PSI - Issue 52

Muhammad Raihan Firdaus et al. / Procedia Structural Integrity 52 (2024) 309–322 M.R. Firdaus et al. / Structural Integrity Procedia 00 (2023) 000–000

310

2

flooded, the float should be able to withstand the aircraft, which resulted in a high level of di ffi culty for an amphibious aircraft to sink. During water landing, the floats are the first amphibious aircraft structures in contact with the fluids. The interaction between the panels of the floats with the water caused a structural deformation Carcaterra et al. (1999); Yu et al. (2022). The structural response of the float will define its integrity due to the hydrodynamic impact that occurs, for instance, the peak hydrodynamic force, maximum structural displacement, and critical stress location Wang and Soares (2017). Several methods can be performed to obtain those structural response parameters, namely the experiment and the numerical modelling of hydrodynamic impact Hassoon et al. (2017). However, each of these hydrodynamic impact methods has its advantages and disadvantages. To model a full scale model of float for the experiment, su ffi cient amount of funds has to be available . Moreover, the facilities required to carry out hydrodynamic impact experiment quite rare to find and costly Judge et al. (2004); Stenius et al. (2013). For the numerical modelling of hydrodynamic impact, similar disadvantage on funding requirement, especially on computational resources, is also established Lu et al. (2000); Engle and Lewis (2003). In addition, completing a numerical analysis of hydrodynamic impact that includes fluid-structure interaction is a time-consuming task. In spite of that, the numerical modelling seems more feasible compared to the experiment, since numerical modelling is repeatable without having to redo experimental set up from zero, especially if new a new specimen is demanded dong Xu et al. (2009). Knowing that the modelling and simulation of hydrodynamic impact or fluid-structure interaction are very time-consuming and heavy in terms of computational resources in the numerical modelling, several strategies to model the numerical simulation on float structure were tried to be attempted. Those modelling strategies are the full shell, full solid, multi-stage multi-scale, and concurrent multi-scale modelling. Modelling complex structures require a very detailed process with huge numbers of elements and thus, resulting in high computational costs, in terms of computational time and resources Shankar et al. (2020). For a rather simpler structure, numerical computation can be computed using a macro-scale model. However, if within this model, there exists such regions where the deformation is very small, then a new e ff ective computation way is needed, where the scale is reduced to a smaller sample Curreli et al. (2018). The hybrid way in computing a structure consisting of a macro-scale and micro-scale mode is called as the multi-scale modelling Said et al. (2018). Mainly multi-scale modelling is aiming in increasing the computational e ffi ciency, while maintaining the accuracy of the simulation Arai et al. (2015). There are two common multi-scale modelling techniques, which are the concurrent multi-scale modelling and multi-stage multi-scale modelling, which is also called as submodelling [Narvydas et al. (2021)]. The submodelling is the technique utilized in the multi-stage multi-scale modelling. This is a technique used to study a local part of a model with a refined mesh based on interpolation of the solution from an initial global model. The usage of this technique is most useful when it is necessary to obtain an accurate and detailed solution in a local region and detailed modelling of that local region has negligible e ff ect on the overall solution Sun et al. (2019). The sub-modelling can use a combination of ABAQUS / Explicit and ABAQUS / Standard with combination of linear and non-linear procedures. Concurrent multi-scale modelling is a method in approaching such multi-scale cases by computing the macro scale and micro-scale exist within one global simultaneously Mishnaevsky et al. (2018); He et al. (2023, 2020). For instance, by implementing the shell-to-solid coupling in computing an aircraft float, as the skin part is modelled using shell elements, while the bulkheads and spars are being approached as solid elements. Moreover, in concurrent multi scale modelling, the quantities needed in the coarse scale model are computed on-the-fly from the fine scale models as the computation proceeds, while still providing information He et al. (2021); Ali and Shimoda (2022). A back-and forth transfer of information is concurrent and essential for the simulation of fracture and failure phenomena Niknafs et al. (2022). The numerical modelling is performed using ABAQUS computer-aided engineering software. Within the numerical simulation world, a considerably complex part is still doable. However, this may compensate the computational time and resources required to perform the analysis due to the need of high number of elements to produce more realistic and accurate results. In overcoming this, such method is introduced, called as the multi-scale modelling, where mainly it is divided into two di ff erent methods called as the multi-stage multi-scale modelling and concurrent multi-scale modelling. Further explanation about these two methods will be explained and delivered on the next sections. In further sections, the details of the modelling strategies, including the basics of fluid-structure interaction modelling in ABAQUS / CAE will also be discussed.