Gagliardi, Giuseppe Maurizio (2024) Multidisciplinary Analysis, Optimization and Testing techniques for Faster Aircraft Design and Certification. [Tesi di dottorato]

Documento PDF PhD_Dissertation_Gagliardi_GM.pdf Visibile a [TBR] Amministratori dell'archivio Download (40MB) | Richiedi una copia

Abstract

The aviation industry is currently undergoing a significant transformation, with an increasing emphasis on sustainability and reduction of carbon emissions. To meet the international goal of achieving net zero emissions by 2050, novel aircraft configurations are being explored and developed. These novel concepts investigate various technological pathways and aerodynamic configurations to support the decarbonization of the entire aviation sector. In light of these new requirements, a greater variety of aircraft designs will emerge in the coming years, necessitating that the aviation industry be adequately prepared to undertake their efficient design and certification. The methodologies and advanced tools proposed in this dissertation aim to significantly enhance and accelerate the Product Development Process (PDP) for aviation companies by reducing time-to-market and expediting certification cycles. The proposed analysis, optimization, and testing methods reduce the development time and costs for traditional and novel airframe configurations. Furthermore, the advanced flutter prediction techniques developed in this research contribute to obtaining safe, reliable, and competitive mobility of passengers, goods, and public services. The research addresses multidisciplinary challenges involving aeroelasticity, structural statics, vibration, buckling, impact dynamics, and aerodynamics. The capabilities and skills developed herein offer a streamlined approach to the design and certification of both innovative aircraft configurations and modified versions of existing aircraft. Significant advancements have been achieved in the area of Multidisciplinary Design Optimization (MDO) by expanding the applicability of analytical sensitivity methods. This dissertation introduces the Continuum Sensitivity Analysis (CSA) method, to compute the sensitivity of structural responses with respect to shape design parameters. This innovative approach allows for the derivation of accurate and efficient derivatives, which can be utilized by gradient-based optimization algorithms to enable efficient and accurate analysis and optimization across a wide range of design problems. Additionally, the sensitivity analysis method has been generalized to include composite materials for the first time. While CSA previously demonstrated excellent results, its applicability was limited to specific cases. This thesis extends the method to accommodate any material type, making it adaptable for real-world industrial applications and compatible with the use of commercial software in a black-box configuration. This dissertation further broadens the CSA method to compute eigensensitivities with respect to the shape. While the original CSA was limited to static problems or dynamic problems in the time domain, this work makes the method more robust to handle frequency domain problems. The method has been applied to beam problems, showing good agreement with reference results for both vibration and buckling analyses. However, eigenvalue problems present additional challenges in accurately estimating spatial gradients, particularly for buckling problems, which often require fine mesh discretization. To overcome this limitation, the dissertation introduces an innovative discrete analytical differentiation strategy. This method computes eigenvalue derivatives directly from the primary analysis results and the derivatives of the structural matrices, eliminating the need for spatial gradients. This approach enhances efficiency and accuracy, providing good results even with coarse mesh discretization. Significant advancements have been made in developing analytical sensitivity methods for flutter problems. Existing analytical aeroelastic shape sensitivity techniques are generally limited to either oversimplified aeroelastic models or detailed time-domain aeroelastic simulations. The former models are too simplistic to support effective shape design optimization. The latter require highly detailed structural and aerodynamic shape representations, which may not align with optimization objectives focused on primary shape parameters. This research addresses this gap by introducing a novel semi-analytical aeroelastic sensitivity method for computing flutter eigenvalue shape sensitivities. The proposed approach is well-suited to the aeroelastic models commonly used in industrial shape optimization, employing frequency domain representations and Doublet Lattice Method (DLM) aerodynamics. Moreover, the method allows both structural and aerodynamic shapes to be considered as design variables, enabling a more comprehensive aeroelastic shape design. Relevant advancements have also been made to enhance the flutter certification process. A novel method, called Master of Modes (MoM), has been developed to improve and facilitate flutter certification. The MoM approach provides a more advanced interpretation of flutter solutions, allowing for a detailed characterization of the behaviour of individual modes and effectively addressing long-standing challenges related to mode sorting. Additionally, the MoM methodology offers a new framework for effectively leveraging experimental data from Ground Vibration Testing (GVT) to enhance the accuracy of flutter calculations. Firstly, experimental modal data can be used to perform aerodynamic flutter calculations directly without the need for a proper structural model. In addition, they can be used to hybridize the structural model by selectively integrating experimental and numerical modal parameters. The hybridization of flutter analysis represents a complete scientific innovation, the MoM approach also provides significant practical benefits for the certification of both new aircraft configurations and modified versions of existing aircraft. The dissertation concludes with an extensive appendix section, presenting practical problems related to shape optimization, aircraft design, and certification. The first subsection demonstrates the application of the CSA approach to a shape optimization problem. Additionally, the vibration testing activities conducted to certify the Next Generation Civil Tiltrotor wing as part of the T-WING European project have been outlined. Furthermore, the Design of Experiments (DoE) approach has been applied to the parachute emergency landing certification test, ensuring a high level of confidence in test results before conducting costly experimental campaigns. Finally, the dissertation illustrates how leveraging commonality facilitated the development of a modified version of an existing aircraft with enhanced take-off and landing capabilities. This approach resulted in a rapid design and certification process, with the aircraft achieving both successful performance and rapid certification.

Downloads

Actions (login required)

Modifica documento