The increasing demand for greener aviation technology has driven the adoption of advanced composite materials in aircraft structures, offering significant weight saving and fuel efficiency improvements. Wing structures made of torsion boxes composed of stiffened panels have shown over the decades the benefits provided by composite technology, which is able to adapt the material properties to the structural constraints to which the structure is subjected. A convenient manufacturing technology for stiffened panels consists of co-bonding stiffeners on pre-cured skin. On these structures, the certification authorities require the demonstration of the residual strength at limit load of the panel with a disbonded stiffener. This is typically a post-buckling problem where the complete failure of the panel is due to a secondary buckling mode or the failure of adjacent stiffeners due to the combination of compressive and tearing loads, the larger buckled panel bay generating pull out loads on the adjacent stiffeners. This demonstration is classically performed by tests on large stiffened panels. In a composite wing box where geometrical parameters and loading modes can vary significantly from one zone to another, the numerical simulation can bring significant benefits to reduce the number of tests. This article presents a numerical damage model able to predict the damage and disbond of a co-bonded stiffener and applicable to a large aircraft structural model that can be integrated into a post-buckling simulation. As a validation case, this model has been applied to a multi stiffened composite panel where a central stiffener has been disbonded. The simulation results gave accurate predictions for the buckling loads and modes as well as for the appearance of damages on stiffeners until the panel failure.
In a context of growing importance of mass reduction and reliability of structures towards greener aircrafts, fatigue of metallic materials is a key issue in the structural optimization. The process used by aeronautic industrials to compute the fatigue life is often based on a large empirical experience and meets a need for efficiency in their application, requiring a compromise between accuracy and ease of use. According to legacy crack initiation methodologies, lifetime computation is based on the analysis of elastic stress fields, calculated analytically or by Finite Element Method. Evaluation of lifetime is calibrated on elementary tests, mainly uniaxial, with geometric specificity (bone, hole, notch…). One of the limits of this approach appears when parts are subjected to multi-axial loads. Nowadays, these particular stress states are justified by conservative approaches to ensure flight safety and by tests on full-scale aircrafts. Whether for the operational maintenance or the structural optimization of new aircrafts, it is intended to enhance crack initiation methodologies, taking into account multiaxiality of loads, stress gradient effects, and complex material behaviours. Dassault-Aviation implements a crack initiation lifetime computation based on a local approach. These developments go hand in hand with a PhD (Nutte, 2023) on a multiaxial fatigue criterion in order to predict crack initiation in metallic assemblies. This work was supported by an innovative dedicated test campaign. The identification of material’s parameters is based on uniaxial and multi-axial mechanical tests, specifically designed to calibrate these models. Then, novel geometry of specimen for bolted assemblies, facilitating various biaxial non-proportional loadings, is used to evaluate the methodology. Also, multiaxial fatigue models require a precise assessment of the local mechanical fields to which the structure is subjected. For this, a finite element analysis must be conducted with a level of complexity associated with the level of accuracy targeted. The material constitutive equations used in the finite element analysis are therefore at the heart of these fatigue substantiation approaches. Applications to complex aeronautical structures such as massive 3D parts or assembly by fastener will highlight the benefits and perspectives for this local fatigue approach. It will require the use of multi-scale data science.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.