Back to search

Georgia Institute of Technology

Certification Oriented Design in the Context of Non-conventional Thermal Management System Architectures

Abstract

dc:description.abstract

The aviation industry is transitioning toward open-rotor and hybrid-electric propulsion to reduce emissions and noise, but these architectures impose new thermal management challenges because they alter (or remove) traditional cooling paths such as abundant bypass flow and nacelle-contained ventilation. In parallel, increased electrification introduces concentrated heat sources (power electronics, generators, motors, batteries) and new failure modes (over-temperature, thermal runaway propagation, and heat-soakback-driven excursions). These shifts raise safety risk and certification burden, particularly for Thermal Management Systems (TMS) distributed across the engine fan and core compartments and aircraft fuselage, where packaging constraints and harsh environments directly shape local heat transfer and thermal margin. This dissertation develops a certification-oriented, simulation-driven framework for early-stage analysis of non-conventional thermal-management system (TMS) architectures. The goal is to systematically leverage knowledge embedded in existing certified systems and physics-based models to (i) rapidly form realistic first-cut TMS architectures for novel propulsion concepts, (ii) identify which thermal limits and locations are most at risk across operating scenarios, and (iii) inform design iteration and test planning by identifying the dominant drivers of temperature margin. The work is positioned as an enabling set of steps toward future expanded use of modeling and simulation in certification contexts, rather than a replacement for physical testing or a stand-alone path to certification. The first objective is to connect legacy engine performance data with first-principles thermal-management principles to rapidly size and compare candidate TMS architectures for a notional open-rotor/hybridized propulsion system. A similarity-anchored decomposition and superposition process is used to identify operationally and physically similar certified engines from historical datasets, then decompose those systems into reusable component priors (pumps, pipes, tanks, heat exchangers, bays/compartments) that can be superimposed onto the novel architecture through updated boundary conditions (cycle states, mission attributes, electrification loads). Legacy-cycle reconstruction methods are used to create internally consistent thermodynamic cycles, which are then passed to a sizing and packaging workflow to define flow rates, station conditions, and geometric envelopes that constrain TMS placement and feasible heat-exchanger integration. The comparative analysis quantifies system-level penalties and benefits (mass, drag, fuel burn, and temperature margins) to test whether similarity-anchored sizing produces early TMS designs that are more feasible and less over-conservative than similarity-blind, equilibrium-only baselines. The second objective is to improve prediction fidelity for engine-compartment component temperatures using a mid-fidelity transient thermal network model that retains conceptual-design speed. A baseline lumped-capacitance network is augmented with geometry-informed airflow estimates and canonical convective correlations to better represent how ventilation and local flow features alter heat transfer around equipment, ducts, and structural surfaces. The model explicitly “pins” critical measurement locations and limit definitions (e.g., EGT measurement station, oil supply temperature at the lubrication/AGB interface, and fuel inlet temperature at the engine boundary) so that the analysis outputs map directly to the constraint logic used in design and operations. The result is a more physically grounded thermal-network representation that can be calibrated using limited higher-fidelity data and used to identify hotspots and time-dependent behaviors such as heat soak-back and reduced ventilation cases. The third objective is to evaluate robustness using scenario-based risk assessment and sensitivity analysis rather than relying on single-point performance claims. A structured set of extreme and off-nominal scenarios is defined (environmental hot-day conditions, ventilation degradation, reduced heat-sink capacity, and transient power bursts), and deterministic uncertainty bounds on key parameters (heat loads, heat exchanger effectiveness, ventilation flow coefficients, and boundary temperatures) are propagated through the coupled cycle–TMS–compartment model. Structured designs of experiments and sensitivity metrics are used to identify which parameters dominate temperature margin erosion, how architectural choices shift those sensitivities, and which combinations of conditions are most likely to expose thermal vulnerabilities. This provides a defensible way to prioritize design mitigations and to focus subsequent testing on the most informative and risk-relevant conditions. Overall, the dissertation contributes a coherent workflow that links (1) historical similarity and decomposition, (2) physics-based thermodynamic and heat-transfer modeling, and (3) scenario-based bounded-uncertainty evaluation to support early identification and mitigation of thermal risks in non-conventional propulsion-system TMS architectures. The framework is intended to reduce iteration risk and improve the traceability of thermal design decisions prior to certification testing, while acknowledging that certification remains a test-backed process and that analysis primarily serves to focus and strengthen that process rather than replace it.

Degree

thesis:*
Name thesis:degree_name
Aerospace Engineering, PhD
Department thesis:degree_department
Aerospace Engineering
Grantor
Georgia Institute of Technology
Year dc:date.issued
2026

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Balkas, Ezgi
Advisor dc:contributor.advisor
  • Mavris, Dimitri N.
Committee members dc:contributor.committeemember
  • Hendricks, Eric
  • Campbell, Angela
  • Gladin, Jonathan
  • Harrison, Evan

Rights

Language dc:language.iso
English

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/1853/82172

Chain of custody

source
Harvested from
Georgia Tech
Base URL
repository.gatech.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Balkas, Ezgi. Certification Oriented Design in the Context of Non-conventional Thermal Management System Architectures. Georgia Institute of Technology, 2026. https://hdl.handle.net/1853/82172