Objectives

Industrial applications where the work will be developed mainly CESMADYN surface ships (DGA, DCN, GTT ..) submarines (DGA, DCN, ..), offshore wind turbines (DCNS, ..) , lighthouses and beacons (CETMEF, ..) and offshore platforms (Saipem, SBM, TECHNIP). Two sub-themes can be extracted from this problem for marine structures, namely their behaviors impact and shock waves (heavy sea, slamming, collision, protection against shock waves aerial and submarine) on one hand and their dynamic behavior and overall vibration (maneuverability, seakeeping, Hydroelasticity, effect of appendages) on the other. The scientific issues associated with such research are simulation and experiment:

• the propagation of a shock wave in a biphasic medium (aerated water, liquid and solid foams, ..);
• free movement of a solid deformable or elastically interacting with water;
• of damage and failure of marine structures (metal, composites, foams, ..) are subjected to impact or shock waves;
• Hydrodynamic cyclic stresses and their consequences.

Finally the hard spots identified scientific concern in particular:

• experimentation and sensors in fast dynamic multi-physics (gas-liquid-solid),
• homogenization liquid-gas and solid-gas dynamic fast
• numerical models "simplified" with,
• unification maneuverability and seakeeping,
• slamming the linearized
• coupling slamming and whipping,
• the seakeeping in nonlinear (buoys and ships)
• numerical models "advanced" with
• the numerical simulation of free surfaces in nonlinear
• super-elements in fluid-deformable structures (anchors, ships)
• coupling dynamics of solid code with CFD codes,
• Finite Element code coupling with boundary element code;
• The criteria for damage and fracture in fast dynamic (metals, composites, foams) that take into account the state of the material after
• development,
• aging marine environment.

The declination of the research activities in the group CESMADYN reads in each sub-theme.

Overall dynamic behavior and vibration

Participants: Steven Kerampran, Jean-Marc Laurens, Jean-Baptiste Leroux, Kostya Roncin

The development of cavitation model and its validation pocket opens promising avenues of research. At first, it is to implement the model of cavitation in a Navier-Stokes solver. The idea is to resume the transpiration velocity method using the calculated steady potential. One can also use the same model by starting the scheme subcavitant calculated Navier-Stokes equations. The second method is certainly more expensive in resources, but will anyway much faster than conventional methods currently in development. Marino, the Netherlands, was very interested in the approach and proposes to work collaboratively. The company Vicusdt, Vigo, Spain, has also been very interested. For purposes of the preceding development, only pockets of partial cavitation were modeled. Steady supercavitating known that hydrodynamic forces on the body collapse and lose its function. From a hydrodynamic point of view the prediction of industrial efforts supercavitating regime has thus apparently of little interest. However, this regime exists when the propeller is overstretched. Then we run the risk of damaging the shaft at very short notice because the blade cavity in a high position when it does not cavitate at lower position due to hydrostatic pressure. This imbalance leads to a bending force on the shaft which is combined with compression already exerted by the propeller thrust. To assess the risk to the tree line it is a first step to extend the model of pocket to the pockets supercavitation then applying the resulting hydrodynamic forces in the boundary condition of a structure calculation involving the line tree and its supporters.

With a potential coupling between simulations and simulations by solving Euler equations, we obtained fluctuations of pressure on the leg of a tractor pod. The coupling is mainly used to avoid large numerical diffusion which manifests itself in codes and finite volume makes it impossible to predict the pressure fluctuations. These results were validated by comparison with experimental results obtained in Canada. However, our method does not place the propeller at a high angle of incidence as there appeared to detachment of the boundary layer profiles on the code potential is not able to simulate. One can imagine an Euler-Euler coupling that is not broadcasting digitally recovering downstream speeds of propulsion in that the impact is reinjected into a structured mesh of the pod. Another promising approach would be to work with ISAE solver who develops a method based on finite volumes sprectral not accusing digital broadcasting standard methods even on tetrahedral elements.

We are thinking for several years to a method of "universal" free movement simulation and / or forced an underwater object. It can be a weapon or output-cons far from a submarine, but the carrier itself or a Type AUV. The hydrodynamic forces are obtained by solving the Navier-Stokes and the movements are obtained by coupling with the equations of motion. We now have a calculation chain that has been verified by considering elementary configurations. We also run simulations on complex three-dimensional configurations, but the resources are so very hungry and need access to computer Ifremer. Only two-dimensional configurations are within the reach of our computer resources. A new means test (two-dimensional vein) will allow to validate the calculation procedure on many configurations, in parallel, it is planned to launch a series of three-dimensional simulations that we can confront pool testing. We are also interested in calculating the trajectory of gliders, AUVs which propel themselves through the ballast but above the lift of their wings. By not taking the viscosity of the condition of Joukowski and friction, unsteady potential solver would prove perfectly adequate and much faster and efficient than a RANS solver for these simulations.

Under the preliminary ship design software used must be fast and responsive to critical load cases. The tools available today are schematically the boundary element methods, linear three-dimensional, computer codes and RANS methods of slices. The last class is older method because less resource-intensive computing and faster to implement. It is, despite the growth of computing resources, a popular method and is still the subject of numerous scientific publications. The main advantage is that it can process nonlinearity specific movements of large amplitude. The nature of the proposed work is to propose a new method of slices covering the nonlinearities of the flow and coupled to the resolution of the SMP in the 6 degrees of freedom. This tool will help address such problems linking seakeeping and maneuverability as parametric rolling or the determination of hydrodynamic loads on the structure. It may be coupled to other tools developed in the laboratory aimed at solving the problem of slamming (slamming) three-dimensional.

We want to develop an experimental validation of models of ships in waves. Currently a level fine enough in determining the hydrodynamic loads when slamming is achieved provided that the geometry and impact velocity are well defined, which requires knowledge of the relative movements of the vessel from the water. The project involves the construction of an experimental system in the form of a sailing robot. The system will be equipped with sensors and measurement systems (central inertial, GPS, anemometer, wind vane) to know at every moment and save its state (position, attitude kinematics torsor, accelerating). The measures thus produced will feed the models developed in the laboratory with more realistic conditions of navigation.

Behavioral impact and shock waves

Participants: Michel Arrigoni, Nicolas Jacques, Christian Jochum, Steven Kerampran Alain Nemea, Mostapha Tarfaoui

Dynamic loads in complex environments:

The experimental study of slamming, supported by the DGA, continues with the development of a system for measuring the hydrodynamic pressure exerted on a hull during its severe impact on a surface of aerated water or not. The deformation of the structure and the total hydrodynamic force will also be measured. These tests will take place on the machine's shock laboratory. The cooling water will be controlled by a bubble generator in the basin. Confrontation test-calculations (ABAQUS codes "lab") will be established.

We want to perform the detailed design of experimental tests allowing, on simple geometric configurations, a representation of the phenomena of whipping (whipping) of a ship following a slam. For this, the software ABAQUS and a simulation module of slamming (codes "lab") will be used. The simulations will help define the shapes of hulls to be tested, sizing assemblies, and also monitor the adequacy of the proposed tests with the aim of validating a simulation module hydro-elastic developed jointly by the BV and ECM. Subsequently, the tests will be on the machine's shock LBMS.

The structural design of propulsion for military buildings requires consideration of dynamic loads resulting from the explosion of a charge near the ship. If the effects of an explosion on a submerged structure are well known, the development of numerical methods to simulate the interaction between the pressure wave resulting from the explosion and an elastic structure is a hot topic. In order to provide usable data in pre-design phase, DCNS / Propulsion develops computational methods coupled fluid / structure applied to study linear interactions Fluid-structure in the case of underwater explosion not in contact. The proposed work will aim to achieve the necessary developments to the generalization of the method of calculation to the case of 2D geometry of form and its extension to the case of 3D geometries, with different models. Also an extension in the context of geometric nonlinearities is considered.

In a significant number of cases, the shocks propagate through several environments governed by different equations of state. This is particularly the case of a shockwave generated aerial detonation of an explosive or a condensed shockwave underwater interacting with a bubble curtain. The objective is to develop a numerical prediction based on the formalism of finite volume, can be interfaced with commercial numerical codes. In this context, several numerical schemes will be tested and validated for a range of equations of state (Gas perfect, JWL, Mie-Grüneisen, ...). Physical means (shock tube, barrel Taylor interfaced with a compartment of water) will be established to obtain experimental data needed to validate the models implemented. This research will be jointly in partnership with CEA-DAM (attenuation of a shock wave in an aqueous foam) and the draft RESIBAD labeled by the competitiveness cluster Mer Bretagne (attenuation a shock wave underwater by a curtain of bubbles).

In the theme of propagation of shocks generated by pulsed power, exploratory research on industrial applications involving pulsed power are considered (Laser Shot Peening, Laser Adhesion Test). These methods involve the use of a pulsed laser for generation. Several partners have expressed interest (DCNS ITHPP, the Institute of Corrosion, Faculty of Dentistry of Brest). The theme would be addressed in collaboration with the MMA. The realization of these projects is contingent on the purchase of a laser pulse.

Damage and Fracture of Marine Structures:

Synthesizing work will be continued on the description of a broad spectrum of critical mechanical phenomena (precursors to the macroscopic scale of damage and / or breaks in the solid structures) via a formulation "unified" the bifurcation of Balance in Speed (BEV). It will address such conditions coplanar buckling of a straight beam elastic, the bifurcation, better known as the aspect instability of Taylor-Couette flow in a cylindrical viscous fluid (micro) cracking of a medium macroscopically brittle elastic, bifurcations Rice type environments (thermal) elasto (visco) plastic isotropic hardening, robustness criterion BEV implanted as failure criterion in the finite element code ABAQUS through its behavior refinement linkages and conditions of existence of a BEV in the macroscopic elastoplastic media with linear kinematic hardening (Prager).

The micromechanical modeling of dynamic damage ductile, made in collaboration with the University of Metz, should continue in the coming years. However, a new direction is given to developing a model to describe the problems of ductile tearing. This work was initiated this year. Problems involving ductile tearing mechanisms of nucleation and growth of cavities differ from those that come into play for scaling (discussed earlier), requiring a thorough review of modeling adopted. A key point is always the consideration of micro-inertial effects in the formulation. The first simulations performed suggest that these effects could play a crucial role in the dynamic propagation of ductile cracks.

At the macroscopic scale, the work on the direct identification of the behavior of composite materials elasticity coupled to damage will be expanded to cases of cyclic loading (repeated shocks, fatigue). In this context, the methods of optical extensometer will be linked to the technique of infrared thermography to identify the parameters governing the initiation and propagation of damage by impact and fatigue. This work involves two underlying developments very little discussed in the literature. The first concerns the application of infrared thermography for quantitative monitoring and characterization of the physical processes of damage by impact. The second involves the construction of thermo-mechanical models incorporating those mechanisms and kinetics of damage. This research will be applied to the use of composite materials in the field of marine renewable energy and more particularly in the context of offshore wind and tidal turbines.

Given the increase in intrinsic performance of the fiber reinforcement and formulations of the corresponding matrices, the use of thick composite (typically more than 5 mm thick) for structural purposes is becoming more common in marine applications and offshore. Nevertheless, the thermo activated and exothermic thermosetting composite cooking must be taken into account to identify couplings between thermal, chemical and mechanical training of the matrix. This expertise has shown that the strategy allowed a fairly good prediction of gradients of mechanical properties and the level of internal stresses induced by cooking. Besides the fact that the modeling of coupling can be further improved (taking into account the aging ocean, deepening the viscoelastic behavior of the matrix in training), it is essential to test the validity of the model implemented with dynamic tests on specimens and structures. This should help answer the question of design of composite structures under dynamic loads by providing a description of three-dimensional gradients of properties and internal stresses present in the composite thick after its manufacture.

It is envisaged to continue micromechanical modeling the dynamic behavior of composite three-dimensional microstructure in order to broaden the range of strain rates studied for different architectures (stitching, z-pins, 2.5D and 3D) and to introduce in the modeling inertial effects of phases of the material. This perspective requires the setting up of specific experimental methods (Hopkinson bar machine, shock ...). The objective of the work to be done is to understand the working mechanisms of the binding of composite preforms on to give the desk study relevant information to optimize the design of an assembly of preforms for their resistance to shocks. More efficient numerical models that take into account the different interaction process-material-structure will be developed to improve the modeling of the impact behavior and fatigue of composite assemblies in order to contribute to the knowledge of their tolerance damage.

For bonded structures, performance mode I and mode II are very different. The experimental results indicate that we have established that for the assemblies studied, the toughness increases with speed. These observations could be augmented by the benefits of experimental results by introducing the speed camera and infrared camera. Finally, it seems interesting to compare the variations in toughness with temperature and speed, changes in mechanical properties. This research should be used for new assembly technologies hybrid (bolted / bonded, sewn / bonded composite / aluminum, ...).