Skip to Main content Skip to Navigation
Theses

Adjoint-based aerostructural sensitivity analysis for wing design

Abstract : This work addresses the development of numerical methods for optimal wing shape design based in compressible ow. It presents the development and the validation of a discrete aero-structural adjoint method for sensitivity analysis with respect to both planform and internal structural design parameters which a ect the cost function, be it aerodynamic or structural. The developments performed in this work are an extension from the aeroelastic adjoint developments performed by Marcelet and are implemented in the CFD code elsA. While the aero-elastic adjoint method assumes that the wing sti ness is frozen, the here developed aero-structural adjoint method allows to consider structural variations. This extension is performed via a structural module, InAirSsi, developed during this work to model the wing box structure. This module is linearized to supply the necessary terms to the adjoint system. The structural module has been validated via the gradient-based sizing of the structural wing box of an Airbus research con- guration. The adjoint solver in the CFD code elsA covers now both planform variations and internal structural variations and allows the gradient computation of either an aerodynamic cost function or a structural cost function The gradient computed by the aero-structural adjoint method are validated systematically through a comparison with nite di erences.
Document type :
Theses
Complete list of metadatas

Cited literature [75 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00925210
Contributor : Imane Ghazlane <>
Submitted on : Tuesday, January 7, 2014 - 4:55:05 PM
Last modification on : Friday, June 26, 2020 - 2:02:05 PM
Long-term archiving on: : Monday, April 7, 2014 - 11:55:45 PM

Identifiers

  • HAL Id : tel-00925210, version 1

Collections

Citation

Imane Ghazlane. Adjoint-based aerostructural sensitivity analysis for wing design. Optimization and Control [math.OC]. Université Nice Sophia Antipolis, 2012. English. ⟨NNT : 01B0JJ0DEY6⟩. ⟨tel-00925210⟩

Share

Metrics

Record views

441

Files downloads

2398