Residual associated with subspace identification.

Assume now that a reference ${\theta }_0$ and a new sample $Y_1,\cdots ,Y_N$ are available. For checking whether the data agree with ${\theta }_0$, the idea is to compute the empirical Hankel matrix ${\widehat{\mathscr{H}}}_{p+1,q}$ defined by

and to define the residual vector

Let $\theta$ be the actual parameter value for the system which generated the new data sample, and ${𝐄}_{\theta }$ be the expectation when the actual system parameter is $\theta$. From (13), we know that ${\zeta }_N\left({\theta }_0\right)$ has zero mean when no change occurs in $\theta$, and nonzero mean if a change occurs. Thus ${\zeta }_N\left({\theta }_0\right)$ plays the role of a residual.

As in most fault detection approaches, the key issue is to design a residual, which is ideally close to zero under normal operation, and has low sensitivity to noises and other nuisance perturbations, but high sensitivity to small deviations, before they develop into events to be avoided (damages, faults, ...). The originality of our approach is to:

• Design the residual basically as a parameter estimating function,
• Evaluate the residual thanks to a kind of central limit theorem, stating that the residual is asymptotically Gaussian and reflects the presence of a deviation in the parameter vector through a change in its own mean vector, which switches from zero in the reference situation to a non-zero value.

The central limit theorem shows  66 that the residual is asymptotically Gaussian:

where the asymptotic covariance matrix $\Sigma$ can be estimated, and manifests the deviation in the parameter vector by a change in its own mean value. Then, deciding between $\eta =0$ and $\eta \ne 0$ amounts to compute the following ${\chi }^2$-test, provided that $𝒥$ is full rank and $\Sigma$ is invertible :

where

3.5.1 Infrared radiation

Infrared is an electromagnetic radiation having a wavelength between $0.2$$\mu m$ and 1 $mm$, this range begins in uv spectrum and it ends on the microwaves domain, see Figure 1.

Show image description

Electromagnetic spectrum - Credit MODEST, M.F. (1993). Radiative Heat Transfer. Academic Press.

For scientific purposes, infrared can be divided in three ranges of wavelength in which the application varies, see Table 1.

Our work is concentrated in the mid infrared spectral band. Keep in mind that Table 1 represents the ISO 20473 division scheme, in the literature boundaries between bands can move slightly.

The Planck's law, proposed by Max Planck in 1901, allows to compute the black body emission spectrum for various temperatures (and only temperatures), see Figure 2 left. The black body is a theoretical construction, it represents perfect energy emitter at a given temperature, cf. Equation (20).

With $\lambda$ the wavelength in m and $T$ as the temperature in Kelvin. The $C_1$ and $C_2$ constants, respectively in W.m${}^2$ and m.K are defined as follow:

with

• $c$ the electromagnetic wave speed (in vacuum $c$ is the light speed in m.s${}^{-1}$),
• $k=1.381e^{-23}$ J.K${}^{-1}$ the Boltzmann (Entropy definition from Ludwig Boltzmann 1873). It can be seen as a proportionality factor between the temperature and the energy of a system,
• $h\approx 6,62606957e^{-34}$ J.s the Planck constant is the link between the photons energy and their frequency.

Show image description

Left: Planck's law at various temperatures - Right: Energy spectrum of the atmosphere

Show image description

Left: Planck's law at various temperatures - Right: Energy spectrum of the atmosphere

By generalizing the Planck's law with the Stefan Boltzmann law (proposed first in 1879 and then in 1884 by Joseph Stefan and Ludwig Boltzmann), it is possible to address mathematically the energy spectrum of real body at each wavelength depending on the temperature, the optical condition and the real body properties, which is the base of the infrared thermography.

For example, Figure 2 right presents the energy spectrum of the atmosphere at various levels, it can be seen that the various properties of the atmosphere affect the spectrum at various wavelengths. Other important point is that the infrared solar heat flux can be approximated by a black body at 5523,15 K.

3.5.2 Infrared Thermography

The infrared thermography is a way to measure the thermal radiation received from a medium. With that information about the electromagnetic flux, it is possible to estimate the surface temperature of the body, see section 3.5.1. Various types of detector can assure the measure of the electromagnetic radiation.

Those different detectors can take various forms and/or manufacturing process. For our research purposes, we use uncooled infrared camera using a matrix of microbolometers detectors. A microbolometer, as a lot of transducers, converts a radiation in electric current used to represent the physical quantity (here the heat flux).

This field of activity includes the use and the improvement of vision system, like in 7.

3.8.1 Mathematical model of electric cables and networks

A cable excited by a signal generator can be characterized by the telegrapher's equations 70

where $t$ represents the time, $z$ is the longitudinal coordinate along the cable, $V(t,z)$ and $I(t,z)$ are respectively the voltage and the current in the cable at the time instant $t$ and at the position $z$, $R\left(z\right),L\left(z\right),C\left(z\right)$ and $G\left(z\right)$ denote respectively the series resistance, the inductance, the capacitance and the shunt conductance per unit length of the cable at the position $z$. The left end of the cable (corresponding to $z=a$) is connected to a voltage source $V_s\left(t\right)$ with internal impedance $R_s$. The quantities $V_s\left(t\right)$, $R_s$, $V(t,a)$ and $I(t,a)$ are related by

At the right end of the cable (corresponding to $z=b$), the cable is connected to a load of impedance $R_L$, such that

One way for deriving the above model is to spatially discretize the cable and to characterize each small segment with 4 basic lumped parameter elements for the $j$-th segment: a resistance $\Delta R_j$, an inductance $\Delta L_j$, a capacitance $\Delta C_j$ and a conductance $\Delta G_j$. The entire circuit is described by a system of ordinary differential equations. When the spatial discretization step size tends to zero, the limiting model leads to the telegrapher's equations.

A wired network is a set of cables connected at some nodes, where loads and sources can also be connected. Within each cable the current and voltage satisfy the telegrapher's equations, whereas at each node the current and voltage satisfy the Kirchhoff's laws, unless in case of connector failures.

3.8.2 The inverse scattering theory applied to cables

The inverse scattering transform was developed during the 1970s-1980s for the analysis of some nonlinear partial differential equations 68. The visionary idea of applying this theory to solving the cable inverse problem goes also back to the 1980s 67. After having completed some theoretic results directly linked to practice 15, 71, we started to successfully apply the inverse scattering theory to cable soft fault diagnosis, in collaboration with GEEPS-SUPELEC 12.

To link electric cables to the inverse scattering theory, the telegrapher's equations are transformed in a few steps to fit into a particular form studied in the inverse scattering theory. The Fourier transform is first applied to obtain a frequency domain model, the spatial coordinate $z$ is then replaced by the propagation time

and the frequency domain variables $V(\omega ,x),I(\omega ,x)$ are replaced by the pair

with the characteristic impedance

These transformations lead to the Zakharov-Shabat equations

with

These equations have been well studied in the inverse scattering theory, for the purpose of determining partly the “potential functions” $q^{\pm }\left(x\right)$ and $q^{*}\left(x\right)$ from the scattering data matrix, which turns out to correspond to the data typically collected with reflectometry instruments. For instance, it is possible to compute the function $Z_0\left(x\right)$ defined in (33), often known as the characteristic impedance, from the reflection coefficient measured at one end of the cable. Such an example is illustrated in Figure 3. Any fault affecting the characteristic impedance, like in the example of Figure 3 caused by a slight geometric deformation, can thus be efficiently detected, localized and characterized.

6.1.1 Vibration extraction from video image flows

Participants: Boualem Merainani, Zhilei Luo, Vincent Baltazart, Michael Doehler, Jean Dumoulin, Qinghua Zhang.

With adequate image processing methods, motion signals can be extracted from video image flows for vibration analysis. In this way, the pixels in selected regions of interest within the images act as a dense network of contactless sensors distributed over the whole structure. In this work, the efficiency of this video-based approach is demonstrated with laboratory experiments on a cantilever beam, in particular, by evaluating its capability for detecting weak damages mimicked by slight mass modifications. To this end, the steerable filter-based method (ST), that recovers displacements from local phase, is first extended to overcome its motion limitation of one pixel size. Then, the performance of the improved motion extraction method is compared with two other well-established methods. Furthermore, a novel image-based motion estimation method that explicitly considers perturbations due to ambient light changes and disturbance by background features is developed. Results are published in 25, 41.

6.1.2 Bias and bottlenecks study in outdoor long term thermal monitoring by infrared thermography: Leveraging opportunistic data for temperature estimation

Participants: Thibaud Toullier, Jean Dumoulin.

This study aims to assess the impact of the targeted scene and environmental variables on in situ longterm thermal monitoring by infrared thermography. It also addresses projection errors arising from camera perspectives and digitization. Over a span of 3 years, extensive measurements were conducted at two instrumented test sites, yielding a valuable dataset for analysis. This work introduces and discusses the model used to convert the collected data into surface temperature estimations and its associated parameters’ sensitivities. Subsequently, it delves into the effects of camera resectioning on infrared measurements. Finally, the study investigates the influence of environmental parameters and proposes a strategy to correct for these effects, with and without additional sensors (opportunistic data) and additional in-situ measurements, published in 28.

6.1.3 Hot box detection on moving targets using way side thermal infrared camera

Participants: Boualem Merainani, Thibaud Toullier, Jean Dumoulin.

In a context of European rail traffic growing rapidly, infrastructure managers are pushed to develop reliable solutions to improve safety and operational performance. In the present study we focus on the bogie component, a complex and important element in rail-road cars. Overheated rail-road car wheels and bearings known as hot boxes, are a major threat for any railway operation. With the advancements in both image sensor technology and processing capabilities, machine vision-based techniques may provide cost-effectiveness and easy solutions for hot wheels and hot axle bearings detection.In this research work, automatic detection, tracking and counting of hot boxes is addressed through the implementation of way side infrared thermal cameras. First a discussion on thermal cameras required performances is proposed and new uncooled fast pixel sensors are introduced. Implementation on a real railway is presented and discussed. A new and affordable thermal vision-based wayside monitoring method is proposed. It combines panoramic thermal images with deep learning (DL)-based object detection algorithm. Panoramas were created, from thermal image sequences collected, by cooled and uncooled cameras, during field experiments on both freight and passenger trains. The results yield promising perspectives, notably in maintaining high safety standards, at a more budget-friendly cost. Publications are 34, 48.

6.1.4 Laboratory and full-scale experiment of a novel hybrid system to harvest energy through concrete pavement

Participants: Domenico Vizarro, Jean Dumoulin.

Havesting energy from the sun through the pavement is a powerful lever to fight against the worldwide energy crisis. Indeed, the pavement surface is daily exposed to the sun light and represents an enormous source of potential harvesting. So far, solutions involving photovoltaic cells placed on the pavement or using a heat-transfert fluid circulating through the structure are studied. Recently, we have developed a novel hybrid system at University Gustave Eiffel combining the two solutions. The structure is made of a porous concrete layer where a heat-transfert fluid is circulating sandwiched between a waterproof structural concrete layer placed below and a photovoltaic layer placed on top. The photovoltaic layer is composed of PV cells protected by a semi-transparent layer designed to support the traffic vehicles, guarantee the skid resistance and enabling the passage of the sun light. At the beginning of the research, the semi-transparent layer and the porous concrete have been optimized in terms of energy harvesting and a laboratory size prototype have been constructed. This work also considers a full scale prototype of the hybrid system tested in real climatic conditions and shows some first results regarding thermal harvested energy measured around 240 W/m2 in the specific conditions tested 36.

6.1.5 Strain transfer model for steel-reinforced optical fiber cable

Participants: Mira Kabbara, Qinghua Zhang, Xavier Chapeleau.

In distributed optical fiber sensors, strain transfer is usually described through a simplified one-dimensional equation, derived from continuum mechanics. This equation, in which a parameter known as the strain-lag parameter takes into account the cable’s geometric and mechanical characteristics, establishes a relationship between the measured longitudinal strain profile within the optical fiber and the real strain profile occurring in the host material. In the case of steel-reinforced optical fiber cables, appreciated for their resistance to breakage during on-site instrumentation, a notable discrepancy between the measured and actual strain profiles is revealed especially in the presence of a strain gradient, indicating that the ability to transfer strain from the host material to the optical fiber is restrained using this type of cable. This work assesses numerically the strain transfer model for steel-reinforced optical fiber sensors in the presence of a strain gradient generated by two void inclusions in a concrete beam. Good accuracy of the strain transfer model is observed by the comparison with a 3D finite element simulation. The result points out also the critical necessity of precisely determining the strain-lag parameter 40.

6.1.6 Fiber optic sensors for strain estimation in pavements

Participants: Xavier Chapeleau.

The need to develop more efficient pavement management systems has driven continuous improvement in structural health monitoring (SHM) systems for road infrastructures. Factors such as low energy consumption, minimal disruption (small-sized and lightweight devices), resistance to environmental conditions, high durability, and sensitivity are some of the possible improvements worthy of exploration. Fiber optics appear to have several of these desired characteristics. This work considers strain measurements performed using fiber optic cables installed at various levels in pavement structures tested on the accelerated pavement testing facility at the Université Gustave Eiffel (or fatigue carousel). The results reveal that strains measured with this technology closely align with those obtained using traditional strain gauges. However, an advantage of fiber optics is that they enable distributed measurements over long distances with high precision (10 m with a measurement interval of 2.6 mm in this study). These measurements highlight a significant variability in pavement strains, seemingly attributable to the heterogeneity of pavement materials and variations in layer thicknesses. Furthermore, the responses of two pavement structures are evaluated. The findings show that the new pavement structure exhibits good bonding and continuous strain distribution, while the rehabilitated pavement structure exhibits partial sliding at the interface and strain discontinuity. For the new pavement structure, the strains are well predicted using a bonded interface, while for the rehabilitated pavement structure, the strains are predicted using a thin elastic interface layer with a low elastic modulus. Results are published in 16, 21.

6.1.7 Investigations on the ageing of GFRP rebar-concrete bond

Participants: Xavier Chapeleau.

Durability investigations are needed to better understand degradation mechanisms and provide optimized safety coefficients. While several researchers have explored the durability of FRP rebars, only few have examined the evolution of the FRP-concrete interfacial bond. The GFRP-to-concrete bond behaviour was experimentally studied, focusing on the combined effects of thermally accelerated ageing in alkaline environment and sustained loading. The evolution of the residual interfacial properties over ageing is presented, allowing for a discussion on the influence of sustained load on the aging process, published in 33.

6.1.8 Development and Validation of a Creep Frame

Participants: Arij Fawaz, Xavier Chapeleau.

In industry, structural bonding can play an important role in repairing and reinforcing metallic structures. One reliable method is to use fracture mechanics tests that allow the determination of fracture toughness in mode I, mode II or mixed-mode. Such strategy can be used during the analysis of the initial design, but also through the aging study. Few studies were conducted to inspect the effect of applied load on the mechanical behavior of bonded joints. An adaptation of the sample’s geometry is required to allow the combination of both ELS and creep configurations. This study adapts the sample’s geometry to allow the combination of both End Load Split test and creep configurations. The new sample configuration is compared to a traditional sample geometry. An optical fiber is employed to monitor crack propagation, thereby enhancing this comparison. The obtained toughness characterizations allowed us proposing a design strategy that was assessed in regards with experimental tests conducted on large size steel samples reinforced with adhesively bonded composite. This work is published in 35, 57.

6.1.9 Static and Fatigue Strength Assessment of Composite Patch Repair

Participants: Xavier Chapeleau.

Marine and Offshore industries are looking for rapid, efficient and qualified repair solutions for damaged or corroded structures. Patch made of composite materials can offer such qualities. Small-scale and large-scale specimens were tested in static and fatigue, and numerical simulations were done in order to develop a robust methodology for the strength assessment of composite patch repairs. The static resistance of the patch designs and manufacturing processes is determined by tensile and bending tests. The results show that long scarf specimens have a higher resistance and that the manufacturing process has little influence. Finally, fatigue tests on long scarf specimens, and for the two studied manufacturing processes, are detailed and compared with a FPSO (Floating Production Storage Offloading) histogram 50.

6.2.1 Generalized Dithering using the Lattice Boltzmann Method

Participants: Romain Noel.

A generalized dithering approach using the Lattice Boltzmann Method (LBM) is proposed. This generalization enables to experiment different models, and to adapt them accordingly to targeted features and problems. As illustrations, 4 models are introduced and tested, including anisotropic gradient-based relaxation time and enhanced equilibrium distribution function, aiming for robust quality and artefacts reduction. These models are implemented on Graphics Processing Units (GPU) and generically interfaced with the OBLiX framework. The several test-cases performed demonstrate that such models yield better qualitative and quantitative results, proving the adaptability of the generalization. By providing a general framework for dithering using the Lattice Boltzmann Method, this allows an adaptation of the algorithms to complex situations deviating from classically encountered cases. Published in 26.

6.2.2 Pore-scale study of phase change materials

Participants: Romain Noel, Clement Rigal.

In recent years, there has been growing interest in the development of reversible thermal storage systems for urban development. One of these systems consists of a fluid flow carrying microcapsules containing a phase-change material (mPCM) circulating in a porous layer integrated into the urban structure. To understand the physical mechanisms at work within this fluid and the solid that surrounds it, which is still incomplete to date, we need to study the various scales involved, particularly that of the pore, which is often difficult to measure experimentally and the subject of much numerical work. The present work aims to investigate the effect of the motion of the mPCM on the the melting process occurring in its inside, focuses on the rotational motion and deals with two scales: the capsule scale and the pore scale. In order to determine the order of magnitude of the angular velocity of small particles moving in a porous medium, a pore-scale numerical study is conducted, using a coupling between lattice Boltzmann method and discrete elements method. The resulting velocities are used in an other numerical study conducted at capsule scale in which an enthalpy-based model provides a realistic solution for the phase change. This work is presented in 62, 63.

6.2.3 A computationally efficient k(ω)-spectral form for partial dispersion analyses within the wave finite element framework

Participants: Alvaro Camilo Gavilan Rojas, Qinghua Zhang, Christophe Droz.

This work addresses the computation of frequency-dependent dispersion curves (i.e., k(ω)) and wave modes within the framework of the Wave Finite Element Method (WFEM) and in the context of highdimensional periodic unit cell models. Numerous applications, ranging from phononics to vibroacoustics, now rely on dispersion analyses or wave expansion over a subset of eigensolutions resulting from the resolution of an eigenvalue problem with a T-palindromic quadratic structure (T-PQEP). To exploit the structure of finite element models, various structure-preserving linearizations have already been developed to achieve partial wave resolution of large T-PQEP, primarily targeting the dominating (least decaying) waves. We derive an alternative linearization of the T-PQEP for the k(ω) problem, which leads to enhanced targeting of the eigenvalues around the unit circle and reduces the inaccuracies induced by root multiplicity. A specific form of the problem is then proposed as an optimal compromise between ease of implementation, numerical stability, convergence and accuracy enhancement. The performance of our proposed linearization is compared against existing ones across various iterative eigensolvers, since the generalized eigenvalue problems involve complex non-hermitian matrices, which are not extensively included in eigensolvers. Results indicate that the proposed linearization should be favored for the WFEM, as it provides numerical enhancements in dispersion and wave vectors computation for large eigenvalue problems, as well as for further wave expansion applications 18, 59.

6.2.4 Computing the dynamic response of a periodic structure coupled with a nonlinear junction

Participants: Vincent Mahe, Adrien Melot, Christophe Droz.

Structures constituted of assembled sub-components are often subject to nonlinear effects due to the presence of joints or contacts, which can induce higher-order harmonics of the source excitation. In the context of periodic structures, these nonlinearities cannot be tackled using the Floquet-Bloch theory in its usual harmonic formulation. This study hence presents an adaptation of the classical Floquet-Bloch theory to address multi-harmonic systems arising from a finite number of localized nonlinearities in periodic waveguides. We adopt the Harmonic Balance Method to recast the governing equations into a nonlinear algebraic system, which can then be solved using numerical continuation algorithms. To demonstrate the validity and benefit of this approach, the predictions derived from the proposed methodology are compared to results obtained through conventional Finite Element analysis. Excellent agreement and a notable reduction in computational cost are observed. This efficient procedure for analysing the nonlinear dynamic response of periodic waveguides opens up new possibilities in the study of damaged composite structures 42.

6.2.5 CoSApp: a Python library to create, simulate and design complex systems

Participants: Mathias Malandain.

CoSApp, for Collaborative System Approach, is an object-oriented Python framework that allows domain experts and system architects to create, assemble, simulate and design complex systems. The API of CoSApp is focused on simplicity and explicit declaration of design problems. Special attention is given to modularity; a very flexible mechanism of solver assembly allows users to construct complex, customized simulation workflows. CoSApp handles steadystate simulation, as well as time-dependent dynamic systems, and multimode systems with event-based mode transitions 20.

6.2.6 Control of isolated response curves through optimization of codimension-1 singularities

Participants: Adrien Melot.

We introduce a computational framework for controlling the location of isolated response curves, i.e. responses that are not connected to the main solution branch and form a closed curve in parameter space. The methodology relies on bifurcation tracking to follow the evolution of fold bifurcations in a codimension-2 parameter space. Singularity theory is used to distinguish points of isola formation and merger from codimension-2 bifurcations and an optimization problem is formulated to delay or advance the onset or merger of isolated response curves or control their position in the state/parameter space. We illustrate the methodology on three examples: a finite element model of a cantilever beam with cubic nonlinearity at its tip, a two-degree-of-freedom oscillator with asymmetry and a two-degree-of-freedom base-excited oscillator exhibiting multiple isolas. Our results show that the location of points of isola formation and mergers can effectively be controlled through structural optimization 22, 60, 45.

6.2.7 Multi-parametric optimization for controlling bifurcation structures

Participants: Adrien Melot.

Bifurcations organize the dynamics of many natural and engineered systems. They induce qualitative and quantitative changes to a system’s dynamics, which can have catastrophic consequences if ignored during design. In this paper, we propose a general computational method to control the local bifurcations of dynamical systems by optimizing design parameters. We define an objective functional that enforces the appearance of local bifurcation points at targeted locations or even encourages their disappearance. The methodology is an efficient alternative to bifurcation tracking techniques capable of handling many design parameters (>10²). The method is demonstrated on a Duffing oscillator featuring a hardening cubic nonlinearity and an autonomous van der Pol-Duffing oscillator coupled to a nonlinear tuned vibration absorber. The finite-element model of a clamped-free Euler–Bernoulli beam, coupled with a reduced-order modelling technique, is also used to show the extension to the shape optimization of more complicated structures. Results demonstrate that several local bifurcations of various types can be handled simultaneously by the bifurcation control framework, with both parameter and state target values. 23

6.2.8 Robust gear design with respect to the primary resonance induced by backlash nonlinearity

Participants: Adrien Melot.

This work investigates the influence of uncertain tooth profile modifications on the nonlinear dynamic response of a spur gear pair induced by a backlash nonlinearity. To this end, an original approach based on bifurcation tracking is developed. The equations of motion are solved in the frequency domain with the harmonic balance method (HBM) coupled to an arc-length continuation algorithm and a bordering technique. The evolution of the bifurcation points with respect to the uncertain parameter is computed deterministically. To characterize the severity of the resonance, we introduce two criteria, namely the amplitude of the response at the resonance peak and the width of the resonance hysteresis. Uncertainty is then introduced by means of probability densities of the tip relief within the tolerance interval. Probability density functions of various indicators of the severity of vibro-impacts can be computed with Monte-Carlo simulation with minimal computational burden. We show that the optimum profile correction differs significantly from the one obtained in the purely deterministic case.46, 61.

6.3.1 Subspace-based modal identification and uncertainty quantification from video image flows

Participants: Boualem Merainani, Bian Xiong, Zhilei Luo, Vincent Baltazart, Michael Doehler, Jean Dumoulin, Qinghua Zhang.

The performance of different motion extraction methods from video image flows is compared in the context of OMA, where natural frequencies, damping ratios and mode shapes with high spatial resolution are estimated together with their uncertainty bounds using covariance-driven subspace identification. The compared methods are evaluated with the help of reference laser displacement measurements as well as a finite element model of the beam, revealing differences in the accuracy of the estimated mode shapes depending on the chosen method for motion extraction. Aiming to investigate early structural damage detection, experiments are carried out under small structural changes and the results are compared to a reference state with the help of estimated uncertainties. Small but statistically significant changes in the modal parameters are detected, showing the potential of the vision based framework for SHM. Furthmermore, the robustness against ambient light changes and disturbance by background features is investigated in the context of OMA. Results are published in 25, 41.

6.3.2 Data-Driven Identification of Noise Covariances in Kalman Filtering for Virtual Sensing Applications

Participants: Michael Doehler.

The optimality of the Kalman filter for state-estimation depends on the knowledge of the process and measurement noise covariance. In applications, these covariances are often treated as tuning parameters, often adjusted in a heuristic manner based on user-defined performance criteria. While several methods to identify them from data exist, some require the use of optimization algorithms, or inversion of large matrices, which is numerically inefficient. In this work we review a simple data-driven subspace identification approach to estimate the process and measurement noise covariance, and apply it in the context of virtual sensing. The estimates of the noise covariances are obtained from the residuals of a regression of the banks of shifted Kalman filter states obtained from data and the model matrices of the monitored system. The performance of the approach is illustrated on a numerical example of a chain system by comparing the system states in unmeasured locations to the Kalman filter-based predictions 37.

6.3.3 Uncertainty-Based Clustering and Tracking of Modal Parameters

Participants: Johann Priou, Michael Doehler.

Automated modal parameter tracking is an essential part of vibration-based structural health monitoring of engineering structures, e.g., bridges, high-rise building, wind turbines, amongst others. In this work, we conceive a robust algorithm for accomplishing this task. First, the modal parameters are extracted with by greedy clustering the stabilization diagram, which is based on covariance-driven subspace identification and the related uncertainty quantification. After the selection of a set of relevant modal parameters to be monitored, the tracking is performed by an active search of stable modes close to the reference ones, again using statistical criteria and automated clustering. The applied strategies allow for a robust method, allowing also for relatively strong variations in the natural frequencies or the MAC without losing track of the monitored modes. Finally, when EOVs (Environmental and Operational Variables) are measured and training and testing periods are defined, a neural network is trained based on the tracked frequencies and the EOVs, with the goal of correcting the environmental influence on the estimated frequencies based on the model predictions. In doing so, a distance metric is defined for damage detection, which is robust to environmental perturbations under the measured EOVs. The whole framework is showcased on monitoring data of several full-scale bridges 51, 52, 53.

6.3.4 Modal mass estimation

Participants: Mikkel Tandrup Steffensen, Michael Doehler.

In operational modal analysis with exogenous inputs, the modal parameters are estimated based on partly measured inputs and outputs. Different algorithms can be used, like combined deterministic stochastic subspace identification (CSI) algorithms or the poly-reference Least Squares Complex Frequency (pLSCF) algorithm. Common for both types of algorithms are the computation of modal mass to normalize the estimated mode shapes according to the inputs. In this work, the modal mass computation for CSI and the pLSCF-algorithm is revisited, where a particular focus is put on the assumption on the intersample behavior that is required for the relation between estimates obtained from the discrete-time system (where the data comes from) and the corresponding continuous-time system (where the modal mass is computed). The discretization methods: zero-order hold, first-order hold, and the impulse invariant discretization are studied and discussed. It is shown that the modal mass estimates are heavily dependent on the discretization approach used. The results are illustrated in the context of Monte Carlo simulation of a six-degrees-of-freedom chain system 58.

6.3.5 Linear time periodic system approximation based on Floquet and Fourier transformations for operational modal analysis and damage detection of wind turbine

Participants: Ambroise Cadoret, Laurent Mevel.

Operational Modal Analysis (OMA) identifies modal properties of mechanical structures from vibration data collected from a few sensors under operation conditions. These methods are widely used to monitor civil engineering structures that are modeled as time-invariant systems. However, when dealing with wind turbines and rotating machines, many dedicated OMA techniques were developed to deal with the time-periodic behavior. Existing methods pre-process the data to adapt them to classical identification techniques. Yet, these methods present limitations as either requiring a high number of measured rotation periods, assuming the assumption of an isotropic rotor (i.e. undamaged), or the knowledge of the rotational speed. This work proposes to lift these difficulties by proving that the application of the classical system identification methods can produce meaningful estimates for anisotropy monitoring. It results in the possibility of using classical identification methods without modification to retrieve the system matrices of the approximated time-invariant system. The identified modes can be used reliably for the monitoring of operating wind turbines and more especially for fault detection, at the expense of losing a part of the complete description of the periodic system. The resulting anisotropy monitoring approach and its capacities are illustrated in two cases 17.

6.3.6 Seismic assessment of bridges through structural health monitoring: a state-of-the-art review

Participants: Michael Doehler.

The present work offers a comprehensive overview of methods related to condition assessment of bridges through Structural Health Monitoring (SHM) procedures, with a particular interest on aspects of seismic assessment. Established techniques pertaining to different levels of the SHM hierarchy, reflecting increasing detail and complexity, are first outlined. A significant portion of this review work is then devoted to the overview of computational intelligence schemes across various aspects of bridge condition assessment, including sensor placement and health tracking 19.

6.3.7 Electric Cable Insulator Damage Monitoring by Lasso Regression

Participants: Qinghua Zhang.

Since the discovery of electricity, electric cables have become ubiquitous in human constructions, from machines to buildings. Insulators play a crucial role in ensuring the proper functioning of these cables, so it is important to monitor their possible damage, which can be caused by environmental contamination, severe temperature variations, and electrical and mechanical stress. While shunt conductance is a direct health indicator of cable insulation, measuring the cable average shunt conductance is not sufficient for the detection of localized insulator damage, since localized conductance variations are diluted over a long cable length in such measurements. The objective of this work is to assess the feasibility of reflectometry techniques for the monitoring of insulator damage in electric cables. To this end, the estimation of localized conductance variations is investigated based on electrical measurements made at one end of a cable. To avoid estimating a large number of discretized conductance values along a long cable, the proposed method relies on sparse regression, which automatically focuses on localized conductance variations at unknown positions caused by accidental insulator damage. Numerical simulations show the potential of this method for fast estimation of localized shunt conductance variations 30.

6.3.8 Interval Estimation for Time-Varying Descriptor Systems

Participants: Qinghua Zhang.

We investigate the state interval estimation problem for discrete-time linear time-varying descriptor systems subject to unknown but bounded system uncertainties. We propose a zonotope-based interval estimation method in an optimization framework. First, we present a novel zonotope-based interval estimator structure, in which the estimated interval bounds of each state component have design parameters independent of those of the other state components. Then the widths of the jointly estimated intervals enclosing every state component are simultaneously minimized by solving parallel L1 optimization problems via linear programming. Finally, a simulation study shows the effectiveness and higher accuracy of the proposed method compared with existing methods. The work is published in 29.

6.3.9 Time-Delay Estimation for Roadway Survey by Ground Penetrating Radar

Participants: Vincent Baltazart.

Time-delay estimation (TDE) using ground penetrating radar (GPR) is of great importance in roadway surveys. We have proposed a general overview of GPR in road engineering over the past 20 years in 64, and describe several existing GPR processing methods and antennas configurations for application to control the flexible pavement structures in layer thickness and interlayer debondings. The most common TDE methods require the signal picking in the time domain, which can be manual or automatic. Advanced TDE processings have been developped in the past to improve the time resolution of GPR and to achieve the control of thin pavement layers. It is now combined with ultra wide-band stepped-frequency radar technique to survey interlayer debonding.

Within this domain, the conventional signal processing methods, namely, FFT-based methods, apply uniform sampling strategy for TDE, which require numerous frequency sampling points, leading to lengthy data acquisition time and large data storage, especially for ultra-wideband (UWB) radar. In 27, the co-prime sampling strategy is developped as an alternative solution to control thin pavement layers by exploiting off-grid sparse Bayesian learning (OGSBL), referred as co-prime-OGSBL. In our scheme, the sampling rate of GPR signals with co-prime sampling strategy is then greatly reduced compared with the uniformsampling, which therefore reduces the data acquisition burden and computational complexity. Both simulation and experimental results demonstrate the efficiency and accuracy of the proposed method in the estimation of time-delays and thickness. Most of these works rely on the existing experimental data base which have been previously collected at UGE 's fatigue carousel and to the ANR Acimp project.

6.4.1 State reconstruction for stochastic nonlinear systems with unknown local nonlinearities via output injection

Participants: Neha Aswal, Adrien Melot, Qinghua Zhang, Laurent Mevel.

This work addresses state estimation for dynamical systems involving localized unknown nonlinearities. Direct application of linear state estimation techniques, e.g., the Kalman filter, would yield erroneous state estimates. Existing approaches in the literature either assume or estimate the nonlinearities. Alternatively, the present paper proposes to reject the unknown nonlinearities as if they were unknown disturbances. By applying an existing disturbance rejection technique, the need to know or to estimate the nonlinearities is avoided. The efficiency of the proposed method is demonstrated through numerical simulations on a nonlinear mechanical system 31.

6.4.2 Rejection of parametric model errors in a recursive Kalman filter for virtual sensing applications

Participants: Michael Doehler.

The performance of the Kalman filter is often hindered by the discrepancies between a model used to realize the filter and the true model of the data-generating system. While some methods to account for those errors exists, the majority is restricted to Luenberger's observers. The objective of this work is to develop a Kalman filter where the effect of parametric model errors is rejected from the state and observation equations before the development of the state estimate, and without the explicit knowledge of the error terms. For this purpose, errors in a physical parameter of a parametrized mechanical system are modelled as additive terms in the state and observation equations by means of first-order perturbation analysis. The rejection step is achieved with an injection of a shaped output nullifying the effect of the added terms. A state observer is then derived under the assumption that the injected shaped outputs and the system states are uncorrelated. The performance of the proposed approach is illustrated on experimental data of a plate system 38.

6.4.3 Mitigating ill-posedness in parameter estimation under sparse measurement for linear time-varying systems employing virtual sensor responses

Participants: Qinghua Zhang, Laurent Mevel.

Monitoring linear time-varying (LTV) systems using model-based approaches typically requires dense instrumentation and high- fidelity support models, resulting in significant computational and financial costs. Bayesian filtering-based methods adopt a joint state-parameter estimation approach for LTV system monitoring wherein states/parameters are observed as well as inferred from the measurements. Eventually, sparsity in measurement while dealing with high-fidelity models can aggravate the ill-posedness in the estimation diverging the estimation to impractical or no solutions. To enhance estimation resolution and precision, an alternative approach can be supporting estimation with virtual sensor measurements sampled from future time steps. To leverage virtual measurements, the estimation needs to be time-delayed, enabling the observation of states/parameters through measurements taken at both current and subsequent time steps, thereby alleviating the ill-posedness. This method has been explored for an LTV spring-mass-dashpot system, where the numerical investigation utilizes an interacting filtering environment to estimate states and parameters in the presence of sparse measurements. The study has shown how incorporating additional information can stabilize the estimation process, leading to improved estimates for both states and parameters 55, 56.

6.4.4 Fault Diagnosability Analysis of Multi-Mode Systems

Participants: Mathias Malandain.

Multi-mode systems can operate in different modes, leading to large numbers of different dynamics. Consequently, applying traditional structural diagnostics to such systems is often untractable. To address this challenge, we present a multi-mode diagnostics algorithm that relies on a multi-mode extension of the Dulmage-Mendelsohn decomposition. We introduce two methodologies for modeling faults, either as signals or as Boolean variables, and apply them to a modular switched battery system in order to demonstrate their effectiveness and discuss their respective advantages 39.

6.4.5 Fast dynamic analysis of damaged 1D periodic waveguides

Participants: Alvaro Camilo Gavilan Rojas, Qinghua Zhang, Christophe Droz.

We extend the Bloch wave-based reduced order models in the Wave Finite Element Method (WFEM) framework for fast wave-damage interaction analysis. It aims at fault detection and diagnosis of periodic structures. A finned tube heat exchanger, which can be seen as a 1D system, is used as a numerical application. Reduced model results and performance are compared to a standard WFEM model. Diffusion curves are obtained more than a hundred times faster with the proposed scheme, moving toward massive generation of wave-damage scenarios and indicators to perform damage detection 54.

6.4.6 Use of bifurcation curves for system identification

Participants: Adrien Melot.

We propose a methodology to carry out nonlinear system identification based on bifurcations. The proposed approach relies on fold bifurcation curves identified experimentally through control-based continuation and an optimization framework to minimize the distance between the experimental and numerical curves computed with bifurcation tracking analyses. The approach is demonstrated on a nonlinear model of base-excited energy harvester with magnetic force nonlinearity 43, 44.

6.5.1 Predictive probability of detection curves based on data from undamaged structures

Participants: Michael Doehler.

We develop a model-assisted approach for determining predictive probability of detection curves. The approach is “model-assisted,” as damage-sensitive features are evaluated in combination with a numerical model of the examined structure. It is “predictive” in the sense that probability of detection (POD) curves can be constructed based on measurement records from the undamaged structure, avoiding any destructive tests. The approach can be applied to a wide range of damage-sensitive features in structural health monitoring and non-destructive testing, provided the statistical distribution of the features can be approximated by a normal distribution. In particular, it is suitable for global vibration-based features, such as modal parameters, and evaluates changes in local structural components, for example, changes in material properties, cross-sectional values, prestressing forces, and support conditions. The approach explicitly considers the statistical uncertainties of the features due to measurement noise, unknown excitation, or other noise sources. Moreover, through confidence intervals, it considers model-based uncertainties due to uncertain structural parameters and a possible mismatch between the modeled and the real structure. Experimental studies based on a laboratory beam structure demonstrate that the approach can predict the POD before damage occurs. Ultimately, several ways to utilize predictive POD curves are discussed, for example, for the evaluation of the most suitable measurement equipment, for quality control, for feature selection, or sensor placement optimization 24.

6.5.2 Predictive Probability of Localization Curves (P-POL) for Structures with Changing Environmental Conditions

Participants: Michael Doehler.

In this paper, an approach to assess the detectability of damages is applied to strain and inclination measurements for the first time, demonstrating the universality of the method. Moreover, it is analyzed how changes in the environmental and operational variables (EOVs) affect the predicted POD. The results demonstrate that the predicted POD are valid even in the presence of environmental changes, but they depend on the user-defined hyperparameters of the algorithms that remove the environmental effects from the measurements. Therefore, the hyperparameter selection is critically discussed and an optimal monitoring strategy is outlined 47.

SNCF: Hot boxes detection

Participants: Jean Dumoulin, Thibaud Toullier, Boualem Merainani.

The main strategic issue is the maintenance in operational condition of the Hot Box Detectors (DBC). The removal of the DBC from the track is part of Tech4Rail's ambition: reducing equipment to the track. The innovation aimed at in this project is to study and develop a measurement solution to be deployed at the edge of a lane out of danger zone and independent of track equipment. Among the scientific obstacles identified are the following three:

• the behavior of the measurement system in deteriorated meteorological conditions in a real site,
• the design and implementation of an automated prototype for in-situ deployment (connection to an existing announcement system, hardware packaging of the system, study and design of a scalable software solution allowing pre-processing data),
• the development of automatic processing tools for the analysis of massive data generated by in-situ measurement systems.

Siemens: Proof of concept monitoring coupled with prediction model for de-icing metro lane surface

Participants: Jean Dumoulin, Thibaud Toullier, Mathias Malandain.

A proof of concept study aims at combining real site monitoring solutions with adjoint state FE thermal model approach to predict optimal heating required to preserve surface from icing in winter conditions. Furthermore, we introduced in our prediction model connection with in-line weather forecast provided by Meteo France Geoservice at different time horizons and spatial scales. Total amount: 124 k€.

CETIM: ultrasonic wave monitoring

Participants: Vincent Le Cam, Arthur Bouché.

Two expertise and training sessions were held on the GERONIMO solution developed jointly by CEA and UGE (also called Ondula for railways monitoring or Ondulys for concrete monitoring), with the purpose of application to emit / receive appropriate ultrasonic waves for detection and localization of defects by Acoustic Emission in pipe structures.

CETIM: fiber optic monitoring

Participants: Xavier Chapeleau.

CETIM is conducting fatigue testing on a tank until it bursts, using existing fiber optic sensors based on Bragg gratings for deformation measurement points. This expertise project aims to enhance this setup with additional fiber optic sensors for distributed deformation measurements in a testing campaign, with the goal to compare the effectiveness of these two monitoring technologies. Total amount: 20k€

CEA List ONDULA2 / Alstom

Participants: Vincent Le Cam.

With CEA-LIST and Alstom-Rail, this project (until 2024) focuses on NDT ultrasonic testing methods for rails. The goal is to deploy several complete rail-sensors in real railway application test benches; another aspect consists in transferring the common knowledge to the final customer Alstom. A daughter board for high frequency ultrasonic emission/reception has been successfully developed and licensed in three industrial transfers.

Sercel: Vibration monitoring

Participants: Michael Doehler, Laurent Mevel.

With the goal of providing a complete SHM system for vibration monitoring with their high-end sensors, we have transferred modal analysis and damage detection algorithms in a technology transfer in three contracts to Sercel, involving technical development and support (2020–2023).

Besides the transfer, an ANR France Relance project with Sercel has been accomplished (2022–2024). Furthermore, several meetings with Sercel have happened to define joint future work, with the objective to launch a "contrat cadre" for research on SHM applications.

Hottinger Brüel & Kjær (HBK): uncertainty quantification for frequency-domain modal analysis

Participants: Michael Doehler, Mikkel Steffensen.

In the context of the PhD of Mikkel Steffensen (DTU Denmark / HBK), a research collaboration with HBK has started on developping methods for uncertainty quantification for input/output frequency-domain modal analysis. Mikkel has spent one month at Inria in 2024 for joint work on the subject.

8.1.1 Inria associate team not involved in an IIL or an international program

8.1.2 Participation in other International Programs

8.2.1 Visits of international scientists

8.2.2 Visits to international teams

8.3.1 Horizon Europe

MTE DGITM CASC: Acoustic Wave for Wirebreak in cables Monitoring

Participants: Vincent Le Cam.

This governmental project aims at testing new algorithms in the CASC platform for detecting and localizing wire breaks in cables of suspension bridges by means of acoustic waves time difference of arrival (TDOA), with the objective to provide a better "time of arrival" time-stamping (by means of the maximmum of likelihood for instance). Another objective is the implementation of a good time-synchronization in wireless sensors while keeping the GPS-energy lower as possible. This was done in the context of the PhD of D. Pallier. A demonstration of acoustic sensors for bridge cable monitoring has been set up, and works for qualification carried out. The project has ended with its final report in 2024.

ANR PRC SWEAT-City

Participants: Romain Noel, Jean Dumoulin.

• Duration: 2024 – 2028
• Budget: 409 k€
• Title: Simulation of Water Evaporation within Artificial ground for Thermo-regulation of the City
• Abstract: The global warming and the more extreme events related implies that cities will be concerned by Urban Heat Island (UHI) effect more often and more intensively. Pavements cover between 30% and 40% of city areas and have a strong effect on the UHI. Studies are showing that two major phenomena can be used and then must be studied to reduce the effect of pavements on UHI: The albedo and the evaporation of water. The increase in albedo has a beneficial effect on the surface temperature of pavements but increases radiation on the vertical surfaces of the city. The present proposal focuses on the effect of water evaporation on UHI. Research on that topic is increasing in the recent years, and only few papers are available on the numerical simulation. However, the evaporation of porous media is a complex phenomenon by its geometry, its interactions between the matrix and fluids, the phase change etc. This complexity leads to macroscopic models with numerous parameters that are hard to obtain experimentally and to optimize. The aim of the project is to develop a heat and mass numerical model that considers evaporation in a construction material. The model is based on a multiscale approach combining the ability of Lattice Bolzman Method at the pore scale and Finite Element at the macro scale. Experiments will be carried out at different scales to validate the modelings. Finally experiments in simulated real life situation within the Sense City facility will be performed and simulated in order to validate the models.

ANR JCJC Archi-Noise

Participants: Christophe Droz.

• Duration: 2024 – 2028
• Budget: 286 k€
• Title: Architected materials with meso-scale interactions for Noise and vibration control
• Abstract: Noise, vibration, and harshness (NVH) impact multiple industries, affecting health, system longevity, and sustainability. Despite advances in NVH mitigation materials, performance gains are plateauing due to constraints like cost, compactness, adaptability, and structural integrity. ArchiNoise seeks to redefine material design by exploring architected meta-structures at meso- and macro-scales, surpassing traditional vibro-acoustic limits. It will adapt nano-scale physics and electromagnetism concepts to structural engineering, creating scattering effects akin to Bragg and locally resonant bandgaps, but independent of periodic unit-cell dimensions or oscillator mass. These novel waveguiding phenomena will target broadband NVH control. ArchiNoise will develop theoretical, computational, and phenomenological tools to design and optimize these materials. By integrating enriched continuum theories, vibroacoustics, inverse wave-based identification, and lattice-based periodic modeling, it will pioneer lightweight NVH solutions.

ANR France 2030 ExcellenceS City-FAB / CD 92

Participants: Jean Dumoulin, Thibaud Toullier, Mathias Malandain.

• Duration: 2024 – 2028
• Partners: CD 92, UGE laboratories
• Budget: 600 k€, 80 k€ for the team
• Title: Analysis of uses, and study of comfort and urban atmosphere on an avenue scale
• Abstract: The objective is to anticipate and adapt road redevelopment projects by aiming at better sharing of mobility spaces, making travel safer and enhancing the environment. These objectives meet the issues of sustainable cities and territories. This project focuses on the environmental effects of developments, in particular concerning thermal comfort, air quality and acoustic comfort. Our contribution to this project focuses on in-situ monitoring and data-driven studies.

ANR SCaNING

Participants: Vincent Le Cam.

• Duration: 2021 – 2025
• Partners: UGE (Coordinator), Université de Toulouse, Aix-Marseille Université, Université de Bordeaux, Andra, EDF
• Inria contact: Vincent Le Cam
• Abstract: Using embedded sensors which will provide information similar to that used in NDE while allowing to continuously evaluate performance indicators (compressive strength and Young’s modulus) and the concrete conditions (porosity and water content) to improve indicator reliability and optimize diagnosis and communicating sensors through fully autonomous, low-power networks makes it possible to consider systems with low installation and operation costs. The project is lead by MAST LAMES laboratory of UGE. The full instrumentation part is ensured by I4S common team.

ANR Convinces

Participants: Jean Dumoulin, Romain Noël.

• Duration: 11/2021 – 10/2025
• Partners: Univ. Lorraine (coordinator), CERTES (UPEC), Univ. Strasbourg, UGE, Cerema.
• Abstract: The ANR project CONVINCES is investigating the influence of convection in suspensions of micro-encapsulated phase change material (mPCM) in urban civil engineering applications. This project will include LBM (Lattice Boltzmann Method) and DEM (Discrete Element Method) in multi-scale simulations plus series of experiments at different scales to study the thermal impact of such mPCM suspensions in porous media. The final objective is the thermal regulation of pavements.

ANR RESBIOBAT

Participants: Jean Dumoulin.

• Duration: 01/2022 – 12/2025
• Partners: UGE (coordinator), CERTES (UPEC), LNE, CSTB, Cerema, Themacs Ingénierie.
• Abstract: The ANR project RESBIOBAT addresses energy and environmental issues. Major advances are expected in the building sector. Reliable in-situ thermal characterization of buildings before and after a renovation action are required. Moreover, construction must be more "sustainable", notably by using bio-sourced materials and raw earth. In this project, we propose an inter-disciplinary technical solution combining modeling, simulations and measurements for a better in-situ evaluation of the energy performances of conventional and sustainable walls. The identification of the thermal characteristics will be performed by an inverse method combining a hygro-thermal model solved in real time by a "reduced bases" technique and sensors selected by "optimal experimental design". After a robustness study via virtual tests, a prototype will be realized and tested on real walls in laboratory and in the Equipment of Excellence Sense-City.

ANR France Relance: Sercel

Participants: Johann Priou, Michael Doehler, Laurent Mevel.

• Duration: 02/2022 – 01/2024
• Partner: Sercel
• Abstract: The objectives of this Action 4 France Relance project are the development of automated and robust algorithms for operational modal analysis under environmental variations.

PIA4: MINERVE

Participants: Vincent Le Cam.

• Duration: 2022–2027
• 22 partners, coordinator: SNCF. Budget: 40 M€, 743 k€ for the team
• Title: Méthodes et outils pour la collaboration sectorielle et la continuité numériuqe sur le cycle de vie (MINERVE)

• Abstract: The six main objectives of the MINERVE project are: - Develop design and construction methods and tools using effective BIM approaches for each business - Anticipate and optimize the construction phase, based on sustainable BIM (digital continuity, frugality of models) - Developing digital twins (exploring the potential of AI for decision support), using opportunities with regard to biodiversity and the environment - Use the digital twin to improve resilience to climate change - Develop an industrializable, standardized and shared vision of interfaces ensuring digital continuity via the BIM model on all phases - Build a collaborative ecosystem around the modeling of linear and particularly railway infrastructure

  The team participates with BIM and monitoring of railway structures by modeling vibrations, defining original ways of operational monitoring including fiber optic sensors.

PIA4: DIAM

Participants: Vincent Le Cam.

• Duration: 2022–2026
• Partners: STIMIO (coordinator), SNIC, UGE. Budget: 3 M€, 693 k€ for the team.
• Abstract: In this project, new ways to diagnose infrastructure deterioration are identified through the use of innovative instrumentation and by merging different data sources. With focus on railway monitoring, the goal is online diagnosis communication of critical trackside elements, and to enrich trackside elements with augmented infrastructure monitoring systems. New algorithms and models for predictive maintenance are developed.

CETIM

Participants: Michael Doehler, Xavier Chapeleau.

• Duration: 09/2024–08/2025
• Partners: CETIM, I4S, UGE/MAST-SMC. Budget: 100 k€
• Abstract: This research collaboration funded by CETIM aims at developing a thesis project focused on data fusion and AI using data from SHM sensors to create predictive models for fatigue damage (initiation and propagation of cracks) in welded structures. This preliminary study focuses on evaluating SHM sensors and the processing and fusion of data related to the considered use case.

CEA

Participants: Romain Noel.

• Partners: CEA/DM2S/STMF.
• Abstract: Within the Inria/CEA collaborative framework, I4S and the LMSF started to work together on CFD methods. This collaboration led to a first M2 internship, and the collaboration continues through the PhD project of Clément Bardet (2024–2027) on the use of thermo-chemical potential in LBM.

IFPEN

Participants: Laurent Mevel.

Collaboration with IFPEN leading to the thesis of A. Cadoret on applying OMA techniques on wind turbines, and a new PhD project has started with PhD candidate N. Delette (2023–2026).

AIS Rennes

Participants: Christophe Droz.

The city of Rennes has allocated 10k€ to C. Droz to facilitate his installation and engage collaborations (2022–2024).

PULSAR jeunes chercheurs Pays de la Loire

Participants: Romain Noel.

The region Pays de la Loire, has allocated 48k€ to R. Noël to his project on numerical simulations of phase change material for thermal regulation of cities (2022–2024).

9.1.1 Scientific events: organisation

9.1.2 Scientific events: selection

9.1.3 Journal

9.1.4 Invited talks

• Christophe Droz
  • A Wave Finite Element Framework for model-based guided wave testing, Mandi, India, 25/09/2024.
• Vincent Le Cam
  • introduction keynote on SHM development in France at the 7th SHM@COFREND day in Bordeaux, 20/03/2024
• Michael Doehler
  • “Statistical uncertainties in vibration-based system identification”, keynote lecture at Digital Bridge congress, Munich, Germany, 17/10/2024
• Jean Dumoulin
  • “Ultra Time Domain Infrared Thermography in outdoor monitoring" invited talk at WCNDT20, 20th World Conference on Non-Destructive Testing, May 2024, Incheon, Corée du Sud.
  • “Thermal monitoring of structures and systems by Infrared thermography in outdoor conditions", Short course Lecture, QIRT 2024, July 2024, Zagreb, Croatie
  • “Monitoring of transport infrastructures using ground-based infrared thermography: results, progress, challenges, and perspectives" invited talk at SPIE Optics and Photonics, Conference 13144: Infrared Remote Sensing and Instrumentation XXXII, August 2024, San Diego, USA.

9.1.5 Leadership within the scientific community

• Vincent Le Cam
  • co-chair of SHM@COFREND: this activity branch of the COFREND (French Confederation for Non-destructive Testing) aims at uniting the national SHM community from academia and industry, and to promote and standardize the SHM sector in France.
  • member of the scientific council of WEN (West Electronic Network) since 2014, which is a cluster of about 200 companies, academics and research laboratories active in electronics
• Michael Doehler
  • co-leader of the working group “GT SHM Data” within SHM@COFREND. The GT focuses on scientific issues, technical challenges and standards of the SHM sector related to handling and processing of structural monitoring data. The GT involves around 50 people from academia and industry in France.

9.1.6 Scientific expertise

• Christophe Droz was scientific expert for NWO Netherlands.

9.1.7 Research administration

• Laurent Mevel
  • deputy head of science of Inria Rennes
  • member of Commisision d'Evaluation at Inria
• Vincent Le Cam
  • deputy co-head of COSYS department at Université Gustave Eiffel
• Jean Dumoulin
  • member of Commisision d'Evaluation des chercheurs du Ministère de la Transition Ecologique (MTE)
• Xavier Chapeleau
  • member of Commisision d'Evaluation des chercheurs du MTE

9.2.1 Teaching

• Jean Dumoulin
  • Licence Professionnelle TAM (Techniques Avancées en Maintenance): thermographie infrarouge active, 30h, Université Paris-Est Créteil (UPEC), France
  • Master 2 ITII, BTP, module Maintenance et réhabilitation des ouvrages, Transferts thermiques dans les Structures : Des principes physiques à l'application sur site réel, 12 h, Ecole Centrale de Nantes (ECN), France.
• Vincent Le Cam
  • Master Electrical Engineering (GEII), 3h CM in M1, 4h CM in M2 on electronic systems and Structural Monitoring, Université Bretagne Sud, Lorient, France
  • M2 ENSIM Le Mans, 5h CM (monitoring des structures par capteurs sans fils)
  • EC Nantes, 4h CM + 32h TP (electronique embarquée, Linux et drivers)
  • Ecole d'Ingénieur Builders, 10h CM, Caen
• Xavier Chapeleau
  • M1 ITI, fibre-optique, 8h, IUT Nantes
  • "Monitoring and auscultation : Optical fiber sensors", 2h, Infrastar Training School, October 2024
• Romain Noel
  • École des Mines de Saint-Étienne, Master 2, Fluid Mechanics Applications: Plenary conference (2h)
  • École des Mines de Saint-Étienne, Master 1, Advanced Fluid Mecanics: Lectures (6h) + practical lessons (6h)
  • École Centrale de Nantes, Master 2, Data Signal and Image: Lectures (4h) + practical lessons (4h)
• Christophe Droz
  • MSc 1, Modélisation en Action, Dpt. Mathématiques, Université de Rennes (50h)
  • MSc 2, Contrôle non destructif et identification, Dpt. Sciences pour la matière, Université de Rennes (4h)
• Alvaro Camilo Gavilan Rojas
  • Master 2, Mécanique et Matériaux, Contrôle non destructif et identification, 1.5h CM + 7.5h TD, Université de Rennes
• Lucas Rouhi
  • L1 SVE, Maths pour la biologie 1, 15h TD, Université de Rennes

9.2.2 Supervision

Preview not available.

9.2.3 Juries

• Christophe Droz
  • PhD Defense Examinator - Institut national des sciences appliquées. Diego Salam Claro, "Wave-based numerical approaches for non-destructive testing of structural assemblies involving straight waveguides and curved joints". Defense May 2024.
  • CSI member of PhD candidate Theo Gherdaoui, "Nonlinear small-time controllability results for multi-controlled systems", ENS Rennes, since 09/2022.
• Laurent Mevel , Qinghua Zhang
  • HDR Defense Examinators - Université de Rennes. Christophe Droz, "A Unified Wave Finite Element Modeling and Identification Framework for Periodic Structures: Towards Metamaterial Twins". Defense December 2 2024.
• Vincent Le Cam
  • PhD Defense Examinator, Université de Nantes, Walid Askri, "Générateurs flexibles hybrides piézo/tribo électrique pour l’auto-alimentation de capteurs communicants", defense 18/12/2024, and CSI member.

9.3.1 Specific official responsibilities in science outreach structures

• Christophe Droz is co-organizator of the Sci-Rennes seminar series at the Inria center of the University of Rennes (since Sep. 2022).

9.3.2 Productions (articles, videos, podcasts, serious games, ...)

• Christophe Droz gave three interviews in regional or national magazines. Articles have been published in:
  • Inria Emergences (May, 2024, "A New Concept for Acoustic Insulation in Transportation")
  • Sciences Ouest (June, 2024, "Atoms, Noise, and Waves: An Acoustic Challenge")
  • Usine Nouvelle (October, 2024, "Archi-Noise: towards low-frequency metamaterials")

9.3.3 Others science outreach relevant activities

• Michael Doehler has given a presentation at the Sci-Rennes seminar series at the Inria center of the University of Rennes (01/2024).
• Vincent Le Cam organizes the stay of high school students for internships on the UGE campus in Nantes.

International journals

• 16 articleM.Marina Al Bacha, M. L.Mai Lan Nguyen, P.Pierre Hornych, O.Olivier Chupin, J.Juliette Blanc and X.Xavier Chapeleau. Comparison of surface layers responses under heavy truck loadings on new and rehabilitated pavement with different interface bond conditions.Canadian journal of civil engineeringOctober 2024HALDOIback to text
• 17 articleA.Ambroise Cadoret, E. D.Enora Denimal Goy, J.-M.Jean-Marc Leroy, J.-L.Jean-Lou Pfister and L.Laurent Mevel. Linear time periodic system approximation based on Floquet and Fourier transformations for operational modal analysis and damage detection of wind turbine.Mechanical Systems and Signal Processing212April 2024, 111157HALDOIback to text
• 18 articleA.Alvaro Gavilán Rojas, Q.Qinghua Zhang and C.Christophe Droz. A computationally efficient k(ω)-spectral form for partial dispersion analyses within the wave finite element framework.Journal of Sound and Vibration593December 2024, 118652HALDOIback to text
• 19 articleC.Christos Karakostas, G.Giuseppe Quaranta, E.Eleni Chatzi, A. C.Abdullah Can Zülfikar, O.Oğuzhan Çetindemir, G.Guido de Roeck, M.Michael Döhler, M. P.Maria Pina Limongelli, G.Geert Lombaert, N. M.Nurdan Memişoğlu Apaydın, V.Vikram Pakrashi, C.Costas Papadimitriou and A.Ali Yeşilyurt. Seismic assessment of bridges through structural health monitoring: a state-of-the-art review.Bulletin of Earthquake Engineering2232024, 1309-1357HALDOIback to text
• 20 articleÉ.Étienne Lac, G.Guy de Spiegeleer, A.Adrien Delsalle, F.Frédéric Collonval, D.-T.Duc-Trung Lê and M.Mathias Malandain. CoSApp: a Python library to create, simulate and design complex systems.Journal of Open Source Software994February 2024, 6292HALDOIback to text
• 21 articleP.Paulina Leiva-Padilla, X.Xavier Chapeleau, M.-L.Mai-Lan Nguyen, J.Juliette Blanc, S.Samir Allam, E.Emmanuel Loison and P.Pierre Hornych. Use of Distributed Fiber Optic Sensors for the Monitoring of an Accelerated Pavement Test.Transportation Research Record267810March 2024, 131-146HALDOIback to text
• 22 articleA.Adrien Mélot, E.Enora Denimal and L.Ludovic Renson. Control of isolated response curves through optimization of codimension-1 singularities.Computers & StructuresApril 2024, 1-19HALDOIback to text
• 23 articleA.Adrien Mélot, E.Enora Denimal and L.Ludovic Renson. Multi-parametric optimization for controlling bifurcation structures.Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences2024, 1-22HALDOIback to text
• 24 articleA.Alexander Mendler, M.Michael Döhler and C. U.Christian U Grosse. Predictive probability of detection curves based on data from undamaged structures.Structural Health Monitoring2332024, 1725-1741HALDOIback to text
• 25 articleB.Boualem Merainani, B.Bian Xiong, V.Vincent Baltazart, M.Michael Döhler, J.Jean Dumoulin and Q.Qinghua Zhang. Subspace-based modal identification and uncertainty quantification from video image flows.Journal of Sound and Vibration569January 2024, 117957HALDOIback to textback to text
• 26 articleR.Romain Noël, A.Alexis Renier-Robin and L.Laurent Navarro. Generalized Dithering using the Lattice Boltzmann Method.Signal, Image and Video Processing1812September 2024, 8507-8523HALDOIback to text
• 27 articleJ.Jingjing Pan, H.Huimin Pan, M.Meng Sun, Y.Yide Wang, V.Vincent Baltazart, X.Xudong Dong, J.Jun Zhao, X.Xiaofei Zhang and H. C.Hing Cheung So. Co-prime Sampling based Time-Delay Estimation for Roadway Survey by Ground Penetrating Radar via Off-Grid Sparse Bayesian Learning.IEEE Transactions on Radar Systems2September 2024, 966-978HALDOIback to text
• 28 articleT.Thibaud Toullier and J.Jean Dumoulin. Bias and bottlenecks study in outdoor long term thermal monitoring by infrared thermography: Leveraging opportunistic data for temperature estimation.Infrared Physics and Technology141September 2024, 105471HALDOIback to text
• 29 articleZ.Zhenhua Wang, Y.Youdao Ma, Q.Qinghua Zhang, W.Wentao Tang and Y.Yi Shen. Interval Estimation for Time-Varying Descriptor Systems via Simultaneous Optimizations of Multiple Interval Widths.IEEE Transactions on Systems, Man, and Cybernetics: Systems546June 2024, 3774-3782HALDOIback to text
• 30 articleQ.Qinghua Zhang and M.Monssef Drissi-Habti. Electric Cable Insulator Damage Monitoring by Lasso Regression.Machines121January 2024, 50HALDOIback to text

International peer-reviewed conferences

• 31 inproceedingsN.Neha Aswal, A.Adrien Mélot, L.Laurent Mevel and Q.Qinghua Zhang. State reconstruction for stochastic nonlinear systems with unknown local nonlinearities via output injection.IFAC-PapersOnLineMICNON 2024 - 4th IFAC Conference of Modelling, Identification and Control of Nonlinear Systems5821Lyon, France2024, 256-261HALDOIback to text
• 32 inproceedingsA.Alex Coiret and V.Vincent Le Cam. Applicability of front-wheel braking procedure to evaluate road available grip.CETRA 2024 - 8th International Conference on Road and Rail InfrastructureCavtat - Dubrovnik, Croatia2024, 1-8HALDOI
• 33 inproceedingsN.Noémie Delaplanque, S.Sylvain Chataigner, M.Marc Quiertant, K.Karim Benzarti, A.Arnaud Rolland, X.Xavier Bourbon, X.Xavier Chapeleau, L.Ludwig Battais and L.Laurent Gaillet. Investigations on the ageing of GFRP rebar-concrete bond under sustained load for a high strength concrete.Procedia Structural IntegritySMAR 2024 – 7th International Conference on Smart Monitoring, Assessment and Rehabilitation of Civil Structures64SMAR 2024 – 7th International Conference on Smart Monitoring, Assessment and Rehabilitation of Civil StructuresSalerno, ItalyNovember 2024, 1492-1499HALDOIback to text
• 34 inproceedingsJ.Jean Dumoulin, B.Boualem Merainani and T.Thibaud Toullier. Study of hot box detection on moving targets using way side thermal infrared camera and image processing methods : application to railway infrastuctures.European Geosciences UnionVienne, AustriaMarch 2024, 1-2HALDOIback to text
• 35 inproceedingsA.Arij Fawaz, E.Emilie Lepretre, X.Xavier Chapeleau and S.Sylvain Chataigner. Development and Validation of a Creep Frame Adapted for ELS (End Load Split) Test.Procedia Structural IntegritySMAR 2024 – 7th International Conference on Smart Monitoring, Assessment and Rehabilitation of Civil Structures64Salerno, ItalyElsevier2024, 89-96HALDOIback to text
• 36 inproceedingsE.Eric Gennesseaux, D.Domenico Vizzari, J.Jean Dumoulin, E.Emmanuel Chailleux, S.Stéphane Lavaud, J.-L.Jean-Luc Manceau and T.Thierry Sedran. Laboratory and full-scale experiment of a novel hybrid system to harvest energy through concrete pavement.ICCP 2024 - 13th International Conference on Concrete PavementsMinneapolis, United States2024HALback to text
• 37 inproceedingsS.Szymon Gres, M.Michael Döhler, V.Vasilis Dertimanis and E.Eleni Chatzi. Data-Driven Identification of Noise Covariances in Kalman Filtering for Virtual Sensing Applications.Proceedings of the 10th International Operational Modal Analysis Conference (IOMAC 2024)IOMAC 2024 - 10th International Operational Modal Analysis Conference515Lecture Notes in Civil EngineeringNaples, ItalySpringer Nature SwitzerlandJune 2024, 375-382HALDOIback to text
• 38 inproceedingsS.Szymon Gres, K. E.Konstantinos E Tatsis, M.Michael Döhler, V.Vasilis Dertimanis and E.Eleni Chatzi. Rejection of parametric model errors in a recursive Kalman filter for virtual sensing applications.EWSHM 2024 - 11th European Workshop on Structural Health Monitoring297Potsdam, GermanyJuly 2024, 1-8HALDOIback to text
• 39 inproceedingsF.Fatemeh Hashemniya, B.Benoît Caillaud, E.Erik Frisk, M.Mattias Krysander and M.Mathias Malandain. Fault Diagnosability Analysis of Multi-Mode Systems.SAFEPROCESS 2024 - 12th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes584Ferrara, ItalyElsevier2024, 210-215HALDOIback to text
• 40 inproceedingsM.Mira Kabbara, X.Xavier Chapeleau, Q.Qinghua Zhang and F.Frédéric Bourquin. Numerical evaluation of strain transfer model for steel-reinforced optical fiber cable embedded in a cylindrical concrete beam with two void inclusions.EPJ Web ConferencesAOP 2024 - 6th International Conference on Applications of Optics and Photonics305Aveiro, PortugalOctober 2024, 00021HALDOIback to text
• 41 inproceedingsZ.Zhilei Luo, B.Boualem Merainani, M.Michael Döhler, V.Vincent Baltazart and Q.Qinghua Zhang. Vision-based operational modal analysis robust to environmental conditions.Proceedings of the 10th International Operational Modal Analysis Conference (IOMAC 2024)IOMAC 2024 - 10th International Operational Modal Analysis ConferenceNaples, Italy2024, 1-8HALDOIback to textback to text
• 42 inproceedingsV.Vincent Mahé, A.Adrien Mélot, B.Benjamin Chouvion and C.Christophe Droz. Computing the dynamic response of a periodic structure coupled with a nonlinear junction using the harmonic balance method and Floquet-Bloch modelling.INTER-NOISE 2024 - 53rd International Congress and Exposition on Noise Control Engineering2706Nantes, FranceOctober 2024, 5622-5628HALDOIback to text
• 43 inproceedingsA.Adrien Mélot, E.Enora Denimal Goy and L.Ludovic Renson. Nonlinear system identification with control-based continuation of bifurcation curves.ENOC 2024 - 11th European Nonlinear Dynamics ConferenceDelft, NetherlandsJuly 2024, 1-2HALback to text
• 44 inproceedingsA.Adrien Mélot, E.Enora Denimal Goy and L.Ludovic Renson. On the use of bifurcation curves for system identification and model updating purposes.ECCOMAS 2024 - 9th European Congress on Computational Methods in Applied Sciences and EngineeringLisbon, PortugalJune 2024, 1-1HALback to text
• 45 inproceedingsA.Adrien Mélot, E.Enora Denimal Goy and L.Ludovic Renson. Structural optimization for controlling isolated response curves.ENOC 2024 - 11th European Nonlinear Dynamics ConferenceDelft, NetherlandsJuly 2024, 1-3HALback to text
• 46 inproceedingsA.Adrien Mélot, E.Emmanuel Rigaud and J.Joël Perret-Liaudet. Robust gear design with respect to the primary resonance induced by backlash nonlinearity.ENOC 2024 - 11th European Nonlinear Dynamics ConferenceDelft, NetherlandsJuly 2024, 1-2HALback to text
• 47 inproceedingsA.Alexander Mendler and M.Michael Döhler. Predictive Probability of Localization Curves (P-POL) for Structures with Changing Environmental Conditions.EWSHM 2024 - 11th European Workshop on Structural Health Monitoring297Potsdam, Germany2024, 1-11HALDOIback to text
• 48 inproceedingsB.Boualem Merainani, T.Thibaud Toullier, S.Sarouthan Sriranjan and J.Jean Dumoulin. Monitoring of moving rail-road cars through infrared thermal vision: Study of hot box detection on reconstructed scenes with and without Deep Learning approaches.QIRT 2024 - 17th Quantitative Infrared Thermography ConferenceZagreb, Croatia2024, 1-2HALback to text
• 49 inproceedingsR.Romain Noël and A.Alex Coiret. Application of the Lattice Boltzmann Method to estimate road capacity decrease depending on lane number and flow density.AIIT 2024 - 4th International Conference, Greening the way forward: sustainable transport infrastructure and systemsRome, Italy2024, 18 p.HAL
• 50 inproceedingsS.Stéphane Paboeuf, M.Maxime Deydier, Q.Quentin Sourisseau, E.Emilie Lepretre, S.Sylvain Chataigner and X.Xavier Chapeleau. Static and Fatigue Strength Assessment of Composite Patch Repair.OMAE 2024 - 43rd International Conference on Ocean, Offshore and Arctic EngineeringSingapour, SingaporeAmerican Society of Mechanical Engineers2024HALDOIback to text
• 51 inproceedingsJ.Johann Priou, S.Szymon Gres, A.Alexander Mendler, M.Matthieu Perrault, L.Laurent Guerineau and M.Michael Döhler. Automated Uncertainty-Based Clustering and Tracking of Modal Parameters Under Strong Variations.Proceedings of the 10th International Operational Modal Analysis Conference (IOMAC 2024)IOMAC 2024 - 10th International Operational Modal Analysis Conference514Lecture Notes in Civil EngineeringNaples, ItalySpringer Nature SwitzerlandJune 2024, 581-588HALDOIback to text
• 52 inproceedingsJ.Johann Priou, S.Szymon Gres, M.Matthieu Perrault, L.Laurent Guerineau, M.Mathieu Desbazeille and M.Michael Döhler. Automated modal parameter tracking with neural network based normalization of environmental perturbations.e-Journal of Nondestructive TestingEWSHM 2024 - 11th European Workshop on Structural Health Monitoring29Proceedings of the 11th European Workshop on Structural Health Monitoring (EWSHM 2024)7Potsdam, GermanyJuly 2024, 1-10HALDOIback to text
• 53 inproceedingsJ.Johann Priou, A.Alexander Mendler, M.Matthieu Perrault, L.Laurent Guerineau and M.Michael Döhler. Normalization of Environmental Effects in Modal Parameter Tracking.Proceedings of the 10th International Operational Modal Analysis Conference (IOMAC 2024)IOMAC 2024 - 10th International Operational Modal Analysis Conference515Lecture Notes in Civil EngineeringNaples, ItalySpringer Nature SwitzerlandJune 2024, 95-102HALDOIback to text
• 54 inproceedingsA. G.Alvaro Gavilán Rojas, Q.Qinghua Zhang, O.Olivier Robin and C.Christophe Droz. Fast dynamic analysis of damaged 1D periodic waveguides.20th IFAC Symposium on System Identification SYSID 2024SysID 2024 - 20th IFAC Symposium on System Identification58IFAC-PapersOnLine15Boston, United StatesElsevier2024, 325-329HALDOIback to text
• 55 inproceedingsO. A.O. A. Shereena, S.Subhamoy Sen and L.Laurent Mevel. Mitigating ill-posedness in parameter estimation under sparse measurement for linear time-varying systems employing virtual sensor responses.Proceedings of the 11th European Workshop on Structural Health Monitoring (EWSHM 2024)EWSHM 2024 - 11th European Workshop on Structural Health Monitoring297Potsdam, GermanyNDT.net2024, 1-11HALDOIback to text
• 56 inproceedingsO. A.O. A. Shereena, S.Subhamoy Sen, Q.Qinghua Zhang and L.Laurent Mevel. A novel lagged estimation framework for sparsely observed systems supplemented with virtual measurements.Proceedings of the 10th International Operational Modal Analysis Conference (IOMAC 2024)IOMAC 2024 - 10th International Operational Modal Analysis Conference514Lecture Notes in Civil EngineeringNaples, ItalySpringer Nature SwitzerlandJune 2024, 581-588HALDOIback to text
• 57 inproceedingsQ.Quentin Sourisseau, E.Emilie Lepretre, X.Xavier Chapeleau, S.Sylvain Chataigner, S.Stéphane Paboeuf and M.Maxime Deydier. Use of equivalent interface samples during fracture mechanics investigations for the design of adhesively bonded composite reinforcements on steel structures.Procedia Structural IntegritySMAR 2024 – 7th International Conference on Smart Monitoring, Assessment and Rehabilitation of Civil Structures64Salerno, ItalyElsevier2024, 893-900HALDOIback to text
• 58 inproceedingsM.Mikkel Steffensen, S.Szymon Gres and M.Michael Döhler. Modal mass estimation from state-space models and frequency response functions.International Operational Modal Analysis ConferenceIOMAC 2024 - 10th International Operational Modal Analysis ConferenceProceedings of the 10th International Operational Modal Analysis ConferenceNaples, ItalySpringer Nature SwitzerlandMay 2024, 1-8HALDOIback to text