Showing papers by "Mines ParisTech published in 2022"

Vertical Transmission and Neonatal Outcomes Following Maternal SARS-CoV-2 Infection During Pregnancy

Abstract: The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has infected over 200 million people worldwide and has likely exposed millions of neonates to SARS-CoV-2 in utero. A large body of literature has examined the possibility of vertical transmission from pregnant women infected with SARS-CoV-2 to their neonates. In this chapter, we review mechanisms of-and evidence for-vertical transmission of SARS-CoV-2, including transplacental, through other biospecimens and breastfeeding, and discuss neonatal outcomes following in utero exposure. Based on the available literature, we conclude vertical transmission of SARS-CoV-2 is rare, and exposed neonates generally show favorable health outcomes.

A review on deep reinforcement learning for fluid mechanics: An update

Abstract: In the past couple of years, the interest of the fluid mechanics community for deep reinforcement learning techniques has increased at fast pace, leading to a growing bibliography on the topic. Due to its ability to solve complex decision-making problems, deep reinforcement learning has especially emerged as a valuable tool to perform flow control, but recent publications also advertise the great potential for other applications, such as shape optimization or microfluidics. The present work proposes an exhaustive review of the existing literature and is a follow-up to our previous review on the topic. The contributions are regrouped by the domain of application and are compared together regarding algorithmic and technical choices, such as state selection, reward design, time granularity, and more. Based on these comparisons, general conclusions are drawn regarding the current state-of-the-art, and perspectives for future improvements are sketched.

Relationship between mechanical response and microscopic crack propagation behavior of hydrogen-related intergranular fracture in as-quenched martensitic steel

Abstract: The present study investigated the relationship between the mechanical response and microscopic crack propagation behavior of hydrogen-related intergranular fractures in high-strength martensitic steel. In contrast to cracks in the uncharged specimen, the cracks in the hydrogen-charged specimen propagated under a small crack opening displacement. Crack tip bluntings were frequently observed in the uncharged specimen, whereas no obvious blunting of the tip was observed in the hydrogen-related intergranular cracks. In addition, a high strain was localized around the hydrogen-related intergranular crack tip. The results indicate that strain localization can induce the formation of new quasi-cleavage cracks inside prior austenite grains ahead of the existing crack tip. The hydrogen-enhanced decohesion at prior austenite grain boundaries and the quasi-cleavage crack formation/propagation ahead of the arrested intergranular crack result in hydrogen-related crack propagation with a small crack opening displacement.

Coated single crystal superalloys: processing, characterization, and modeling of protective coatings

Abstract: Superalloy parts are typically employed under high-temperature conditions, often in hot gas paths of gas turbines for aviation or energy conversion. The thermal load of such parts is reduced by active back side cooling and additional application of thermal barrier coating systems. While the use of thermal barrier coating systems is industrial praxis, technological and scientific challenges are pushing the development of new coating systems and improved characterization and lifetime prediction methods. This chapter addresses some new developments which aim at allowing higher in-service temperatures and improving resistance to harsh environments. Further, innovation in characterization, testing, and modeling of protective high-temperature coatings is reviewed.

Policy-based optimization: single-step policy gradient method seen as an evolution strategy

Abstract: This research reports on the recent development of black-box optimization methods based on single-step deep reinforcement learning and their conceptual similarity to evolution strategy (ES) techniques. It formally introduces policy-based optimization (PBO), a policy-gradient-based optimization algorithm that relies on a policy network to describe the density function of its forthcoming evaluations, and uses covariance estimation to steer the policy improvement process in the right direction. The specifics of the PBO algorithm are detailed, and the connections to evolutionary strategies are discussed. Relevance is assessed by benchmarking PBO against classical ES techniques on analytic functions minimization problems, and by optimizing various parametric control laws intended for the Lorenz attractor and the classical cartpole problem. Given the scarce existing literature on the topic, this contribution definitely establishes PBO as a valid, versatile black-box optimization technique, and opens the way to multiple future improvements building on the inherent flexibility of the neural networks approach.

Crystal plasticity models: phenomenological approach

Abstract: This chapter introduces the constitutive equations tailored to single crystals that are inspired by classical macroscopic approaches, while incorporating the basic mechanisms of plastic deformation of the single crystal, that is, the presence of slip planes and systems. These models are a specific case in the class of constitutive equations with multiple potentials. Isotropic and kinematic hardening, viscosity, and more complex effect, like ageing, can be introduced in this framework. For the sake of simplicity, the presentation is limited to a classical formulation, keeping in mind that it must be simple enough to be used in structural calculations (turbine blades). First, the generic expressions are shown, then special forms of isotropic hardening (slip resistance and drag stress) and kinematic hardening (back stress) are proposed. Applications include yield surfaces and then parameter identification for two alloys, AM1 and CMSX4.

On the Security of OSIDH

Abstract: The Oriented Supersingular Isogeny Diffie–Hellman is a post-quantum key exchange scheme recently introduced by Colò and Kohel. It is based on the group action of an ideal class group of a quadratic imaginary order on a subset of supersingular elliptic curves, and in this sense it can be viewed as a generalization of the popular isogeny based key exchange CSIDH. From an algorithmic standpoint, however, OSIDH is quite different from CSIDH. In a sense, OSIDH uses class groups which are more structured than in CSIDH, creating a potential weakness that was already recognized by Colò and Kohel. To circumvent the weakness, they proposed an ingenious way to realize a key exchange by exchanging partial information on how the class group acts in the neighborhood of the public curves, and conjectured that this additional information would not impact security. In this work we revisit the security of OSIDH by presenting a new attack, building upon previous work of Onuki. Our attack has exponential complexity, but it practically breaks Colò and Kohel’s parameters unlike Onuki’s attack. We also discuss countermeasures to our attack, and analyze their impact on OSIDH, both from an efficiency and a functionality point of view.

A twin-decoder structure for incompressible laminar flow reconstruction with uncertainty estimation around 2D obstacles

Abstract: Over the past few years, deep learning methods have proved to be of great interest for the computational fluid dynamics community, especially when used as surrogate models, either for flow reconstruction, turbulence modeling, or for the prediction of aerodynamic coefficients. Overall exceptional levels of accuracy have been obtained but the robustness and reliability of the proposed approaches remain to be explored, particularly outside the confidence region defined by the training dataset. In this contribution, we present an autoencoder architecture with twin decoder for incompressible laminar flow reconstruction with uncertainty estimation around 2D obstacles. The proposed architecture is trained over a dataset composed of numerically-computed laminar flows around 12,000 random shapes, and naturally enforces a quasi-linear relation between a geometric reconstruction branch and the flow prediction decoder. Based on this feature, two uncertainty estimation processes are proposed, allowing either a binary decision (accept or reject prediction), or proposing a confidence interval along with the flow quantities prediction (u, v, p). Results over dataset samples as well as unseen shapes show a strong positive correlation of this reconstruction score to the mean-squared error of the flow prediction. Such approaches offer the possibility to warn the user of trained models when provided input shows too large deviation from the training data, making the produced surrogate model conservative for fast and reliable flow prediction.

Elementary deformation processes in high temperature plasticity of Ni- and Co-base single-crystal superalloys with γ/γ′ microstructures

Abstract: In the present chapter, we review elementary dislocation processes which govern high-temperature strength of Ni- and Co-base single crystals, focusing on yield and creep in different stress/temperature regimes. We show how our understanding has evolved starting with early results which were obtained for lower γ ′ volume fraction polycrystalline Ni-base alloys, where pairwise cutting was first documented. We then move on to the state-of-the-art single-crystal superalloys, with their high volume fraction of cuboidal γ ′ particles. We discuss elementary dislocation processes in the light of three boundary conditions which govern our systems: the need for dislocations to move through narrow γ channels, the possibility to cut into the ordered γ ′ phase, and the fact that dislocation plasticity is affected by misfit stresses, which result from the lattice misfit between the γ and γ ′ phases. Microstructural aspects are discussed with special emphasis placed on differences in deformation mechanisms in different stress–strain rate–temperature regimes. Both the nucleation and movement of 〈112〉 dislocation ribbons at low temperature and high stresses and the rafting of the γ ′ particles at high temperatures and low stresses receive attention. The role of planar fault energies in the γ ′ cutting is discussed and the role of key alloy elements receives special attention. Conclusions about alloy development are drawn.

A general framework for SPDE-based stationary random fields

Abstract: This paper presents theoretical advances in the application of the Stochastic Partial Differential Equation (SPDE) approach in geostatistics. We show a general approach to construct stationary models related to a wide class of linear SPDEs, with applications to spatio-temporal models having non-trivial properties. Within the framework of Generalized Random Fields, a criterion for existence and uniqueness of stationary solutions for this class of SPDEs is proposed and proven. Their covariance are then obtained through their spectral measure. We present a result relating the covariance of the solution in the case of a White Noise source term with the covariance in a generic case through convolution. Then, we obtain a variety of SPDE-based stationary random fields. In particular, well-known results regarding the Matern Model and Markovian models are recovered. A new relationship between the Stein model and a particular SPDE is obtained. New spatio-temporal models obtained from evolution SPDEs of arbitrary temporal derivative order are then obtained, for which properties of separability and symmetry can be controlled. We also obtain results concerning stationary solutions for physically inspired models, such as solutions to the heat equation, the advection-diffusion equation, some Langevin’s equations and the wave equation.

A Comparison of Eight Optimization Methods Applied to a Wind Farm Layout Optimization Problem

Abstract: Abstract. Selecting a wind farm layout optimization method is difficult. Comparisons between optimization methods in different papers can be uncertain due to the difficulty of exactly reproducing the objective function. Comparisons by just a few authors in one paper can be uncertain if the authors do not have experience using each algorithm. In this work we provide an algorithm comparison for a wind farm layout optimization case study between eight optimization methods applied, or directed, by researchers who developed those algorithms or who had other experience using them. We provided the objective function to each researcher to avoid ambiguity about relative performance due to a difference in objective function. While these comparisons are not perfect, we try to treat each algorithm more fairly by having researchers with experience using each algorithm apply each algorithm and by having a common objective function provided for analysis. The case study is from the IEA Wind Task 37, based on the Borssele III and IV wind farms with 81 turbines. Of particular interest in this case study is the presence of disconnected boundary regions and concave boundary features. The optimization methods studied represent a wide range of approaches, including gradient-free, gradient-based, and hybrid methods; discrete and continuous problem formulations; single-run and multi-start approaches; and mathematical and heuristic algorithms. We provide descriptions and references (where applicable) for each optimization method as well as lists of pros and cons to help readers determine an appropriate method for their use case. All the optimization methods perform similarly, with optimized wake loss values between 15.48 % and 15.70 % as compared to 17.28 % for the unoptimized provided layout. Each of the layouts found were different, but all layouts exhibited similar characteristics. Strong similarities across all the layouts include tightly packing wind turbines along the outer borders, loosely spacing turbines in the internal regions, and allocating similar numbers of turbines to each discrete boundary region. The best layout by AEP was found using a new sequential allocation method, discreet exploration-based optimization (DEBO). Based on the results in this study, it appears that using an optimization algorithm can significantly improve wind farm performance, but there are many optimization methods that can perform well on the wind farm layout optimization problem given that they are applied correctly.

Selfit - Accounting for Sexual Dimorphism in Personalized Motor Skills Learning.

Abstract: The development and the upkeep of psychomotor skills are fundamental for performing safely and efficiently daily life, professional, or leisure movements. Artificial intelligence offers the potential to develop adaptive systems that can support people's abilities to perform essential and specialized movements, by providing individualized and optimized learning sessions. Research on men is the building block of most theories and practices in the psychomotor development domain. However, it is assumed that women's inclusion will create potential interference due to their physiological variability (i.e., menstrual cycle and the effect of oral contraceptive). Thus, personalization of motor skills learning must consider significant differences between men's and women's physiology. Our study describes female-specific issues on psychomotor skills development (e.g., strength level, menstrual cycle, and female athlete triad). The sexual dimorphism-based individualization principles and parameters are afterward presented, together with their implementation in the Selfit intelligent tutoring system for psychomotor skills development. A risk injury monitoring module was implemented using computer vision technologies to track user movements and evaluate risk. A simulation with a virtual population of hundreds of men and women was performed, demonstrating that considering sexual dimorphism in the tutoring module is critical for injury prevention and for providing personalized sessions for women training.

Past, present, and future of SX superalloys

Abstract: The present chapter provides a short review of Ni-based superalloys developments, their introduction for the design of aero-engines components as well as improvements in their chemical compositions and processing routes. A specific focus is then paid to French developments of Ni-based single crystal superalloys systems involving public laboratories (Ecole des Mines de Paris, ONERA) and Safran Aircraft Engines (formerly Snecma). Finally, current trends in the use of Ni-based SX superalloys are presented, as well as some possible candidate materials to replace them for (ultra) high temperature applications.

Regional coupled surface-subsurface hydrological model fitting based on a spatially distributed minimalist reduction of frequency-domain discharge data

Abstract: Abstract. Although integrated water resources models are indispensable tools for water management at various scales, it is of primary importance to ensure their proper fitting on hydrological variables, avoiding flaws related to equifinality. An innovative step-wise fitting methodology is therefore proposed, which can be applied for any river basin model, from catchment to continental scale as far as hydrological models or land surface models are concerned. The methodology focuses on hydrosystems considering both surface water and groundwater, as well as internal water fluxes such as river baseflow. It is based on the thorough analysis of hydrological signal transformation by various components of a coupled surface--subsurface hydrosystem, in a nested approach, that considers the conditionality of parameter fields on their input forcing fluxes. The methodology is based on the decomposition of hydrological signal in the frequency domain with the HYMIT (HYdrological MInimalist Transfer function) method (Schuite et al., 2019). Parameters derived from HYMIT are used to fit the coupled surface–subsurface hydrological model CaWaQS3.03 using a step-wise methodology, which relies on successive Markov chain Monte Carlo optimizations, related to various objective functions representing the dependency of the hydrological parameters fields on forcing input fluxes. This new methodology enables significant progress to be made in terms of the spatial distribution of the model parameters and the water balance components, at the regional scale. The use of many control stations such as discharge gauging stations with HYMIT leads to a coarse parameter distribution that is then refined by the fitting of CaWaQS parameters on its own mesh. The step-wise methodology is exemplified with the Seine River basin (~76,000 km2). In particular, it made it possible to spatially identify fundamental hydrological values, such as rainfall partitioning into actual evapotranspiration, as well as runoff and aquifer recharge through its impluvium, in both the time and frequency domains. Such a fitted model facilitates the analysis of both the overall and detailed territorial functioning of the river basin, including explicitly the aquifer system. A reference piezometric map of the upmost free aquifer units and a water budget of the Seine basin are established, detailing all external and internal fluxes up to the exchanges between the eight simulated aquifer layers. The results showed that the overall contribution of the aquifer system to the river discharge of the river network in the Seine basin varies spatially within a wide range (5–96 %), with an overall contribution at the outlet of the basin of 67 %. The geological substratum greatly influences the contribution of groundwater to the river discharge.

Fast and Accurate Background Reconstruction Using Background Bootstrapping

Abstract: The goal of background reconstruction is to recover the background image of a scene from a sequence of frames showing this scene cluttered by various moving objects. This task is fundamental in image analysis, and is generally the first step before more advanced processing, but difficult because there is no formal definition of what should be considered as background or foreground and the results may be severely impacted by various challenges such as illumination changes, intermittent object motions, highly cluttered scenes, etc. We propose in this paper a new iterative algorithm for background reconstruction, where the current estimate of the background is used to guess which image pixels are background pixels and a new background estimation is performed using those pixels only. We then show that the proposed algorithm, which uses stochastic gradient descent for improved regularization, is more accurate than the state of the art on the challenging SBMnet dataset, especially for short videos with low frame rates, and is also fast, reaching an average of 52 fps on this dataset when parameterized for maximal accuracy using acceleration with a graphics processing unit (GPU) and a Python implementation.

Microstructure and chemical characterization

Abstract: This chapter is aimed at describing the microstructural features observed in single-crystal Ni-base superalloys, as they are essential to understand the mechanical properties, and the experimental tool implemented for each microstructural feature. The experimental tools that can be used to characterize both the microstructure and its defects are described from the larger scale to the smaller one. Most of the experimental techniques used in the case of single-crystal Ni-base superalloys are commonly used in other metallic alloys. They are well and widely documented in many textbooks or major publications. The intent of this chapter is not to develop every detail that can be found in the literature. It is rather an attempt to give a useful overview of the techniques, from the standard to innovative ones, in the case of single-crystal Ni-base superalloys.