Martin Lévesque’s research while affiliated with Polytechnique Montréal and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (179)


X-Ray Tomography-Based Characterization of the Porosity Evolution in Composites Manufactured by Fused Filament Fabrication
  • Article

December 2024

·

19 Reads

Experimental Mechanics

·

·

·

[...]

·

M. Lévesque

Fused filament fabrication delivers composites with incomplete interface bonding prone to delaminate under loading due to the non-isothermal molecular entanglement during deposition. We aim to localize the mesoscale porosity in 3D-printed composites and quantify its volumetric growth under loading to investigate whether incomplete filament adhesion can lead to delamination. We measured the porosity volumic content by X-ray tomography testing. To distinguish between damage nucleated at the crack tip and mesoscale interface delamination, we quantified the local, 3D strain concentration region size at the crack tip by 2D digital image correlation of slice images over orthogonal planes. Through image segmentation, we observed that the mesoscale porosity resulting from the deposition process clustered at the filament interfaces and doubled from roughly 7% to 14% from an applied opening load of 700 N to 1400 N due to the stress concentration at the filament interfaces. Digital image correlation emphasized the strain concentration over a reduced area at the notch, up to the damage nucleation for an applied load of 1400 N, before the sudden brittle failure. The presented contactless characterization technique emphasizes that mesoscale porosity concentrates at the filament interface, which is a critical delamination nucleation site under loading. This fracture mechanism is even more severe for high-performance composites such as carbon fiber reinforced PEEK.



Schematic overview of the automated shot peen forming workflow. The numerical model of the target shape (a) is input into the simulation software (b). The latter computes an optimal peening pattern (c), that must be applied to form the plate into the target shape. The peening pattern is input into the translation software (d), which generates the peening program (e). The peening program includes the nozzle paths and a set of peening parameters that allow the peening robot (f) to reproduce the peening pattern. The peening program is loaded into the peening robot in the form of a programming code. When the treatment is finished, we characterize the curved shape of the plate using a 3D scanner (g) and trace an error map (h), which shows discrepancy between the target and the scanned shapes. The simulation and the translation softwares require experimental calibration, which consists in characterizations of the shot stream (i), of the eigenstrain (j) induced by the peening treatment, and of the eigenstrain anisotropy (k). The shot stream is characterized with flat dummy specimens, the eigenstrain is measured using 76×19\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$76 \times 19$$\end{document} mm aluminum coupons, and the anisotropy is quantified with the help of square aluminum plates. Art by IMPAKT Scientifik
A triangular mesh element and the vectors serving for computation of the local fundamental forms. The vectors v0\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textbf{v}_{0}$$\end{document}, v1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textbf{v}_{1}$$\end{document}, v2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textbf{v}_{2}$$\end{document} define positions of the element vertices in the global coordinate system. The edge vectors e0\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textbf{e}_{0}$$\end{document}, e1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textbf{e}_{1}$$\end{document}, e2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textbf{e}_{2}$$\end{document} are computed as the difference between the corresponding vertex vectors. The vectors n0\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textbf{n}_{0}$$\end{document}, n1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textbf{n}_{1}$$\end{document}, n2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\textbf{n}_{2}$$\end{document} are traced in the center of the corresponding edges and define the local curvature of the mesh. They are normal to the edges, but their angle of inclination with respect to the element constitutes a degree of freedom
Generation of the initial mesh. The mesh of the curved target shape (a) is projected on the flat initial geometry (b) using the LSCM algorithm. The distortion of each element in the projected mesh is then minimized with respect to the target mesh using the L-BFGS algorithm. The resulting mesh (c) is used as the initial mesh during the inverse problem resolution
Two types of nozzle paths generated by the translation software. The “Zigzag” path type (left) implies straight parallel nozzle paths. The “Circular” path type (right) is a set of concentric closed lines. The example pattern implies treatment from the top and bottom sides with the same peening regime
Determination of the dimensionless error Ωz=dH/Δz\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Omega _z = d_H/\Delta z$$\end{document} between the simulated shape and the shape of the treated plate. a) The simulated shape and its bounding box. The deflection Δz\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta z$$\end{document} is computed as the smallest dimension of the bounding box. b) A photo of the plate treated following the computed pattern that leads to the simulated shape. The non-reflective surface is the treated part. c) The meshed shape scanned with a 3D scanner. d) Map of the local distances d between the simulated and the scanned shapes that were optimally aligned. The map is traced on the simulated shape. The Hausdorff distance dH\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$d_H$$\end{document} is the maximal value of local distances d

+11

Simulation and automation of aluminum panel shot peen forming
  • Article
  • Publisher preview available

April 2024

·

55 Reads

·

2 Citations

International Journal of Material Forming

We present a methodology for automated forming of metal plates into freeform shapes using shot peening. The methodology is based on a simulation software that computes the peening pattern and simulates the effect of its application. The pattern generation requires preliminary experimental characterization of the treatment. The treatment is applied by a shot peening robot. The program for the robot is generated automatically according to the peening pattern. We validate the methodology with a series of tests. Namely, we form nine aluminum plates into doubly curved shapes and we also shape model airplane wing skins. The article describes the complete workflow and the experimental results.

View access options





Eigenstrain‐based analysis of why uniformly shot peened aluminium plates bend more in the rolling direction

June 2023

·

20 Reads

·

3 Citations

Strain

Shot peen forming is a process widely used to shape aircraft components such as wing skins, yet its fundamental working is still crudely understood. It is understood that a light conventional peen forming treatment applied uniformly over an initially flat plate will induce isotropic in‐plane stretching of the surface layer and will thus lead to a panel curving with identical curvatures in all directions. However, [1] made the startling observation that uniformly peen formed aluminium plates of different aspect ratios all bent along their laminating direction irrespective of the peening direction. This experimental result is counterintuitive because the residual stresses due to the lamination process are 1 order to 2 orders of magnitude smaller than those induced by the shot peen forming treatment. In the present study, we apply the eigenstrain theory to estimate the effect of the different sources of anisotropy on uniformly peen formed aluminium plates. Potential sources of anisotropy included the plastic anisotropy of rolled aluminium, nonequibiaxial initial stresses that redistribute when their equilibrium is disturbed by peening, the geometry of the specimens and externally applied prestress. For the alloy and peening conditions considered, we show that plastic anisotropy had no discernible influence on the resulting shape of the peen formed specimens. Initial residual stresses, on the other hand, caused slightly larger bending loads in the rolling direction of the alloy. Although the magnitude of these loads was approximately 30 times smaller than peening‐induced loads, it was sufficient to overcome the geometric preference for rectangular sheets to bend along their long side and cause all unconstrained specimens to bend along the rolling direction instead. Our analysis highlights the importance of the history of the material that is being peened. Residual stresses already present in the part before peening must be considered to ensure good simulation predictions.




Citations (74)


... Currently, there are limited studies on continuous fiber non-planar AM using a three-axis printer. On the other hand, a limitation of non-planar AM using a three-axis printer is that the deposition angle is restricted, making it challenging to produce printing layers with pronounced curvature [46]. Therefore, it is necessary to study the manufacturing capability and applicable scope of non-planar AM of continuous fiber composites based on a three-axis printer. ...

Reference:

Non-planar additive manufacturing of pre-impregnated continuous fiber reinforced composites using a three-axis printer
Non-planar material-extrusion additive manufacturing of multifunctional sandwich structures using carbon-reinforced polyetheretherketone (PEEK)
  • Citing Article
  • April 2024

Additive Manufacturing

... Instead, the material was generally considered to have the same properties in these 'bonded regions'. Only some recently published FEA studies using phase-field models for 3D-printed materials actually distinguished between cross-layer or bulk layer failure (i.e., filament or bead failure) and inter-layer or inter-phase failure [50,51] in their formulations. Similar observations were made in a very recent publication by Monaldo and Marfia [25], who, to the best of the authors' knowledge, were the first to explicitly incorporate both the porosity and inter-facial bond properties (i.e., inter-layer and intra-layer bond strengths) simultaneously in numerical (microstructural) models for 3D-printed polymers produced by material extrusion. ...

Phase-field modeling of fracture in fused filament fabricated thermoplastic parts and experimental validation

Engineering Fracture Mechanics

... The computed Poisson's ratio (0.36) is slightly higher than the values reported in the literature for 3D printed short fiber-reinforced composites [27,41]. We compared the experimental results with the elastic properties prediction using the 260 multiscale fast Fourier transform (FFT) homogenization approach developed ad-hoc for additively manufactured composites described in [42]. Table 3 summarizes the average E l , E t , and E 0°−90°r esulting from the dual-scale homogenization of real meso and microstructures inspected by X-ray tomography, with the corresponding 95% confidence intervals. ...

Multiscale Fast Fourier Transform homogenization of additively manufactured fiber reinforced composites from component-wise description of morphology
  • Citing Article
  • September 2023

Composites Science and Technology

... They noted that the effects of plastic anisotropy, initial residual stress distribution, and prestress on curvature anisotropy were substantial. Miao et al. [24] applied inherent strain (or eigenstrain) theory to evaluate the effects of various sources of curvature anisotropy in uniformly peen-formed aluminum plates and showed that the initial residual stress distribution generated by the roll direction had a substantial effect. ...

Eigenstrain‐based analysis of why uniformly shot peened aluminium plates bend more in the rolling direction
  • Citing Article
  • June 2023

Strain

... In the field of fracture mechanics bending testing, the studies from Lampron et al. [11] for FFF-printed single-edge notched bending (SENB) samples made of PLA, from Refat et al. [12] for printed polyetheretherketone (PEEK), from Marsavina et al. [13] for printed PLA and from Nurizada and Kirane [14] for printed ABS SENB samples are particularly noteworthy. The occurrence of different failure modes is, however, not limited to fracture mechanics bending tests; numerous studies [15][16][17][18][19] have also observed varying failure modes in fracture mechanics tensile specimens, such as compact tension (CT) specimens. Overall, this brief literature review demonstrates that both failure mechanisms − crack deflection and penetration − can occur depending on various process-related parameters, such as the layer height or the infill angle/orientation, but none of these studies focused on investigating or explaining this phenomenon in depth. ...

Multiscale characterization of the fracture mechanics of additively manufactured short fiber-reinforced composites
  • Citing Article
  • May 2023

Engineering Fracture Mechanics

... Wielhorski et al. (2022) reviewed various meso-and microscale modeling strategies of dry 3D woven composite reinforcements. Trofimov et al. (2023) reviewed methods for the generation of representative volume elements (RVE) of 3D woven structures at the meso-and microscales with a special focus on the RTM process, i.e., resin injection, curing, cooling/ demolding, and the RTM process. ...

A review on the Representative Volume Element-based multi-scale simulation of 3D woven high performance thermoset composites manufactured using resin transfer molding process
  • Citing Article
  • February 2023

Composites Part A Applied Science and Manufacturing

... The RECT specimen, as the simplest design to mechanically characterise the material, and the DENT specimen, as the needed notched geometry to determine certain fracture parameters, along with an accurate acquisition system, provide reliable results. Furthermore, the obtained elastic properties of 3Dprinted PLA are in line with the literature [79,82,118]. In addition, the calculated K c is remarkably similar to the plane strain mode I fracture toughness of PLA reported in other works [51,82,118], validating the followed procedure. ...

Characterization of the non-isotropic tensile and fracture behavior of unidirectional polylactic acid parts manufactured by material extrusion

Additive Manufacturing

... Honeycomb sandwich composites, featuring light-weight, high-strength, and customizable design, are extensively utilized in critical secondary load-bearing structures of the aerospace industry. These include the leading and trailing edges of wings, control surfaces such as flaps and rudders, elevators, fuselage panels, auxiliary power unit (APU) doors, and radar hoods [1][2][3][4][5][6]. The sandwich construction typically consists of an upper panel, a lower panel, a centrally located honeycomb core, and adhesive layers interspersed between these components. ...

Material extrusion additive manufacturing of multifunctional sandwich panels with load-bearing and acoustic capabilities for aerospace applications
  • Citing Article
  • December 2022

Additive Manufacturing

... Various surrogate models have been utilized [17,18,19], with a notable recent increase in the use of neural networks (NNs) [20,21,22,23,24,25], due to their ability to capture complex, nonlinear relationships in data. When it comes to calibration, most surrogate-based works still use gradient-free optimization with genetic algorithms (GAs) [26,21,24,27], replacing the costly FE simulation with the surrogate's efficient prediction. Others suggest training a separate ML model in the inverse setting, directly predicting FE model input parameters from its outputs [28,29]. ...

Surrogate optimization of a lattice foot orthotic
  • Citing Article
  • November 2022

Computers in Biology and Medicine

... Extensive research has focused on the mechanical characterization of MEX-produced components along the building direction [14], consistently highlighting the significance of process parameter selection on mechanical behavior in this orientation [15][16][17][18][19][20]. The anisotropy observed in MEX components arises from distinct mechanisms: when loaded along the filament deposition direction, mechanical properties closely approximate those of the parent materials, with variations primarily attributable to porosity and differential crystalline phase formation, particularly in semi-crystalline polymers. ...

Additive manufacturing and characterization of high temperature thermoplastic blends for potential aerospace applications
  • Citing Article
  • November 2022

Composites Science and Technology