Novel structural modelling strategies to enable early stage design of integrated airframe-propulsion systems

Abstract

Novel aircraft configurations exhibit coupling between airframe and propulsion system aerodynamic, flight-, and structural dynamics responses which exceed the capability of existing design processes. Physics-based (aeroelastic) analysis offers a flexible solution to this challenge; however, the representativeness of such approaches is dependent on constituent modelling strategies. Thus, the design of novel configurations necessitates appropriate modelling strategies to inform architectural design decisions. Such strategies must representatively and robustly acquire a suite of key metrics which capture: (a) novel airframe-propulsion system responses, and; (b) impacts of architectural decisions which are implicitly considered by empirical approaches.

A range of propulsion system structural modelling strategies have been employed in literature with limited consideration of their appropriateness – dependent on their ability to representatively idealise aeroelastic responses, mass and stiffness distributions. This thesis aims to identify representative, robust, and efficient propulsion system structural modelling strategies to assist the early-stage design of conventional and novel airframe-propulsion system configurations.

This research employs verified, whole-aircraft-level, aeroelastic models to capture pertinent propulsion system loads under manoeuvre, dynamic gust, and extreme event cases. The representativeness, robustness, and efficiency of a large suite of propulsion system structural modelling strategies – including novel strategies developed herein alongside those commonly employed in literature – is evaluated relative to a Global Finite Element Model.

Initial studies performed herein indicate low fidelity, concentrated mass propulsion system idealisations acquire traditional, early design stage Key Structural Performance Metrics within ±2% of a Global Finite Element Model. Mid-fidelity, 1D idealisations fail to capture key metrics representatively – condensation of connection stiffness to a single plane misrepresents structural responses to all loading. Higher fidelity, predominantly 2D shell idealisations are required to representatively acquire propulsion system lateral inertial loads, modal responses, sizing loads, and internal deformations.

Sensitivity analyses identify appropriate strategies for two case-study, novel airframe-propulsion system configurations by evaluating: (a) the robustness of effective (representative and efficient) modelling strategies, and; (b) significant design parameters for each Key Structural Performance Metric. The most effective modelling strategies for each metric are generally robust to architectural changes. While low fidelity strategies do not representatively capture propulsion system model responses, they robustly indicate changes in power plant swing eigenfrequency. However, it is necessary to model nacelle stiffness to robustly acquire relative airframe-propulsion system lateral displacement.
Key Structural Performance Metrics of Truss-Braced Wing configurations employing conventional and novel under-wing mounted propulsion systems are most sensitive to propulsion system location. Thus, low-fidelity, concentrated mass and higher fidelity, predominantly 2D modelling strategies – which are appropriate for conventional configurations – are likely to be appropriate for the acquisition of key metrics for Truss-Braced Wing configurations. A predominantly 2D shell propulsion system idealisation which explicitly models core casings, pylon, shaft/rotor, and primary nacelle component masses is recommended to robustly capture modal responses; lower fidelity strategies exacerbate coupling with airframe modes.

This thesis addresses key limitations in the modelling and analysis of integrated airframe-propulsion systems during early design stages. Appropriate (representative, robust, and efficient) propulsion system structural modelling strategies may be employed in novel aircraft design frameworks to assist architectural decisions. The overall approach employed in this research is sufficiently flexible to permit application to novel configurations beyond the scope of this research.

Date of Award	Jul 2023
Original language	English
Awarding Institution	Queen's University Belfast
Sponsors	Engineering and Physical Sciences Research Council
Supervisor	Damian Quinn (Supervisor) & Declan Nolan (Supervisor)