Laboratoire de Mécanique de Normandie

This paper reviews the flow behavior and mathematical modeling of various metals and alloys at a wide range of temperatures and strain rates. Furthermore, it discusses the effects of strain rate and temperature on flow behavior. Johnson-Cook is a strong phenomenological model that has been used extensively for predictions of the flow behaviors of metals and alloys. It has been implemented in finite element software packages to optimize strain, strain rate, and temperature as well as to simulate real behaviors in severe conditions. Thus, this work will discuss and critically review the well-proven Johnson-Cook and modified Johnson-Cook-based models. The latest model modifications, along with their strengths and limitations, are introduced and compared. The coupling effect between flow parameters is also presented and discussed. The various methods and techniques used for the determination of model constants are highlighted and discussed. Finally, future research directions for the mathematical modeling of flow behavior are provided.

Abstract The purpose of the Reliability-Based Design Optimization (RBDO) is to find the best compromise between safety and cost. Therefore, several methods, such as the Hybrid Method (HM) and the Optimum Safety Factor (OSF) method, are developed to achieve this purpose. However, these methods have been applied only on static cases and some special dynamic ones. But, in real mechanical applications, structures are subject to random vibrations and these vibrations can cause a fatigue damage. So, in this paper, we propose an extension of these methods in the case of structures under random vibrations and then demonstrate their efficiency. Also, a Robust Hybrid Method (RHM) is then developed to overcome the difficulties of the classical one. A numerical application is then used to present the advantages of the modified hybrid method for treating problem of structures subject to random vibration considering fatigue damage.

This study aims at examining the impact behavior of hybrid carbon and glass fibers woven-ply reinforced PolyEther Ether Ketone (PEEK) thermoplastic quasi-isotropic laminates. An instrumented Charpy pendulum is specifically designed to estimate its capability to perform low velocity impact tests. Through the comparison of different impact methods (Quasi-static indentation tests, Charpy and drop tower impacts), the influence of impact velocity on the impact behavior of this hybrid composite material is investigated. From the obtained results, it appears that the macroscopic impact response is similar in terms of force-displacement response. Indeed, the impact velocity is significantly higher (2.5 times higher) with falling weight impact testing. In PEEK-based laminates whose mechanical behaviour is time-dependent, slow loading rates (e.g. Charpy impact testing) are instrumental in ruling the dissipated energy (+20% at 35 and 40J) as well as in increasing the permanent indentation (1.6 times higher) that is always higher than the Barely Visible Impact Damage.

The numerical simulation of multiphysics problems has grown steadily in recent years. This development is due to both the permanent increase of IT resources and the considerable progress made in modeling, mathematical and numerical analysis of many problems in fluid and solid mechanics. The phenomena related to fluid/structure mechanical coupling occurs in many industrial situations, and the influence it may have on the dynamic behavior of mechanical systems is often significant. In this paper, a numerical vibratory study is conducted on a three-dimensional aircraft’s wing subjected to aerodynamic loads. Finite volume method (FVM) is used for the discretization of the fluid problem, and finite element method (FEM) is used for the structure’s approximation. In this context, a deterministic model has been proposed in our study, then stochastic analysis has been developed to deal with the statistical nature of fluid–structure interaction parameters. Moreover, probabilistic-based reliability analysis intends to find safe and cost-effective projects.

The vibration of wind turbine towers is relevant to the reliability of the wind turbine structure and the quality of power production. It produces both ultimate loads and fatigue loads threatening structural safety. This paper aims to reduce vibration in wind turbine towers using an active damper named the twin rotor damper (TRD). A single degree of freedom (SDOF) oscillator with the TRD is used to approximate the response of wind turbines under a unidirectional gusty wind with loss of the electrical network. The coincidence between the wind gust and the grid loss is studied to involve the maximum loading on the structure. The performance of the proposed damping system under the maximum loading is then evaluated on the state-of-the-art wind turbine NREL 5 MW. The effectiveness of the TRD is compared to a passive tuned mass damper (TMD) designed with similar requirements. The numerical results reveal that, at the 1st natural mode, the TRD outperforms the passive TMD by three to six times. Moreover, the results show that the TRD is effective in reducing ultimate loads on wind turbine towers.

In the reliability-based design optimization (RBDO) model, the mean values of uncertain variables are usually applied as design variables, and the cost is optimized subject to prescribed probabilistic constraints as defined by a nonlinear mathematical programming problem. Therefore, an RBDO solution that reduces the structural weight in non-critical regions provides not only an improved design, but also a higher level of confidence in the design. Solving such nested optimization problems is extremely expensive for large-scale multidisciplinary systems that are likewise computationally intensive. This article focuses on the study of a particular problem representing the failure mode of structural vibration analysis. A new method is proposed, called safest point, that can efficiently give the reliability-based optimum solution under frequency constraints, and then several probability distributions are developed, which are mathematically nonlinear functions, for the proposed method. Finally, the efficiency of the extended approach is demonstrated for probability distributions such as log-normal and uniform distributions, and its applicability to the design of structures undergoing fluid–structure interaction phenomena, especially the design process of aeroelastic structures, is also demonstrated.

Many geotechnical problems involve pavements in Algerian arid regions; the main among them being related to the scarcity of standard materials. Recently, numerous studies were carried out on the valorization of Saharan local materials for pavement design. In this context, dune sand, tuff and quarry waste are promising alternatives. In this study, the design of experiments was used to optimize formulations to meet the current unconfined compressive strength (UCS) recommendations according to grading distribution. For the studied local material, the use of UCS criterion extended the valid domain of formulations to poorly graded materials where the amount of dune sand could reach 54.5%. An optimal formulation was proposed and tested, composed of 46% dune sand, 36% tuff, 11% quarry waste and 7% water, fulfilling all current recommendations related to the pavement design in the Sahara region. Finally, the results show that the design of experiments can be a useful tool to optimize mixture formulations and valorize local materials such as dune sand.

In the last few years, many studies have focused on the behaviour of unsaturated sands subjected to dynamic loading and there is consensus that soils with saturation degree lower than 100% can be liquefied. However, it is still necessary to have systematic studies for better understanding of this phenomenon. In this study, some experiments were carried out to survey the sand mechanical behaviour subjected to cyclic loading as well as the residual strength after liquefaction under monotonic loading. All the samples were prepared by the wet tamping method then the Skempton parameter B was measured to evaluate the saturation degree of the sample. After that, sample consolidation and cyclic loading were conducted step by step. All tests show that the soil liquefaction susceptibility goes up with the increase of the sample saturation degree. Through these tests, the relationship between cyclic stress ratio and Skempton’s parameter B is also highlighted. After liquefaction, the samples were loaded monotonically to assess the residual strength. The results show that the sand recuperates its strength when the pore water pressure dissipates after liquefaction.

High-cycle fatigue behaviour of structures can be obtained through vibratory tests using frequency response functions or transmissibilities. To this end, this study deals with the qualification of the vibratory fatigue bench developed at the Laboratory of Mechanic of Normandy. This bench uses an electrodynamic shaker which can reach excitation frequencies that are higher than conventional fatigue machines. To carry out relevant tests, the capacities of the test bench must be well known. Correlations between excitation and response were investigated to determine the allowable setpoints to maintain linearity between the input and output signals to validate our system. The difficulties related to the experiments were presented and discussed.

La mcatronique est une discipline qui combine entre la mcanique, l'lectronique et l'informatique. L'apparition des systmes mcatronique donne naissance des phnomnes de dfaillance et de dgradation qui se dveloppe avec le temps et qui ne sont pas bien maitrises. Pour tudier ces dfaillances on va utiliser la mthode des lments finis, est un schma (mthode) numrique qui permet de simuler (rsoudre) via l'outil informatique des problmes de la physique compliqus. Et ce, en approximant le modle mathmatique bas sur une quation aux drives partielles dont le nombre d'inconnus est infini par un modle algbrique matriciel dont le nombre d'inconnus est fini. La mise en application de cette mthode ce fait avec deux logiciels COMSOL et ANSYS, et les simulations vont nous permettre d'observer le comportement de notre composant ainsi dtecter l'origine des dfaillances. ABSTRACT. Mechatronics is a discipline that combines mechanics, electronics and computer science. The appearance of mechatronic systems gives rise to failure and degradation phenomena that develop over time and are not well controlled. To study these failures we will use the finite element method, is a numerical scheme (method) that allows to simulate (solve) complicated physics problems via the computer tool. This is made by approximating the mathematical model based on a partial differential equation whose number of unknowns is infinite with a matrix algebraic model whose number of unknowns is finite. The implementation of this method is made with two software programs COMSOL and ANSYS, and the simulations will allow us to observe the behavior of our component and detect the origin of the failures. MOTS-CLS.

Cette thèse vise à développer une approche pour une évaluation efficace et précise de la fiabilité du jacket des éoliennes offshore. Le jacket d'une éolienne offshore est soumis à diverses incertitudes. L’analyse de la fiabilité de la fondation du jacket est généralement effectuée en utilisant des approches d'approximation traditionnelles (par exemple FORM/SORM) ou la méthode des simulations de Monte Carlo. La modélisation mécanique du jacket d'une éolienne offshore nécessite généralement des modèles de simulations complexes et une analyse dynamique très couteuse en temps de calcul. L’utilisation des méthodes traditionnelles (FORM/SORM, simulations de Monte Carlo, etc.) pour l'analyse de fiabilité de ces structures peut être inadaptée. En effet, les méthodes d'approximation souffrent souvent de problèmes de convergence numérique surtout lorsque l'analyse dynamique des structures est impliquée, voire de précision lorsque le problème contient plusieurs points de défaillance. Les méthodes de simulations de Monte Carlo sont robustes, toutefois elles sont très coûteuses en temps de calcul et elles sont impraticables pour calculer des faibles probabilités de défaillance. La première partie de cette thèse vise la comparaison de trois approches de simulation des charges utilisées pour l'analyse dynamique des structures d'éoliennes offshore. Les approches étudiées sont respectivement la méthode non couplée, l'approche séquentielle et l'approche entièrement couplée. Ensuite, deux modèles numériques du jacket sont développés afin d'étudier l'influence des techniques de modélisation du jacket. Le premier modèle utilise des éléments de poutre de Timoshenko pour l'ensemble des éléments du jacket. Le deuxième modèle utilise une modélisation avancée à l'aide de la technique des super-éléments. Les éléments du jacket sont modélisés par des éléments de poutre et les assemblages entre ces éléments sont modélisés à l'aide des éléments de coque. Des comparaisons entre ces deux modèles sont également effectuées. La comparaison des approches de simulation des charges a montré que les résultats de l'approche séquentielle sont pour la plupart en bon accord avec ceux de l'approche entièrement couplée. L’approche non couplée peut conduire à des erreurs importantes dans les réponses extrêmes de l'analyse dynamique. En outre, pour la comparaison entre les deux modèles du jacket, nous constatons que les réponses du modèle du jacket à super-éléments sont différentes de celles du modèle de poutre, en particulier pour les déplacements du jacket. La deuxième partie de ce travail de thèse propose deux approches d'apprentissage actif pour l'évaluation de la fiabilité basée sur des modèles de substitution d'ensemble. Les modèles de krigeage (kriging) et les réseaux de neurones artificiels (ANN : Artificiel Neural Network) sont combinés pour constituer le modèle de substitution d'ensemble. L’efficacité et la précision des approches proposées sont dé­montrées par 4 exemples académiques et le modèle de poutre du jacket. Les résultats de l'analyse de fiabilité des exemples traités en utilisant les modèles de substitution d'ensemble avec les approches proposées d'apprentissage actif montrent l'efficacité et la robustesse de ces méthodes. D'ailleurs, même pour des problèmes de grande dimension et d'événement rare (probabilité de défaillance très très faible), ces approches montrent des performances numériques remarquables par rapport au modèle de substitution unique avec des approches d'apprentissage actif (par exemple AK-MCS). La dernière partie de cette thèse est dédiée à l’analyse de fiabilité système défini par plusieurs fonctions de performances. Une nouvelle fonction d’apprentissage composite est proposée pour le krigeage basé sur l’apprentissage actif avec la fonction U. Le modèle de krigeage à apprentissage actif avec la fonction d’apprentissage dite H est également adapté à l’analyse de fiabilité système...

Face aux exigences concurrentielles et économiques actuelles dans le secteur industriel, Les outils de simulation numérique, tels que les méthodes des éléments finis, sont de plus en plus largement (...)

The present paper aims at describing the use of a Synthetic-Eddy-Method (SEM), initially proposed by Jarrin et al. [12], in the 3D Lagrangian Vortex method framework. The SEM method is used here in order to generate a far-field incoming flow with a prescribed ambient turbulence intensity. However, for the account of the diffusive term in the Navier-Stokes equations, a classical Particle Strength Exchange model with a LES eddy viscosity is used.\nFirstly, the general characteristics of the Synthetic-Eddy-Method will be presented together with its integration in the framework of the developed 3D unsteady Lagrangian Vortex software [27]. The capability of the ambient turbulence model to reproduce a perturbed flow that verifies any turbulence intensity I∞ and any anisotropic ratio (σu :σv :σw ) will be discussed and validated. Then, the capability of the presented ambient turbulence model to compute turbine wakes will also be presented together with first results. Finally, comparisons will be made between the obtained numerical results against experimental data [22, 23] for two levels of ambient turbulence, namely I∞ = 3% and I∞ = 15%. Although the present study was initially performed in the framework of tidal energy, its application to wind energy is straightforward.\n

This paper aims at finding an optimal design of a shape memory alloy (SMA) actuator that guarantees a required reliability level under uncertainties parameters. The SMA actuator is composed of two NiTi membranes which, at room temperature, are in a martensitic state. They have an initial at shape and are bonded together with an intermediate spacer. The motion of the actuator is caused by the heating or cooling of its membranes. This paper proposes a methodology to find the best adjustment between cost and safety. Accordingly, several methods are developed, such as the hybrid method (HM) and the robust HM (RHM) to reach this purpose. The objective of this work is to present an extension of these methods in case of the SMA material. In fact, in this case, these methods may lead to an infeasible solution. Then, a method called improved RHM (IRHM) is proposed to overcome the difficulties of other ones. The obtained results showed the effectiveness and efficiency of the IRHM for treating the reliability-based design optimization problem of the SMA actuator.

Abstract This paper presents a lifting‐line implementation in the framework of a Lagrangian vortex particle method (LL‐VP). The novelty of the present implementation lies in the fluid particles properties definition and in the particles shedding process. In spite of mimicking a panel method, the LL‐VP needs some peculiar treatments described in the paper. The present implementation converges rapidly and efficiently during the shedding sub‐iteration process. This LL‐VP method shows good accuracy, even with moderate numbers of sections. Compared to its panel or vortex filaments counterparts, more frequently encountered in the literature, the present implementation inherently accounts for the diffusion term of the Navier‐Stokes equations, possibly with a turbulent viscosity model. Additionally, the present implementation can also account for more complex onset flows: upstream ambient turbulence and upstream turbine wakes. After validation on an analytical elliptic wing configuration, the model is tested on the Mexnext‐III wind turbine application, for three reduced velocities. Accurate results are obtained both on the analytical elliptic wing and on the New MEXICO rotor cases in comparison with other similar numerical models. A focus is made on the Mexnext‐III wake analysis. The numerical wake obtained with the present LL‐VP is close to other numerical and experimental results. Finally, a last configuration with three tidal turbines in interaction is considered based on an experimental campaign carried out at the IFREMER wave and current flume tank. Enhanced turbine‐wake interactions are highlighted, with favourable comparisons with the experiment. Hence, such turbine interactions in a farm are accessible with this LL‐VP implementation, be it wind or tidal energy field.

• Introduces a surrogate model for the production of γ-valerolactone (GVL). • Latin Hypercube Sampling method to include parameters uncertainty in the optimization. • A multi-objective optimization to maximize production rate and minimize thermal risk. • Pareto charts demonstrate the trade-offs between performance and safety. • Computational time reduced by 10 4 times compared to conventional kinetic modelling. In chemical process optimization, identifying conditions that balance production rate and thermal risks is crucial. This paper presents a surrogate-assisted optimization methodology that integrates parameters uncertainty, specifically focusing on synthesizing γ-valerolactone (GVL) in adiabatic and batch modes. A surrogate model was established to elucidate the relationships between input variables, production rate and risk index, which reduces the computational burden associated with complex differential equations. The Latin Hypercube Sampling method was employed to assess how uncertainties propagate through the processes. This study formulates a multi-objective optimization model that seeks to find a balance between the highest possible GVL production rate and the lowest probability of failure under deterministic and uncertain scenarios. The results in Pareto charts illustrate the possible operating conditions and determine the optimized initial conditions. This approach serves as a model for optimizing complex chemical processes, balancing production capacity and safety while considering uncertainty management.

An efficient strategy is presented in the recent study. This strategy leads to determine the best design based on thermal management system of phase change material (PCM) for cooling electronic devices. The phase change material is filled inside the heat sink made of Aluminum. Salt Hydrate is used as PCM to keep the temperature of such devices below the critical temperature by absorbing the thermal energy released by electronic components. In the last few decades, designers try to find the optimal design of such a structure. Then, many methods have been developed to reach this goal. Deterministic Design Optimization (DDO) used proposes to minimize an objective function based on physical constrain (thermal constrain). In fact, the maximum temperature of the electronic device should not reach the critical one, to guarantee its best performance. The proposed strategy is programmed using MATLAB codes which is coupled with the finite element software Ansys. Its efficiency is then verified using a numerical application of PCM based heat sink.

In order to assess wave impacts on coastal structures that are coupled with a marine energy device, for instance an oscillating water column (OWC), a smoothed particle hydrodynamics (SPH) software named JOSEPHINE (Cherfils et al., 2012) is used. In the present study, only a vertical wall will be considered as the front wall or draft of an OWC. In order to clearly identify impact phenomena, a breaking solitary wave will be used, so as to have a single phenomenon. And comparisons with experimental results issuing from Kimmoun et al. (2009) will be used as a matter of validation of our numerical study on solitary wave impacts. The present paper will focus first on the accuracy and convergence of wave propagation within the SPH framework, as a continuation of the work of Antuono et al. (2011), both for a regular wave train and for a solitary wave. For regular waves, the second-order dispersion relation is well recovered, up to the third order for the higher amplitudes. For solitary waves, comparisons with analytic and experimental results are also performed. Several types of impact are obtained similarly to those mentioned in the literature by changing the wave-maker parameters or the water depth in numerical wave flume. However, most of the effort was used for the validation of an impact case well documented in the literature. New experimental results issuing from the previous study of Kimmoun et al. were also used. Some intense and rapid impact phenomena are reproduced with our SPH single-phase numerical approach. The conclusion of this work is that a two-phase compressible approach is finally necessary to accurately compute such phenomena.

Pipelines play a crucial role in transporting essential resources such as water, oil, and gas across industrial, urban, and environmental infrastructures. Clogging remains a persistent challenge, potentially resulting in catastrophic failures, operational disruptions, increased maintenance costs, and serious safety risks. This study presents a novel prognostic and health monitoring approach that utilizes bubble-induced acoustic emissions to detect and characterize pipeline blockages. An analytical model is developed to capture the acoustic signatures of detaching bubbles, revealing features highly sensitive to clogging. Finite element simulations using Abaqus further investigate how different clogging conditions affect acoustic wave propagation. These insights drive the development of a machine learning-based predictive maintenance strategy, validated on real-world datasets. The results demonstrate exceptional accuracy, with most classifiers achieving 100% detection rates for clogging presence, shape, and severity. Additionally, model generalization tests show that machine learning algorithms adapt more effectively to varying clogging thickness than clogging shape. This research paves the way for a highly accurate, non-destructive monitoring solution, enhancing predictive maintenance and ensuring the reliability of industrial pipelines.

La présente étude porte sur le rendement de transmission dynamique d'un multiplicateur d'éolienne double étage en considération l'incertitude. Le travail consiste à considérer l'incertitude relative à la (...)