Search

Mutual information (MI) has been previously used to select the relevant features for the task of speech emotion recognition (SER). However, the procedure does not deliver the optimal number of relevant features. We propose MI based criterion for estimating this number defined as the minimum number of features that explains the variable of the class indices. In order to minimize the MI estimation errors, we also search the best histogram binning choice considering three formulas: Sturges, Scott and LMSE. Four selection strategies MMI, CMI, JMI and TMI have been implemented and applied on 39-features vectors and on large dimension vectors. The feature selection results have been validated on independent text SER system, based on GMM classifier and evaluated on EMO-db database. Results demonstrate that LMSE bin choice gives the best MI estimation and ensures a minimal number of features with slight performance drop. Particularly, using the proposed stopping criterion, the CMI strategy achieves reduction of 48.72% in the case of the 39-features vectors size and 67.86% in the case of large dimension vectors. Moreover, using the recognition rate criterion, the JMI strategy gives a comparable feature reduction with slight improvement of performance but requiring very high computation capabilities.

This paper addresses the problem of blind stereoscopic image quality assessment (NR-SIQA) using a new multi-task deep learning based-method. In the field of stereoscopic vision, the information is fairly distributed between the left and right views as well as the binocular phenomenon. In this work, we propose to integrate these characteristics to estimate the quality of stereoscopic images without reference through a convolutional neural network. Our method is based on two main tasks: the first task predicts naturalness analysis based features adapted to stereo images, while the second task predicts the quality of such images. The former, so-called auxiliary task, aims to find more robust and relevant features to improve the quality prediction. To do this, we compute naturalness-based features using a Natural Scene Statistics (NSS) model in the complex wavelet domain. It allows to capture the statistical dependency between pairs of the stereoscopic images. Experiments are conducted on the well known LIVE PHASE I and LIVE PHASE II databases. The results obtained show the relevance of our method when comparing with those of the state-of-the-art. Our code is available online on https://github.com/Bourbia-Salima/multitask-cnn-nrsiqa_2021.

International audience

The objective of the present work was to experimentally investigate flow field instantaneous characteristics and their associated heat transfer for pulsating flows representative of engine exhaust flow operating conditions. The experimental apparatus consists of a stationary turbulent hot air flow, with a variable Reynolds number, excited through a pulsating mechanism and exchanging thermal energy with a water cooled steel pipe. Pulsation frequency ranges from 10 to 95 Hz. Simultaneous time-resolved measurements of velocity and temperature were achieved by mean of coupled hot-wire anemometry and micro-unsheathed thermocouples measurements to investigate the impact of the pulsation frequency on heat transfers. Flow pulsation was found to enhance heat transfers in the entire range of frequencies. Significant improvements were observed when the flow was excited with a frequency equal to a resonant mode of the system. With the increase in velocity-amplitude ratio, an increase in the convective heat transfer was observed, with a maximum enhancement corresponding to a factor three and higher, with respect to the relative Nusselt number. Combining experimental results with an analytical formulation, derived from the 1D energy balance equation for a turbulent pulsating flow, the velocity amplitude ratio was identified as the characteristic term representative of the predominant heat-transfer enhancement mechanism.

In the field of fingerprint identification, local histograms coding is one of the most popular techniques used for fingerprint representation, due to its simplicity. This technique is based on the concatenation of the local histograms resulting in a high dimension histogram, which causes two problems. First, long computing time and big memory capacities are required with databases growing. Second, the recognition rate may be degraded due to the curse of dimensionality phenomenon. In order to resolve these problems, we propose to reduce the dimensionality of histograms by choosing only the pertinent bins from them using a feature selection approach based on the mutual information computation. For fingerprint features extraction we use four descriptors: Local Binary Patterns (LBP), Histogram of Gradients (HoG), Local Phase Quantization (LPQ) and Binarized Statistical Image Features (BSIF). As mutual information based selection methods, we use four strategies: Maximization of Mutual Information (MIFS), minimum Redundancy and Maximal Relevance (mRMR), Conditional Info max Feature Extraction (CIFE) and Joint Mutual Information (JMI). We compare results in terms of recognition rates and number of selected features for the investigated descriptors and selection strategies. Our results are conducted on the four FVC 2002 datasets which present different image qualities. We show that the combination of mRMR or CIFE feature selection methods with HoG features gives the best results. We also show that the selection of useful fingerprint features can surely improve the recognition rate and reduce the complexity of the system in terms of computation cost. The feature selection algorithms may reach 98% of time reduction by considering only 20% of the total number of features while also improving the recognition rate of about 2% by avoiding the curse of dimensionality phenomena.

International audience; This paper provides multi-level modeling of microcracked viscoelastic masonries generally present in historical masonries or refractory linings. It is an extension of the Cecchi and Tralli (2012) or Cecchi & Barbieri (2008) and Cecchi & Taliercio (2013) works for the typical Burgers and Modified Maxwell models followed by both undamaged and microcracked masonries. For the sake of simplicity and in order to provide rigorous analytical global estimates, only the mortar is assumed to be viscoelastic and microcracked. Bricks are assumed to be undamaged, elastic or quasi-rigid. The distribution of microcracks is assumed to be isotropic. The effective behavior of the viscoelastic microcracked masonry is provided by two steps. The first one relies on the coupling between the Griffith’s brittle fracture theory and linear mean-field homogenization scheme in order to account for the effect of microcracks on the macroscopic deformation of the mortar and establishes a linear relation between apparent macroscopic stress and strain. This step allows to easily and fast determine the effective creep function of the microcracked mortar without recourse to ’complex’ or heavy numerical inversion of the Laplace–Carson transform. The second step is based on the coupling between asymptotic analysis and homogenization theory applied for a periodic masonry. The proposed models provide analytical solutions - explicit functions of the crack density parameter - for the effective orthotropic behavior of a microcracked viscoelastic periodic masonry cell. This study proves that the Cecchi & Tralli’s and Cecchi & Taliercio’s extension estimates are close and that the later are softer. Such overall properties are used to perform finite element computations on a compressed masonry panel as a first application. These models allow then the prediction of mostly stressed and deformed areas in microcracked masonry structures. This study demonstrates that modeling a mortar (at its undamaged or microcracked state) with this Burgers formulation is only suitable for a masonry with too high values of the Maxwell’s relaxation time otherwise it yields to vanishing effective properties with the increase of time and crack density leading thus to a premature collapse of the masonry. On the other hand, the Modified Maxwell model permits the masonry to preserve a certain resistance for every range value of Maxwell’s relaxation time. These conclusions are valid for masonry either with elastic or quasi-rigid undamaged bricks.

The fourth industrial revolution is obviously well underway. Every day, we have the demonstration through new modes of commercial and industrial organizations. We are faced with major breaks as well related to the evolution of technologies as to the globalization of the economy with the emergence and the maturity of new actors: the challenge of the energy transition, the digital revolution, the protection of our planet, the convergences between human sciences and sciences. This paper presents different applications implemented for “Internet of Things” with Profinet network. These applications can be used on all of the company's processes as part of the Industry of the Future. The OPC standard appeared in the mid-1990s to facilitate exchanges between the world of automation and that of PC-based supervision. The latest OPC UA (Unified Architecture) specifications are now validated. With this new version, the OPC Foundation is revolutionizing its standard of communication between industrial equipment. OPC UA breaks the links that made it inseparable from Microsoft Windows. It can be deployed on all types of platforms. A Programmable Logic Controller (PLC) manages a manufacturing process. Seven operator panels also manage the various production recipes. Client n°1 makes the link to other Clients (reporting, E-mail, opening up to a scientific platform). The results are presented and detailed for numerous opportunities for industrial developments.

International audience

International audience; This study focuses on the actual exhaust emissions of regulated and unregulated pollutantsfrom earthwork machines (EM). In addition to providing a realistic assessment of gaseousand particulate emissions from key EM and filling critical data gaps in emission inventories,it emphasizes the importance of breaking down the average operating emission factors intotask-related emission factors (TEF). The latter serve many goals, namely by: categorizingEM in relation to their emission dynamics, finding emission determinants and proposingon-site emission mitigation strategies. These have also led to a more refined analysis ofboth the effectiveness of EU and US emissions standards and the accuracy ofNONROAD2008 model. Ultimately, by using task-related emission factors in conjunctionwith data available from the literature and with the methodology developed by US-EPA,a lower bound was estimated for gaseous exhaust emissions from the major constructionvehicles operable at earthwork sites.

International audience; When an explosion occurs in a tunnel, the study of the blast wave quickly becomes complicated, owing to the multiple propagation patterns of the blast wave (Incident wave, regular and Mach reections) and to the geometrical conditions. Considering this problem, two patterns can be revealed. Near the explosive, the well known free-eld pressure wave can be observed. After multiple reections on the tunnel's walls, this overpressure behaves like a one-dimensional (1D) wave. One aim of this paper is to determine the position of this transition spherical-to-planar wave propagation in a tunnel using both numerical and reduced-scale experiments, and thereby validate the dedicated law established in a previous work. For this purpose, a detonation of TNT in a tunnel with a cross-section of up to 55 m 2 is considered. Results show good agreement between the numerical simulations and experiments. The transition zone between the three-dimensional (3D) and the 1D wave is well detected. An application to a simplied subway station is also investigated which shows that signicant planar waves can be transmitted to the neighboring stations via the junction tunnels.

International audience; In order to support the optimization of mortarless brick linings of steel producing furnaces it is proposed in this paper to investigate only the compressibility of dry refractory joints. To this end, optical measurements based on the Digital Image Correlation method were carried out during compression of Magnesia-Carbon brick samples with dry joints. The second main objective is to assess the accuracy of the secant linearization schemes (the classical secant procedure and its modified extension) to reproduce the reference local and global behaviour of refractory mortarless linings accounting for the identified inelastic convex powerlaw behaviour of the Magnesia-Carbon dry joint. The reference nonlinear solution is obtained by means of finite elements method. Under normal compressive loading, unlike for usual (concave) power-law viscoplastic composites for which the secant schemes are known to provide too stiff results, it was found that the modified secant scheme leads to good overall predictions. The classical secant procedure underestimates the reference local and overall behaviour. To improve the latter result, an empirical improved secant formulation was proposed and implemented. It leads to better estimates at local and global levels.

IntroductionSurgery is the standard curative treatment for lung cancer but is only possible in patients with local tumour and preserved exercise capacity. Improving fitness before surgery can reduce postoperative complications and mortality. However, preoperative rehabilitation remains difficult to implement for several reasons. We aim to investigate the effectiveness of an intensive 3-week home-based preoperative exercise training programme on hospital discharge ability, postoperative complications and physical performance in patients with chronic obstructive pulmonary disease (COPD) who are eligible for lung cancer surgery.Methods and analysisWe designed a multicentre randomised controlled trial. The randomisation sequence will be generated and managed electronically by a research manager independent of assessments or interventions. We will recruit 90 patients with COPD and a diagnosis of lung cancer from four university hospitals. The rehabilitation group (R group) will receive a standardised preoperative home exercise programme for 3 weeks, combining both high-intensity training and usual physical therapy. The R group will perform 15 training sessions over 3 weeks on a cycloergometer. A physical therapist experienced in pulmonary rehabilitation will visit the patient at home and supervise one session a week. The R group will be compared with a control group receiving preoperative usual physical therapy only. The primary outcome will be hospital discharge ability assessed with a 10-item list. Secondary outcomes will be postoperative course (complication rate and mortality) as well as pulmonary function, exercise capacity and quality of life assessed 1 month before and the day before surgery.Ethics and disseminationThis protocol has been approved by the French health authority for research (2016-A00622-49) and the research ethics committee/institutional review board (AU1267). Adverse events that occur during the protocol will be reported to the principal investigator. The results will be published in an international peer-reviewed journal.Trial registration numberNCT03020251.

To reduce the levelized cost of energy, the energy production, robustness and lifespan of horizontal axis wind turbines (HAWTs) have to be improved to ensure optimal energy production and operational availability during periods longer than 15–20 years. HAWTs are subject to unsteady wind loads that generate combinations of unsteady mechanical loads with characteristic time scales from seconds to minutes. This can be reduced by controlling the aerodynamic performance of the wind turbine rotors in real time to compensate the overloads. Mitigating load fluctuations and optimizing the aerodynamic performance at higher time scales need the development of fast-response active flow control (AFC) strategies located as close as possible to the torque generation, i.e., directly on the blades. The most conventional actuators currently used in HAWTs are mechanical flaps/tabs (similar to aeronautical accessories), but some more innovative concepts based on fluidic and plasma actuators are very promising since they are devoid of mechanical parts, have a fast response and can be driven in unsteady modes to influence natural instabilities of the flow. In this context, the present paper aims at giving a state-of-the-art review of current research in wind turbine-oriented flow control strategies applied at the blade scale. It provides an overview of research conducted in the last decade dealing with the actuators and devices devoted to developing AFC on rotor blades, focusing on the flow phenomena that they cause and that can lead to aerodynamic load increase or decrease. After providing some general background on wind turbine blade aerodynamics and on the atmospheric flows in which HAWTs operate, the review focuses on flow separation control and circulation control mainly through experimental investigations. It is followed by a discussion about the overall limitations of current studies in the wind energy context, with a focus on a few studies that attempt to provide a global efficiency assessment and wind energy-oriented energy balance.

In a robotized ultrasound scanning system and more generally in any system which interacts with a human, the used control laws have to satisfy a safety of operating and functioning. Moreover these control laws have to guarantee a high precision in the medical gesture tracking, and especially within the framework of the ultrasound scanning, they assure the preservation of the ultrasound plane. The mechanical architecture of the designed robot takes into account the specifications of the medical application of the remote ultrasound scan. This robot presents singular configurations on its work space. These configurations perturb the follow-up of the medical remote gesture and lead to a loss of precision which induce errors in the follow-up of trajectory as well as in the ultrasonic plan. To solve this problem we set up specific control laws which take into account these singular configurations and allow to reproduce the movements generated by the medical expert with a great accuracy. The first part of this paper presents the geometric and kinematic modelling of the Protech robot used for the remote ultrasound and an analysis of its work space and its singular configurations. The second part describes the various approaches used for the management of these singular configurations. The third and last part presents the simulations of these approaches and the obtained results.

International audience; In this paper, the problem of eco-driving for a conventional vehicle equipped with an internal combustion engine is studied. The associated optimal control problem is formulated and solved using Dynamic Programming (DP). The impact of the mesh choice on the optimality of the DP solution is investigated in order to find a trade-off between the optimality of the DP solution and its computation time. The eco-driving speed trajectories obtained were tested on a high-frequency HIL (Hardware-In-the-Loop) engine test bench to quantify the real reductions in fuel consumption. Simulations and experiments are compared.

International audience; In this paper, a first practical extension of the Equivalent Consumption Minimization Strategy (ECMS) is proposed to include thermal dynamics (engine and catalyst temperatures) in the optimal design of an Energy Management System (EMS) for a parallel Hybrid-Electric light-duty Vehicle (HEV). The task of this novel multi-state ECMS is to achieve a sufficient level of performance with respect to pollutant emissions while keeping fuel consumption within acceptable limits. The extension suggested here is based on correlations between the thermal state and their corresponding adjoint states, observed along extremal calculated from extensive offline solutions of optimal control problems. Simulation results stress that the obtained performance is sufficient to satisfy the environmental norms while keeping fuel consumption sub-optimality relatively marginal.

ASME Turbo Expo 2017: Turbomachinery Technical Conference and ExpositionVolume 4A: Combustion, Fuels and EmissionsCharlotte, North Carolina, USA, June 26–30, 2017Conference Sponsors: International Gas Turbine InstituteISBN: 978-0-7918-5084-8; International audience; A solution for CCS (Carbon Dioxide Capture and Sequestration of CO2) is oxycombustion. Due to the high cost of pure O2 production, however, other approaches recently emerged such as post-combustion coupled with Oxygen Enhanced Air (OEA). This is the solution studied in this paper, which presents an innovative gas turbine cycle, the Oxygen Enriched Air Steam Injection Gas Turbine Cycle (OEASTIG). The OEASTIG cycle is composed of Methane combustion with OEA (Oxygen Enhanced Air), EGR (Exhaust Gas Recirculation) and H2O coming from a STIG (Steam Injection Gas Turbine). CO2 capture is achieved by a membrane separator. The final aim of this work is to predict NO and CO emissions in the gas turbine by experimental and numerical approaches. Before carrying out this study, the validation of a reaction mechanism is mandatory. Moreover, this new gas turbine cycle impacts on the combustion zone and it is therefore necessary to understand the consequences of H2O and CO2 dilution on combustion parameters. While a large number of papers deal with CO2 dilution, only a few papers have investigated the impact of water dilution on methane combustion. A study of the influence of H2O dilution on the combustion parameters by experimental and numerical approaches was therefore carried out and is reported in the present paper. The paper is divided in three parts: i) description of the innovative gas turbine (OEASTIG) cycle and determination of the reactive mixtures compatible with its operation; ii) validation of the reaction mechanism by comparing laminar methane flame velocity measurements performed in a stainless steel spherical combustion chamber with calculations carried out in a freely propagating flame using the Chemical Workbench v.4.1. Package in conjunction with the GRIMech3.0 reaction mechanism; iii) Extrapolation to gas turbine conditions by prediction of flame velocities and determination of the feasible conditions from a gas turbine point of view (flame stability). In particular, mixtures (composed of CH4/O2/N2/H2O or CO2) leading to the same adiabatic temperature were investigated. Lastly, the influence of oxygen enrichment and H2O dilution (compared to CO2 dilution) were investigated.

International audience; The paper addresses the 2D observer-based control of a magnetic microrobot navigating in a cylindrical blood vessel along a reference trajectory. In particular, this robot faces the nonlinear drag force induced by the pulsatile blood flow, which can hardly be measured. Consequently, a mean value theorem (MVT) based observer to estimate the blood velocity from the sole measurement of the robot position is proposed. Also, the stability of the observer-based backstepping controller is proved. The resulting estimation and tracking are then illustrated through simulations, as well as robustness to parametric uncertainty, measurement noise, and dynamical errors when the pulsatile blood flow is incorrectly modeled.

International audience; Scramjets are suitable for hypersonic flight, but their use requires the ability to ensure theirthermal protection. In this context, a remotely controlled fuel-cooled combustor, suitable for theexperimental analysis of the pyrolysis-combustion coupling characterizing a regenerativelycooled combustion chamber when a hydrocarbon propellant is used, has been designed.Similitude rules were used. Ethylene is used as fuel, air as oxidizer, with an equivalence ratiobetween 1.0 and 1.5 and a fuel injection pressure between 1 and 10 bar. Experiments arerealized by varying operating conditions, to determine their impact on combustor heat transferdynamics. Previous numerical results have been confirmed. A hysteresis effect has beendemonstrated. It has been observed that system response time to fuel mass flow rate increases islower (of about 40 to 50%) than system response time to fuel mass flow rate decreases and thata rise in equivalence ratio from 1.0 to 1.25 produces an increase in system response intensitythat is, following the operating conditions, in the range from 90% to 170%. It has also been seenthat an increase in equivalence ratio from 1.0 to 1.5 produces a raise of the temperature of thefuel-coolant (of about 40 to 50%), due to the increase in the emissivity of the flame.

This paper presents a mixed numerical approach to model the blast waves generated by the detonation of a spherical stoichiometric mixture of propane and oxygen, impacting a plate-liquid system. The problem is split into two parts. The first calculation part relies on the modeling of the blast load and its propagation. Over-pressure distribution, in this part, is presented and reveals a very good level of agreement with experimental results. The time and space scales of the blast load data must be compatible with the plate-liquid system. This compatibility is ensured by an appropriate spatio-temporal interpolation technique. This technique is presented and its effectiveness and accuracy are demonstrated. The second part consists in modeling the response of the coupled plate-liquid system under the numerical blast load model. Experiments at reduced scale are carried out in two configurations in order to assess the effectiveness of this mixed numerical approach. Convincing results are obtained and discussed.

International audience; A simple multi-0D model of a 3-Way Catalytic Converter (3WCC) is built from physical equations, integrating the temperature dynamics and a pollutant emission conversion map. The validated model involves complexity and performances suitable to be integrated in a high fidelity powertrain model of a gasoline-Hybrid Electric Vehicle (HEV). Next, a pollutant constrained optimal energy management is derived from the Pontryagin Minimum Principle. The approach allows the joint minimization of pollution and fuel consumption with only one parameter to tune, by considering all the standardized pollutant emissions. The proposed strategy significantly reduces pollutant emissions with only a slight fuel consumption increase. Using a complex HEV model shows the feasibility of the pollution constraint integration in on-line energy management.

A new automated method to generate smooth Finite Element meshes for the meso-scale representation of textile composites unit cells with preformed and compacted reinforcements is presented. The reinforcement geometry is obtained from textile geometry preprocessors or by modeling of the preforming step. It is then transformed into a geometric model of the entire composite unit cell, taking into account the real yarn shape after preforming and compaction, the contact interactions between the yarns, as well as the resulting local variations of fiber volume fraction. A consistent mesh of the complete cell, conformal between the parts, is ensured, without the need for inserting artificial matrix layers between the yarns. The methodology is illustrated by mechanical modeling of the representative unit cell of a multi-layer composite with a compacted and nested woven reinforcement.

International audience; This paper presents an investigation on the use of new light-weight construction material, composed of lime, water and cereal straw fiber. Two types of fibers were used: wheat and barley straw. The influence of some parameters such as fiber types, binder types (lime and/ or lime-cement), fiber to binder ratio (F/B) and Water to Binder ratio (W/B) on the mechanical and thermal properties is studied. Compressive strength, thermal conductivity and density of the material were investigated. The results indicated that the thermal conductivity of the straw-lime composites decreased with increasing straw content. The result comparisons also revealed that the composite reinforced by wheat straw fibres has the highest compressive strength.

International audience; Image understanding has many real industrial applications (video-monitoring, image retrieval . . . ). Given an image and an associated ground truth, it is possible to quantify the quality of understanding results provided by different algorithms or parameters. To this end, it is necessary to take into account many factors for each object in the image: localization and recognition errors and under or over-detection of objects. In order to define an evaluation metric for quantifying the quality of an image understanding result, we have to set, as for example, the weights of each kind of error in the global score. For a correct parameters setting of an evaluation metric we defined previously, we conducted a subjective evaluation of image understanding results involving many experts in image processing. We present in this paper the developed method and analyze the obtained results to weight the various errors in an appropriate way. We show the benefit of this kind of study to define the correct parameters of the metric in order to have a judgment as reliable the one provided by experts. Experimental results on many images from the PASCAL VOC Challenge show the good behavior of this metric compared to the human judgment.

International audience; The paper proposes an energy management strategy (EMS) for hybrid electric vehicles (HEV) and plug-in hybrid electric vehicles (PHEV) taking into account battery health through an additional soft constraint on battery internal temperature, considered as one of the prime factors influencing battery aging. Battery cell temperature is modeled and considered as a second state constraint with the state of energy (SOE) in the optimization problem solved on-line using the equivalent consumption minimization strategy (ECMS). Simulation results are presented to highlight the contribution of the strategy including battery thermal management compared to the standard approach.

International audience; This research work aims to study the impact of the heating system on the fuel consumption of a hybrid electric vehicle (HEV). The thermal engine is less used in an HEV than in a thermal vehicle, thus the cabin heating is partly ensured by electrical resistances. However, because the battery is partly charged by the thermal engine, this electrical heating has an impact on the fuel consumption. In the present work, a multi-domain model is proposed to analyze the impact of the heating system on the fuel consumption of a HEV. The models of the different physical subsystems are organized and unified by energetic macroscopic representation (EMR). Experimental validations, with an accuracy of 95%, are provided for each subsystem model. The validated simulation models are used to study the impact of the heating system for a specific driving cycle and climatic condition. For a simple energy management strategy (EMS), there is an over-consumption of 19% that is due to the heating system. When a more efficient EMS is used, the over-consumption is reduced to 12%. This study shows the interest in developing advanced energy management strategies that couple the traction and the heating functions of the vehicle.

This paper provides finite elements predictions for the effective tangent properties of microcracked viscoelastic masonry. This model relies on two steps. The first one is based on the identification at the short and long terms of an approximate analytical creep function for the mortar. Following (Nguyen et al., 2011) for a microcracked homogeneous material, this step rests on the coupling between the Griffith’s brittle fracture theory and stress-based dilute homogenization scheme. It allows to avoid recourse to ‘heavy’ numerical inversion of the Laplace–Carson transform. The second step provides orthotropic overall properties of masonry by means of finite elements method used to carry out periodic homogenization on masonry. This step relies on the calculation of the localization strain tensors in each phase (brick and mortar). This paper proposes an alternative model to an incremental homogenization of masonry which is heavier to carry out since it depends on additional parameters such as time increment and prestresses in viscoelastic phases added to the crucial computation step of strain localization tensors in each phase. In this work, for the sake of simplicity and as a first approach, only time-dependent crack densities (Shrive et al., 1997; Reda Taha and Shrive, 2006) are considered at short and long-terms. Results provided by the proposed finite elements model can be considered as reference solution enabling a rigorous assessment of recently analytical model proposed by Nguyen et al. (2014) and Rekik et al. (2016) as a generalization of the Cecchi & Taliercio model (Cecchi and Taliercio, 2013) initially available for uncracked masonries. This comparison is carried out for different mortar thicknesses and brick’s dimensions. The case of a compressed wall is also considered at the long term.

International audience; Earth mover's distance is one of the most effective metric for comparing histograms in various image retrieval applications. The main drawback is its computational complexity which hinders its usage in various comparison tasks. We propose fast earth mover's distance computation by providing better initialization to the transportation simplex algorithm. The new approach enables faster EMD computation in Visual Memory (VM) compared to the state of the art methods. The new proposed strategy computes earth mover distance without compromising its accuracy.

International audience; This paper considers the problem of eco-driving for electric cars. This problem is formulated as an Optimal Control Problem (OCP) aiming at minimizing the vehicle's energy consumption over fixed time and distance horizons. The impact of battery parameter variations and auxiliary power demands on the optimal vehicle velocity computation are studied from a model complexity viewpoint. Simulation results are presented and discussed to illustrate the suggested simplifications.

International audience; In this paper, two simplified methods based on Dynamic Programming (DP) to solve an Eco-driving problem for a conventional vehicle equipped with an internal combustion engine are studied. The first method is based on the transformation of a time-based Optimal Control Problem (OCP) into a distance-based OCP while the second is based on solving the time-based OCP directly. The Pontryagin Minimum Principle (PMP) is used to decrease the complexity of the OCP formulation. Based on simulations, the two methods are compared in terms of optimality (fuel consumption) and the time needed to run the DP. The impact of the mesh choice on the optimality of the solution is also investigated.

We propose to apply the Gaussian Mixture Model (GMM) to surface electromyography (sEMG) signals in order to detect the muscular activation (MA) onset, timing off and intervals. First, classical time and frequency features are extracted from the sEMG signals, beside the Teager-Kaiser energy operator (TKEO) is evaluated and added as a new feature which enhances the detection performance. All the obtained features are then used as the input for the GMM to conduct the binary clustering. Finally, a decision theory is applied in order to declare sEMG activation timing of human skeletal museles during movement. Accuracy and precision of the algorithm are assessed by using a set of synthetic simulated sEMG signals and real ones. A comparison with two previously published techniques is conducted: wavelet transform-based method and double threshold-based method. Our experimental results prove that the proposed GMM-based algorithm is able to accurately reveal the MA timing with performance beyond that of the state-of-the-art methods. Moreover, this proposed algorithm is automatic and user-independent.

International audience; Legislation concerning the pollutant emissions of diesel passenger cars is becoming increasingly restrictive, especially for nitrogen oxide (NOx) and particulate matter (PM). A methodology that combined physical approach and a dynamic output feedback based observer for the modelling and estimation of the pollutant emissions in diesel engines is thus proposed. A grey box control-oriented model of NOx emissions is developed; this model was designed based on a deep analysis of the crank-angle discrete combustion process of the test-bench diesel engine. This model will be used to help the understanding of the thermal NOx emission formation process. It provides the basics for the further development of the diesel engine emissions control-oriented models, but it couldn't be used to predict real-time pollutant emissions on a vehicle. The model is calibrated by experimental measurements and a complete cylinder pressure and temperature analysis. Then, to solve the vehicle integration problem (real-time calculation problem) this model and a linear model of the diesel engine were used to develop an dynamic observer to estimate the quantities of particulate matter.

In this paper, we propose a modeling technique for the surface electromyographic (sEMG) signals based on the fractional linear prediction (FLP). To our knowledge, this is the first time application (use) of the FLP modeling to sEMG Data. This study is motivated by the ability of FLP modeling for characterizing a waveform with a reduced set of parameters. The FLP is applied on real sEMG data recorded on the Soleus muscles under walking conditions and preliminary results are obtained. The dynamics of FLP coefficients for persons suffering from Parkinson's disease (PD) were shown to be lower than those for healthy subjects. This suggests less adjustment possibilities in the neuromuscular response of the PD subjects compared to the healthy subjects. Perspectives are the evaluation of fractal components of nonstationary EMG data in connection with FLP evolution.

A new efficient and user-independent technique for the detection of muscle activation (MA) intervals is proposed based on Gaussian Mixture Model (GMM) and Ant Colony Classifier (AntCC). First, time and frequency features are extracted from the surface electromyography (sEMG) signals. Then, GMM is used to cluster these extracted features into burst & non burst. Those features with their class name are then used as the input for the AntCC algorithm in order to derive classification rules. Finally, the obtained rules are used to detect sEMG activation timing of human skeletal muscles during movement. The performance of the proposed technique is demonstrated by means of synthetic simulated sEMG signals and real ones. The proposed technique is then compared to two previously published techniques: wavelet transform-based method [1] & double threshold-based method [2]. It is concluded that our technique outperforms those methods and significantly improves the accuracy of good MA timing detection. Moreover, to our knowledge, the proposed technique is the first user-independent one since no tuning parameters are required. Our findings show that the proposed method is convenient for automatically processing large amounts of sEMG signals with performance beyond that of the state of the-art methods.

The modernized Global Positioning System (GPS) and Galileo satellite navigation systems have recommended the interoperable Multiplexed-Binary Offset Carrier (MBOC) modulation on L1C/E1 frequency band for their open service signals. MBOC signal multiplexes BOC( 6 , 1 ) and BOC( 1 , 1 ) according to a certain proportion, and can be implemented in a number of ways, including Composite BOC (CBOC) and Time-Multiplexed BOC (TMBOC). The complexity and the accuracy of the theoretical signal modeling pose a growing concern to scrutinize and to optimize signal characteristics and receiver performances. In this paper, the analytical expressions of the optimal correlation function (CF) for the new modernized MBOC signal are proposed. Also derived are the analytical models of the discriminator functions (DFs) and the multipath error envelopes (MEEs) in both coherent and non-coherent code tracking configurations. The validation of the proposed models is performed and the simulation results show that the proposed analytical models coincide with the numerical ones.

This paper deals with the design process adapted to medical robots. The large diversity of kinematic architectures that can be encountered in medical robotics leads us to seek a robust method dedicated to tool-guidance medical robot design. First, we detail a proposed design process adapted especially for handling the inherent needs in tool-guidance in medical robotics. This proposed method ties together the phases of the design process with their respective tools. We describe the spectrum of medical robots and particularly the variety of kinematic architectures used. Each phase of the design process is detailed through application examples in the domains of tele-echography and minimally invasive surgery, which exhibit a number of commonalities. The use of tools for accomplishing the various steps of the design process is detailed, with emphasis on medical gesture analysis. This is followed by topological and dimensional synthesis. This study illustrates how the type of medical robot can impose specific requirements and a particular approach in the design process. We expect through this paper to bring a significant contribution to the design of medical tool-guidance robots and to facilitate their integration in the clinical environment. The main contribution of this work is to propose a design process method for robotic medical tool-guidance manipulators.

International audience; This work investigates the tow buckling defect that may take place during the forming of complex shapes. The defect is studied independently of the process with a device specially designed. A specific instrumentation was associated to the device. Structure light interferometry was chosen to measure the elevation of the tows exhibiting the buckling defect all along its growth. The device and its instrumentation were validated in this work and a preliminary study was performed to investigate the origin of the tow buckle’s appearance and its growth kinematic. The growth kinematic of the buckle’s appearance consisting on a double simultaneous rotation of the tow exhibiting the buckle around the Z and the Y axis was established. It was shown that the in-plane bending of the tow is a key parameter that can probably be considered as a preliminary criterion that conditions the appearance and the growth of the tow buckling defect.

International audience; This work focuses on the problem of automatically extracting human 3D poses from a single 2D image. By pose we mean the configuration of human bones in order to reconstruct a 3D skeleton representing the 3D posture of the detected human. This problem is highly non-linear in nature and confounds standard regression techniques. Our approach combines prior learned correspondences between silhouettes and skeletons extracted from 3D human models. In order to match detected silhouettes with simulated silhouettes, we used Krawtchouk geometric moment as shape descriptor. We provide quantitative results for image retrieval across different action and subjects, captured from differing viewpoints. We show that our approach gives promising result for 3D pose extraction from a single silhouette.

International audience; To support optimisation of refractory masonry structures compressibility of dry joints of magnesia-carbon and magnesia-chromite bricks have been investigated. Laboratory scale tests and finite element modelling have been performed. Measurements done in wide temperature range have shown that the exponential form of the joint closure curve results from gradual closure of initially non-parallel surfaces. The stress needed to close the joint was found to be proportional to the material stiffness. Temperature influences the joint closure by changing the stiffness of material and by reducing the initial joint gap due to thermal expansion.

This paper deals with thermo-hydromechanical behaviour of a stone masonry wall assembled with mortar. Macroscopic modelling is proposed that allows an evaluation of the impact of everyday thermal and hydraulic variations on the generation of variations in effective stress. The thermo-hydromechanic parameters of the unsaturated stone have been estimated by experimental tests for tuffeau, such as that found in the Chambord monument. An inverse identification has been used for the optimisation of these parameters based on in- situ measurements, before using the model for long-term analysis. The results of this numerical study show that the contrast between the properties alone is not sufficient to generate a stress that is superior to the stone strength. Otherwise, the variations in thermal and hydraulic conditions on the surface of the stone could lead to a fatigue-like degradation of historic monuments. Cet article s’interesse au comportement thermo-hydromecanique d’un mur maconne en pierre liee avec un mo...

International audience; One of the major challenges of the wind energy sector is to accurately predict the wind potential. This task is especially difficult in mountainous terrains where the topography can imply complex reliefinduced flows. Wind tunnel testing is one of the possibilities to simulate and predict the wind for wind turbine micro-siting. Most advanced quantitative measurement techniques can be used in the wind tunnel, however, measuring the whole terrain to find the highest wind potential zones is very timeconsuming. This paper proposes to use a very simple, quick and cheap technique to detect and evaluate the high wind speed areas over an entire model. Commonly used for pedestrian wind comfort assessment, the sand erosion technique is here applied to wind resource assessment. The technique can provide valuable qualitative information but can also give an order of magnitude of the local speedup. It is first applied to a backward facing step flow and then on a mountainous terrain. An amplification factor and the fractional speed-up ratio (FSR) can be calculated over the entire mountain. For high speed positions results extracted from sand erosion appears to be comparable the one calculated by particle image velocimetry. The technique is repeatable, able to perform a detection of the high speed area, and capable of giving an estimation of the amplitude of the wind. The technique allows to restrict the use of quantitative measurements to the most interesting areas.

International audience; The first stage of the RTM process concerns the preforming of the part. During the preforming of multilayered reinforcements, frictions between the plies occur. An experimental device designed to analyze the ply/ply, ply/tool, and yarn/ yarn frictions has been built. Specific contact behavior for ply/ply friction is directly related to shocks taking place between overhanging yarns of each sample. Within the signal, two particular periods can be exhibited. This work shows for four different architecture woven fabrics that the two periods are directly related to the meso-architecture of the studied woven fabrics and more precisely to the characteristics of the fabric unit cell. The friction response appears to be very sensitive to the relative positioning and orientation of the samples. It can also lead to manufacturing defects such as unweaving or wrinkles. Up to now, finite element codes predicting the forming behavior of woven reinforcement fabrics do not take into account this friction coefficient evolution to model accurately the process. This point could therefore be addressed in future works on this topic with the view to optimize the multi-ply forming of composite parts.

International audience; The market demand for thicker complex shaped structural composite parts is increasing. Processes of the Liquid Composite Moulding (LCM) family, such as Resin Transfer Moulding (RTM) are considered to manufacture such parts. The first stage of the RTM process consists in the preforming of the part. During the pre-forming of multilayered reinforcements, frictions between the plies occur and need to be taken into account for the forming simulation. An experimental device designed to analyse the ply/ply and ply/ tool frictions has been set up. The different set up steps of the device are described. First results are presented, which show the ply/ply friction behaviour for a glass plain weave fabric. A specific contact behaviour has been observed for dry reinforcement fabric in comparison to non-technical textiles. A honing effect classically observed in dry fabric testing has also been pointed out through cyclic experiments. It can be attributed to both fibre material abrasion and fibre reorganisation inside the yarn.

In this paper, a neural network model is developed to simulate the accumulation of apparent fatigue damage of 3D trabecular bone architecture at a given bone site during cyclic loading. The method is based on five steps: (i) performing suitable numerical experiments to simulate fatigue accumulation of a 3D micro-CT trabecular bone samples taken from proximal femur for different combinations of loading conditions; (ii) averaging the sample outputs in terms of apparent damage at whole specimen level based on local tissue damage; (iii) preparation of a proper set of corresponding input–output data to train the network to identify apparent damage evolution; (iv) training the neural network based on the results of step (iii); (v) application of the neural network as a tool to estimate rapidly the apparent damage evolution at a given bone site. The proposed NN model can be incorporated into finite element codes to perform fatigue damage simulation at continuum level including some morphological factors and some bone material properties. The proposed neural network based multiscale approach is the first model, to the author’s knowledge, that incorporates both finite element analysis and neural network computation to rapidly simulate multilevel fatigue of bone. This is beneficial to develop enhanced finite element models to investigate the role of damage accumulation on bone damage repair during remodelling.

The transformation of the sheet into a product without failure and excess of material in a deep drawing operation means that the initial blanks should be correctly designed. Therefore, the initial blank design is a critical step in deep drawing design procedure. Consequently, an easy approach for engineers in predicting the initial blank shape is necessary to reduce wastage in material and to overcome the large time consumed in the classical approaches. Thus, the aim of the present investigation is to propose an automatic procedure for the quick sheet metal forming optimization. In fact, a metamodel will be build based on artificial neural networks which will be coupled then with an optimization procedure in order to predict the initial blank shape in a rectangular cup deep drawing operation. The metamodel is built from the finite element simulations using ABAQUS commercial code. This procedure allows a significant reduce of the CPU time compared to classical optimization one. The results show that the desired shape is in good agreement with the one calculated using the optimized blank shape.

International audience; The present study examines the global effect of a laboratory-simulated exhaust gas recirculation (EGR) on the homogeneous charge compression ignition (HCCI) combustion of n-heptane in a transparent monocylinder diesel engine. The investigations were carried out during the engine cycle for two EGR rates, 0 and 30%, at a constant 0.3 equivalence ratio. The analysis is focused on the relation between natural combustion emissions during the cold and main combustion phases and the OH* chemiluminescence and the processes involved inside the combustion chamber of a compression engine. By observing cool and main flame emissions, it was observed that the dilution by burnt gas delays and degrades the combustion phenomenon. For both cases, the natural flame emission is not uniformly distributed, because of the nonhomogeneity of the gas temperature. By coupling kinetic analysis, provided by a zero-dimensional model, it was shown that the natural emissions of combustion are sufficiently sensitive to yield combustion process analysis.

A core operational definition of active and healthy aging (AHA) is needed to conduct comparisons. A conceptual AHA framework proposed by the European Innovation Partnership on Active and Healthy Ageing Reference Site Network includes several items such as functioning (individual capability and underlying body systems), well-being, activities and participation, and diseases (including non-communicable diseases, frailty, mental and oral health disorders). The instruments proposed to assess the conceptual framework of AHA have common applicability and availability attributes. The approach includes core and optional domains/instruments depending on the needs and the questions. A major common domain is function, as measured by the World Health Organization Disability Assessment Schedule 2.0 (WHODAS 2.0). WHODAS 2.0 can be used across all diseases and healthy individuals. It covers many of the AHA dimensions proposed by the Reference Site network. However, WHODAS 2.0 does not include all dimensions proposed for AHA assessment. The second common domain is health-related quality of life (HRQoL). A report of the AHA questionnaire in the form of a spider net has been proposed to facilitate usual comparisons across individuals and groups of interest. (C) 2015 Published by Elsevier Inc. on behalf of AMDA - The Society for Post-Acute and Long-Term Care Medicine.

International audience; The energy management strategy (EMS) of a Hybrid electric vehicle (HEV) is a supervisory control meant to steer power flows among the different energy sources of the powertrain. The EMS is designed for the purpose of reaching the maximum potential of a given powertrain architecture. Fuel consumption is not the only criterion to be considered for evaluating the global performance of an EMS, as the EMS also impacts other features such as drivability, through engine starts/stops and gear shifts. This paper proposes a general approach to include drivability constraints in the EMS of powertrains provided with discrete automated transmissions. As a first step, determinist dynamic programming (DDP) is used to compute the optimal solution of the resulting optimization problem. This optimal solution is then used to assess the performance of the proposed on-line strategy which is based on the Equivalent Consumption Minimization Strategy (ECMS). Finally, simulation results are completed by a discussion of experimental results, as the proposed on-line strategy has been implemented in a prototype vehicle designed at Renault and tested on a dynamometric bench.

International audience; The aim of this work is, while using an optimal approach with dynamic programming (DP), to determine the best eco-driving cycle starting from a given driving cycle. To find the cycle with the minimum fuel consumption, two ways of computing an eco-driving cycle are compared. For a Hybrid Electric Vehicle (HEV), is it more efficient to compute an HEV based eco-driving cycle or to compute a conventional vehicle based eco-driving cycle and then to adapt it to a hybrid vehicle? Hence, two different DP programs are developed, that lead to two different eco-driving cycles: hybrid and conventional. Finally, the fuel consumption of an HEV with two Energy Management Strategies (EMS) will be compared: an off-line optimal one based on the Pontryagin Minimum Principle and an on-line suboptimal Equivalent Consumption Minimization Strategy (ECMS). The paper shows that in order to generate the best speed trajectory in terms of fuel consumption, it is necessary to take the fact that the vehicle is an HEV into account.

International audience; Forming thick, complex shapes with several layers is needed in high technology fields. During forming, defects can occur and have to be taken into account because they can significantly affect the mechanical performance of the part. This experimental study shows that, when working with dry fabric forming, the type and number of defects is a function of the punch geometry, the process parameters, the orientation of the fabric with respect to the punch and the inter-ply friction. Inter-ply friction has a huge effect on the quality of the preform when inter-ply sliding occurs. This inter-ply friction leads to several overhanging yarn shocks that generate high tangential forces, which inhibit the relative sliding of plies. In addition, to reduce the number and amplitude of defects, the layers subjected to severe defects can be placed in the inner position where they are subjected to the compression applied by the upper layers.