A machine learning-based surrogate model for efficient wood microstructure generation compatible with a physics-based model is developed. The model is based on the U-Net neural network, a variant of convolutional neural network which, due to its architecture, is suitable for image-to-image transformation, and focuses on the crucial step of the microstructure generation, which is distortion of the wood slices according to the prescribed distortion map. For training the U-Net, a dataset consisting of a variety of wood microstructures slices before and after the distortion is generated, along with the corresponding distortion maps, using the previously developed parametric model. Transfer learning is shown to improve the performance of the U-Net, especially if the dataset of wood slices is small. The best results with transfer learning are obtained if either the whole U-Net is fine-tuned or only the bottleneck block is frozen during the fine-tuning. The surrogate model is a promising tool for generating a large dataset of wood microstructures within the structural parameter ranges it was trained on in a relatively short time compared to the original method. That could in turn be useful for a parametric study and optimization of the physical properties of wood-based materials.
Background: Digital Volume Correlation (DVC) is a powerful experimental technique for quantifying 3D full-field volumetric displacements and strains. In light of its increased adoption in metrological applications, there is a critical need for benchmark datasets to systematically evaluate the performance of various DVC algorithms across different materials, imaging modalities, and deformation scenarios. Objective: Building on the foundations of DVC Challenge 1.0, the DVC Challenge 2.0 initiative aims to create a repository of DVC datasets to enable researchers to validate and refine their DVC algorithms against common benchmarks. This can help in expanding the scope and performance of DVC and foster innovation in volumetric deformation measurement. Methods: DVC Challenge 2.0 compiles a diverse collection of volumetric image sets contributed by the global research community. These datasets encompass different materials, loading conditions, and imaging modalities, including confocal/multiphoton microscopy, X-ray computed tomography (XCT), neutron tomography, and synthetically generated images. These datasets present various metrological challenges, such as complex deformation fields, poor image quality, and anisotropic or sparse speckle patterns. All datasets are published in an open repository, with a uniform image format and a common data framework. Results: The resulting repository provides benchmark datasets for validating and comparing DVC algorithms, facilitating the exploration of DVC capabilities in diverse and challenging scenarios. Conclusion: By promoting collaboration and open data sharing, DVC Challenge 2.0 will drive innovation in volumetric deformation measurement techniques and broaden the impact of DVC. It will also help establish a baseline for comparison of DVC algorithms and codes.
Realistic 3D microstructure models of wood fiber networks (WFNs, e.g., paper, molded fibers, hot-pressed fibers, etc.) are of interest for numerical modeling of mechanical, optical, and other physical properties. One challenge is to numerically describe 3D high-density WFN (HD-WFN) models with complex fiber shapes without fiber overlapping. An efficient method is proposed to generate 3D HD-WFN microstructures with porosities as low as 21.5% while without fiber overlapping. The HD-WFN microstructures are obtained by compressing an initially sparse structure using a geometrically designed 3D displacement field. The sparse structure can be optimized to obtain a very low-porosity HD-WFN by reducing the local fiber clustering. The method can generate HD-WFNs with designated shapes (e.g., molded shapes) and varied structural parameters, including fiber width, orientation, location, and material thickness. Each interfiber bond and local material axis in the HD-WFN models can be determined for numerical simulation of properties. The optical scattering of transparent HD-WFN models (polymer matrix composites, transparent paper) is numerically studied using ray tracing methods. The open-source code is available on GitHub, and it can be used to obtain a large dataset for deep learning modeling.
Mechanical behavior of high-density oriented spruce and aspen fiber networks from mildly delignified holocellulose fibers is investigated. Such recyclable, eco-friendly fiber networks are of interest for molded fiber materials and biocomposites. The aspen holocellulose fiber network showed excellent mechanical properties comparable to spruce despite much shorter fiber length. This contrasts with lower density “paper” structures from short fibers which show lower strength than spruce fiber networks. Present results are explained by improved interfiber shear strength and reduced critical fiber length. Microstructures and damage mechanisms were analyzed for materials design purposes using FE-SEM, wide-angle X-ray scattering (WAXS) and tensile testing with strain-field measurements using Digital Image Correlation (DIC).
Transparent composites that combine optical transmittance with mechanical performance are increasingly important for applications in optical devices, sustainable building materials, and photonic engineering. However, predicting light scattering in such materials remains a challenge due to complex, multi-scale microstructural interactions. Here, we present a physically grounded and computationally efficient analytical model. It predicts angular light scattering in transparent composites based on the Average Interface Number (AIN), a single governing microstructural metric derived in this work from geometrical optics. The model accurately captures angular scattering behavior in both fiber- and particle-reinforced composites, as well as in transparent wood. We further introduce the Equivalent Average Interface Number (EAIN), combining AIN with refractive index mismatch into a unified parameter for fast haze prediction. Deep neural network (DNN) analyses confirm AIN as the dominant feature influencing optical scattering. The model predictions are supported by ray-tracing simulations and experimental trends from literature. Finally, we demonstrate the application of our model in fast image rendering simulations through transparent composites. This work provides a compact and practical toolbox for optical design and optimization of transparent structural materials.
Transparent wood (TW) is a sustainable composite material with high optical transmittance and excellent mechanical properties. Nanoparticles, dyes and quantum dots can be added in a controlled manner for new functionalities relying on the light scattering properties of the composite. The scattering properties of 3D TW models of cellular microstructure are investigated numerically using geometrical optics. A group of 3D TW material models with controlled microstructural parameters are generated based on an analytical method. A ray tracing approach is adopted to model scattering in these complex materials. Effects from different material parameters on ray scattering are analyzed. A virtual camera or virtual eye to render images positioned behind a TW plate is simulated using backward ray tracing. The blurred impression in human eyes of real objects viewed through a TW "window" can then be mimicked.
The accuracy of numerical predictions in sheet metal processes involving multiaxial stress-strain states (e.g., blanking, riveting, and incremental forming) heavily depends on the characterisation of plastic anisotropy under multiaxial loading conditions. A fully calibrated 3D plastic anisotropy model is essential for this purpose. While in-plane material behaviour can be conventionally characterised through uniaxial and equi-biaxial tensile tests, calibrating out-of-plane material behaviour remains a significant challenge. This behaviour, governed by out-of-plane shear stress and associated material parameters, is typically described by out-of-plane shear yielding. These parameters are notoriously difficult to determine, leading researchers to frequently assume isotropic behaviour or identical shear parameters for in-plane and out-of-plane responses. Although advanced calibrations may utilise crystal plasticity modelling, there remains a critical need for macro-mechanical characterisation methods. This paper presents an out-of-plane shear testing and material characterisation procedure based on full-field strain measurements using digital image correlation (DIC). Strains within the shear zone are measured via DIC and employed in the Finite Element Model Updating (FEMU) to identify out-of-plane shear parameters of a 2.42 mm thick, cold-rolled AW5754-H22 aluminium alloy sheet, using the Yld2004-18p yield criterion. Given that the characteristic strain response at this scale may be influenced by local crystal structure behaviour on the surface, this paper evaluates the feasibility of such measurements. Finally, to test the validity of the full-field-based approach, the FEMU-identified parameters are compared against results obtained through a classical optimisation procedure based on force-elongation measurements from the shear zone.
Digital Image Correlation (DIC) typically has poor accuracy when the speckle pattern is degraded, as in cases involving fractures, speckle melting or oxidation in high-temperature measurements, speckle slip, speckle obstruction due to surface roughness, pixel overexposure, or low-quality sensors with dead or defective pixels. We propose a pixel-removing DIC (PR-DIC) method that can accurately match the images with these challenging issues. The PR-DIC dynamically discards unreliable pixels within each subset and keeps only the good pixels for subset matching, enabling a better correlation. Numerical tests shows that PR-DIC has good matching accuracy and efficiency even under severe speckle pattern degradation, which obviously outperforms the classical DIC. A real test of joint root demonstrates reliable performance under practical fracture conditions when the images exhibit severe speckle degradation and crack-induced discontinuities.
Finite element model updating (FEMU) is an advanced inverse parameter identification method capable of identifying multiple parameters in a material model through one or a few well-designed material tests. The method has become more mature thanks to the widespread use of full-field measurement techniques, such as digital image correlation. Proper application of FEMU requires extensive expertise. This paper offers a review of FEMU and a guide to practice. It also presents FEMU-DIC, an open-source software package. We conclude by discussing the challenges and opportunities in this field with the intent of inspiring future research.
The implementation of digital image correlation (DIC) involves several different problems, such as image prefiltering, image intensity gradient calculation, image intensity interpolation at subpixel positions, shape function construction and strain calculation. This paper offers a unified insight into the nature of several key problems in DIC technique. We treat all the problems involved in the former mentioned key steps as fundamentally the same problem, that is, reconstructing an analytical description from a discrete and noisy sampling of a signal, such as discrete image intensity and displacement at some scattered nodes. From the reconstructed analytical description, the gradient and physical value at subpixel or integral pixel position can be analytically calculated without extra error. Here, we solve all these problems using the same mathematical tool, the meshfree method, leading to a unified DIC (U-DIC) method. This method introduces errors solely during the steps involving the determination of the continuous description. However, it effectively avoids errors in the remaining steps. It holds the best balance between spatial resolution and measurement resolution for both displacement and strain measurements compared to 28 state-of-the-art DIC algorithms based on the benchmark tests on DIC Challenge 2.0. It also holds all the unique advantages of the advanced meshfree DIC (MF-DIC) compared to conventional local DIC and global DIC. The novel concept and excellent balance between spatial resolution and measurement resolution make U-DIC an attractive replacement for conventional DIC methods. Additionally, the consistency of the implementation procedures in U-DIC can simplify the parameter selection in DIC, which is of potential for the standardization of DIC technique.
Meshfree digital image correlation (MF-DIC) is a recently-proposed advanced DIC technique that deeply integrates meshfree method with DIC. MF-DIC is of potential due to its close relationship with computational mechanics and the excellent balance between spatial resolution and measurement resolution. A new MF-DIC algorithm based on another advanced meshfree method, namely the reproducing kernel particle method (RKPM), is proposed. The RKPM-based MF-DIC is proven to be equivalent to the recently-proposed Element Free Galerkin Method (EFGM) based MF-DIC in certain conditions. This work also briefly introduces two additional MF-DIC methods based on smooth particle hydrodynamics (SPH) and the point interpolation method (PIM). In certain cases, they are the degenerative derivations of the RKPM-based MF-DIC. Benchmark tests on DIC Challenge 2.0 show that these MF-DIC methods also have a superior balance between spatial resolution and measurement resolution compared to state-of-the-art DIC algorithms.
Increased use of multi-phase, wood-based biocomposites may contribute to sustainable development. The porous microstructure offers unique possibilities for modification, but global properties are often predicted based on simplified unit cells and homogenization. For materials design, simulations based on complex 3D microstructures with statistical variability are alternatives to better understanding physical properties. Parametric models are developed in a distortion-map-based method to represent 3D wood microstructures. Basic structures of uniform tubular cells and other features are generated followed by distortion mapping. These maps are highly adaptable and can generate realistic features and variability. Fibers, vessels, and ray cells are realistically distributed. The models are realistic, versatile, and scalable, as well as can be used to simulate the mechanical, optical, and hydrodynamic properties of complex composites. The model is promising for generating large sets of data to train deep learning networks for multi-physics research.
Based on the recently proposed mirror-assisted multi-view digital image correlation (MV-DIC), we establish a cost-effective and easy-to-implement mirror-assisted multi-view high-speed digital image correlation (MVHS-DIC) method and explore its applications for dual-surface full-field dynamic deformation measurement. In contrast to the general requirement of four expensive high-speed cameras for dual-surface dynamic deformation field measurement, the established mirror-assisted MVHS-DIC halves the cost by involving only two synchronized high-speed cameras and two planar mirrors. The two synchronized high-speed cameras can dynamically measure the front and rear surfaces of a sheet sample simultaneously through the reflection of the two mirrors. The results on the two surfaces are then transformed into the same coordinate system, leading to the required dual-surface 3D dynamical deformation fields. The effectiveness and accuracy of the established system are validated through modal tests of a cantilever aluminum sheet. The vibration measurement of a drum and dual-surface transient deformation measurement of a smartphone in the drop-collision process further prove its practicability. Benefiting from the attractive advantages of multi-view dynamic deformation measurement in a cost-efficient way, the established mirror-assisted MVHS-DIC is expected to encourage more comprehensive dynamic mechanical behavior characterization of regular-sized materials and structures in vibration and impact engineering fields.
Polymer shape-memory aerogels (PSMAs) are prospects in various fields of application ranging from aerospace to biomedicine, as advanced thermal insulators, actuators, or sensors. However, the fabrication of PSMAs with good mechanical performance is challenging and is currently dominated by fossil-based polymers. In this work, strong, shape-memory bio-aerogels with high specific surface areas (up to 220 m2/g) and low radial thermal conductivity (0.042 W/mK) were prepared through a one-step treatment of native wood using an ionic liquid mixture of [MTBD]+[MMP]-/DMSO. The aerogel showed similar chemical composition similar to native wood. Nanoscale spatial rearrangement of wood biopolymers in the cell wall and lumen was achieved, resulting in flexible hydrogels, offering design freedom for subsequent aerogels with intricate geometries. Shape-memory function under stimuli of water was reported. The chemical composition and distribution, morphology, and mechanical performance of the aerogel were carefully studied using confocal Raman spectroscopy, AFM, SAXS/WAXS, NMR, digital image correlation, etc. With its simplicity, sustainability, and the broad range of applicability, the methodology developed for nanoscale reassembly of wood is an advancement for the design of biobased shape-memory aerogels.
TW transparent wood/polymer biocomposite laminates are of interest as multifunctional materials with good longitudinal modulus, tensile strength and optical transmittance. The effect of filling the pore space in wood with a polymer matrix on fracture toughness and crack growth is not well understood. Here, we carried out in-situ fracture tests on neat birch wood and laminates made of four layers of delignified birch veneers impregnated with poly(methyl methacrylate) (PMMA) and investigated crack growth in the tangential-radial (TR) fracture system. Fracture toughness KIc and JIc at crack initiation were estimated, including FEM analysis. SEM microscopy revealed that cracks primarily propagate along the ray cells, but cell wall peeling and separation between the PMMA and wood phases also take place. A combination of in-situ tests and strain field measured by digital image correlation (DIC) showed twice as long fracture process zone of TW laminates compared with neat birch.
Owing to the hierarchical structure, easy multi-functionalization and favorable mechanical properties, wood could harvest electricity from mechanical energy through piezoelectric behavior. In this work, a scalable method to synthesize wood/ZnO composite with multilayered ZnO morphologies is reported for efficient mechanical energy conversion. The synthesis includes charged wood template fabrication, precursor infiltration, and ZnO hydrothermal growth, resulting in controlled ZnO morphologies and dis-tributions while maintaining the hierarchical structure of the wood. Stereo-digital image correlation (stereo-DIC) investigated the relationship between deformation and piezoelectric performance, which revealed the homogeneous distribution of multilayered ZnO enhance piezoelectric performance. The out-put voltage of wood/ZnO was 1.5 V under periodic mechanical compression (8???10 N) for 300 cycles, while the output current was 2.91 nA. The scalable synthesis strategy and piezoelectric performance are signif-icant for the design of advanced wood nanocomposites for sustainable and efficient energy conversion systems. ?? 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Functional metal oxide particles are often added to the polymers to prepare flexible functional polymer composites with adequate mechanical properties. ZnO and cellulose nanofibrils (CNF) outstand among these metal oxides and the polymer matrices respectively due to their various advantages. Herein, we in situ prepare ZnO microrods in the presence of CNF, which resultes in a layered composite structure. The ZnO microrods are sandwiched between the CNF layers and strongly bind to highly charged CNF, which provides a better stress transfer during mechanical activity. Digital image correction (DIC) and finite element analysis-based computational homogenization methods are used to investigate the relationship between mechanical properties and composite structure, and the stress transfer to the ZnO microrods. Full-field strain measurements in DIC reveal that the in situ ZnO microrods preparation leads to their homogenous distribution in the CNF matrix unlike other methods, which require external means such as ultrasonication. The computational homogenization technique provides a fairly good insight into the stress transfer between constituents in microstructure as well as a good prediction of macroscopic mechanical properties, which otherwise, would be challenging to be assessed by any ordinary mechanical testing in the layered composites. Finally, we also demonstrate that these composites could be used as physiological motion sensors for human health monitoring.
We propose an element-removal (ER) global digital image correlation (DIC) method to improve the measurement accuracy of discontinuous deformation fields, such as crack propagation. The occurrence of cracks in materials or structures inevitably deteriorates the tracking accuracy, and, consequently, the strain field accuracy obtained by regular subset and global DIC. The proposed ER-global-DIC algorithm iteratively identifies and removes all the elements covering the crack, during the updating of displacement fields. In the remaining elements, the continuous shape function is applicable for accurate deformation measurement. In principle, although elements that contain the cracks are removed, the algorithm preserves the same number of nodes since the nodes are retained by the remaining elements. Synthetically deformed images based on analytical discontinuous displacement fields validate the effectiveness and accuracy of the proposed method. The ER-global-DIC is further applied to measure the discontinuous displacement fields containing a crack deflection, generated from a finite element model with a cohesive zone model. The results demonstrate the potential of the proposed method for discontinuous deformation measurement on advanced materials, e.g., fiber-reinforced composites.
A multifunctional soft material with high ionic and electrical conductivity, combined with high mechanical properties and the ability to change shape can enable bioinspired responsive devices and systems. The incorporation of all these characteristics in a single material is very challenging, as the improvement of one property tends to reduce other properties. Here, a nanocomposite film based on charged, high‐aspect‐ratio 1D flexible nanocellulose fibrils, and 2D Ti 3 C 2 T x MXene is presented. The self‐assembly process results in a stratified structure with the nanoparticles aligned in‐plane, providing high ionotronic conductivity and mechanical strength, as well as large water uptake. In hydrogel form with 20 wt% liquid, the electrical conductivity is over 200 S cm −1 and the in‐plane tensile strength is close to 100 MPa. This multifunctional performance results from the uniquely layered composite structure at nano‐ and mesoscales. A new type of electrical soft actuator is assembled where voltage as low as ±1 V resulted in osmotic effects and giant reversible out‐of‐plane swelling, reaching 85% strain.
Background The association of advanced digital image correlation (DIC) and numerical simulation has been widely used for inverse parameter identification. Objective It is attractive to develop an accurate DIC method sharing the common features with numerical simulation, which can lead to better synergy between experiments and simulations. Methods A new meshfree digital image correlation (MF-DIC) using element free Galerkin method (EFGM) is proposed for deformation measurement. The EFGM is a classical meshfree method in numerical studies, and it is directly used to construct the shape function in MF-DIC from a set of scattered nodes for image matching. The MF-DIC is principally different from the classical local DIC and global DIC since it does not rely on the concept of a subset or an element. Results In MF-DIC, the C^1 -continuous displacement for every point is constructed based on a group of scattered nodes in a small support domain surrounding it. The continuous strain map can then be directly derived from the displacement, instead of using an additional smoothing technique as required in classical local DIC or post-processing used in global DIC. A performance assessment based on the Metrological Efficiency Indicator (MEI), as defined in DIC Challenge 2.0, shows that the proposed MF-DIC yields an excellent balance between spatial resolution and measurement resolution for both displacement and strain measurements. Conclusions Given that the proposed MF-DIC shares common features with the classical meshfree method in computational mechanics, it paves the way for an enhanced synergy between experiments and simulations required for robust inverse parameter identification methods.