representative distorted inlet conditions.
A journal version is planned for submission to the ASME Journal of Turbomachinery following the conference presentation.
D. A. Stewart et al., "Surrogate-Based Multi-Objective Optimisation of a Novel Surface Heat Exchanger Geometry," planned for presentation at ASME Turbo Expo 2027, Calgary, Canada.
Develops Gaussian Process surrogate models of aerodynamic loss and thermal effectiveness from a design-of-experiments campaign over a novel surface heat exchanger geometry, solved using a RANS CFD model validated against experimental data from the custom annular wind tunnel. The surrogates drive a multi-objective (NSGA-II-style) search to identify Pareto-optimal geometries, trading aerodynamic loss against thermal performance across representative installation flow conditions.
A journal version is planned for submission to the ASME Journal of Turbomachinery following the conference presentation.