Estimation of Local Strain Fields in Two-Phase Elastic Composite Materials Using UNet-Based Deep Learning

2021 ◽  
Author(s):  
Mayank Raj ◽  
Sanket Thakre ◽  
Ratna Kumar Annabattula ◽  
Anand K Kanjarla
Keyword(s):  
Deep Learning ◽  
Composite Materials ◽  
Local Strain ◽  
Two Phase ◽  
Strain Fields ◽  
Elastic Composite

Delamination in the scoring and folding of paperboard

TAPPI Journal ◽  
2012 ◽  
Vol 11 (1) ◽  
pp. 61-66 ◽  
Author(s):  
DOEUNG D. CHOI ◽  
SERGIY A. LAVRYKOV ◽  
BANDARU V. RAMARAO
Keyword(s):  
Finite Element ◽  
Plane Deformation ◽  
Strain Analysis ◽  
Local Strain ◽  
Uniaxial Stress ◽  
Material Model ◽  
Element Model ◽  
Plastic Behavior ◽  
Stress Strain ◽  
Strain Fields

Delamination between layers occurs during the creasing and subsequent folding of paperboard. Delamination is necessary to provide some stiffness properties, but excessive or uncontrolled delamination can weaken the fold, and therefore needs to be controlled. An understanding of the mechanics of delamination is predicated upon the availability of reliable and properly calibrated simulation tools to predict experimental observations. This paper describes a finite element simulation of paper mechanics applied to the scoring and folding of multi-ply carton board. Our goal was to provide an understanding of the mechanics of these operations and the proper models of elastic and plastic behavior of the material that enable us to simulate the deformation and delamination behavior. Our material model accounted for plasticity and sheet anisotropy in the in-plane and z-direction (ZD) dimensions. We used different ZD stress-strain curves during loading and unloading. Material parameters for in-plane deformation were obtained by fitting uniaxial stress-strain data to Ramberg-Osgood plasticity models and the ZD deformation was modeled using a modified power law. Two-dimensional strain fields resulting from loading board typical of a scoring operation were calculated. The strain field was symmetric in the initial stages, but increasing deformation led to asymmetry and heterogeneity. These regions were precursors to delamination and failure. Delamination of the layers occurred in regions of significant shear strain and resulted primarily from the development of large plastic strains. The model predictions were confirmed by experimental observation of the local strain fields using visual microscopy and linear image strain analysis. The finite element model predicted sheet delamination matching the patterns and effects that were observed in experiments.

scholarly journals Efficiency of the Needle Probe Test for Evaluation of Thermal Conductivity of Composite Materials: Two-Scale Analysis

2014 ◽  
Vol 36 (1) ◽  
pp. 55-62 ◽  
Author(s):  
Dariusz Łydżba ◽  
Adrian Różański ◽  
Magdalena Rajczakowska ◽  
Damian Stefaniuk
Keyword(s):  
Thermal Conductivity ◽  
Composite Materials ◽  
Numerical Simulations ◽  
Conductivity Measurement ◽  
Scale Analysis ◽  
Equivalent Conductivity ◽  
Inclusion Type ◽  
Two Phase ◽  
Needle Probe ◽  
Probe Test

Abstract The needle probe test, as a thermal conductivity measurement method, has become very popular in recent years. In the present study, the efficiency of this methodology, for the case of composite materials, is investigated based on the numerical simulations. The material under study is a two-phase composite with periodic microstructure of “matrix-inclusion” type. Two-scale analysis, incorporating micromechanics approach, is performed. First, the effective thermal conductivity of the composite considered is found by the solution of the appropriate boundary value problem stated for the single unit cell. Next, numerical simulations of the needle probe test are carried out. In this case, two different locations of the measuring sensor are considered. It is shown that the “equivalent” conductivity, derived from the probe test, is strongly affected by the location of the sensor. Moreover, comparing the results obtained for different scales, one can notice that the “equivalent” conductivity cannot be interpreted as the effective one for the composites considered. Hence, a crude approximation of the effective property is proposed based on the volume fractions of constituents and the equivalent conductivities derived from different sensor locations.

scholarly journals Elasto-elastic composite materials with a polymer matrix based on ultrafine iron and polyethylene

2021 ◽  
Vol 11 (1) ◽  
pp. 17-21
Author(s):  
Yuri Kabirov ◽  
Evgeny Sidorenko ◽  
Natalya Prutsakova ◽  
Mark Belokobylsky ◽  
Andrey Letovaltsev ◽  
...  
Keyword(s):  
Composite Materials ◽  
Polymer Matrix ◽  
Elastic Composite

scholarly journals Deep Network Guided Proof Search

10.29007/8mwc ◽  
2018 ◽  
Author(s):  
Sarah Loos ◽  
Geoffrey Irving ◽  
Christian Szegedy ◽  
Cezary Kaliszyk
Keyword(s):  
Deep Learning ◽  
Program Analysis ◽  
Network Models ◽  
Theorem Prover ◽  
Neural Network Models ◽  
Two Phase ◽  
First Order ◽  
Proof Search ◽  
System Verification ◽  
Theory Reasoning

Deep learning techniques lie at the heart of several significant AI advances in recent years including object recognition and detection, image captioning, machine translation, speech recognition and synthesis, and playing the game of Go.Automated first-order theorem provers can aid in the formalization and verification of mathematical theorems and play a crucial role in program analysis, theory reasoning, security, interpolation, and system verification.Here we suggest deep learning based guidance in the proof search of the theorem prover E. We train and compare several deep neural network models on the traces of existing ATP proofs of Mizar statements and use them to select processed clauses during proof search. We give experimental evidence that with a hybrid, two-phase approach, deep learning based guidance can significantly reduce the average number of proof search steps while increasing the number of theorems proved.Using a few proof guidance strategies that leverage deep neural networks, we have found first-order proofs of 7.36% of the first-order logic translations of the Mizar Mathematical Library theorems that did not previously have ATP generated proofs. This increases the ratio of statements in the corpus with ATP generated proofs from 56% to 59%.

An Orthotropic Integrated Flow-Stress Model for Process Simulation of Composite Materials—Part I: Two-Phase Systems

2019 ◽  
Vol 141 (3) ◽  
Author(s):  
Sina Amini Niaki ◽  
Alireza Forghani ◽  
Reza Vaziri ◽  
Anoush Poursartip
Keyword(s):  
Composite Materials ◽  
Flow Stress ◽  
Interactive Effect ◽  
Fluid Phase ◽  
Solid Material ◽  
Material System ◽  
Two Phase ◽  
Stress Development ◽  
Thermoset Resin ◽  
Three Phase

An integrated flow-stress (IFS) model provides a seamless and mechanistic connection between the two distinct regimes during the manufacturing process of composite materials, namely, fluid flow in the pregelation stage of the thermoset resin and stress development in the composite when it acts as a solid material. In this two-part paper, the two- and three-phase isotropic IFS models previously developed by the authors are extended to the general case of composite materials with orthotropic constituents. Part I presents the two-phase, fluid-solid, orthotropic model formulation for the case where the fluid phase solidifies during the course of curing. Part II extends the orthotropic formulation to a three-phase model that includes a gas phase as the third constituent of the composite material system. A broader definition of material properties in poroelasticity formulation is adopted in the development of the general orthotropic formulation. The model is implemented in a two-dimensional (2D) plane strain u-v-P finite element (FE) code and its capability in predicting the flow-compaction behavior and stress development is demonstrated through application to a case study involving an L-shaped unidirectional laminate undergoing curing on a conforming convex tool. Comparison of the results with those obtained from sole modeling of the stress development reveals the importance of capturing the simultaneous and interactive effect of the mechanisms involved during the entire process cycle using an IFS modeling approach presented in this paper.

scholarly journals Two-Phase Object-Based Deep Learning for Multi-Temporal SAR Image Change Detection

2020 ◽  
Vol 12 (3) ◽  
pp. 548 ◽  
Author(s):  
Xinzheng Zhang ◽  
Guo Liu ◽  
Ce Zhang ◽  
Peter M. Atkinson ◽  
Xiaoheng Tan ◽  
...  
Keyword(s):  
Deep Learning ◽  
Change Detection ◽  
Speckle Noise ◽  
Sar Image ◽  
Phase Object ◽  
Sar Images ◽  
Two Phase ◽  
Object Based ◽  
Multi Temporal ◽  
Image Change Detection

Change detection is one of the fundamental applications of synthetic aperture radar (SAR) images. However, speckle noise presented in SAR images has a negative effect on change detection, leading to frequent false alarms in the mapping products. In this research, a novel two-phase object-based deep learning approach is proposed for multi-temporal SAR image change detection. Compared with traditional methods, the proposed approach brings two main innovations. One is to classify all pixels into three categories rather than two categories: unchanged pixels, changed pixels caused by strong speckle (false changes), and changed pixels formed by real terrain variation (real changes). The other is to group neighbouring pixels into superpixel objects such as to exploit local spatial context. Two phases are designed in the methodology: (1) Generate objects based on the simple linear iterative clustering (SLIC) algorithm, and discriminate these objects into changed and unchanged classes using fuzzy c-means (FCM) clustering and a deep PCANet. The prediction of this Phase is the set of changed and unchanged superpixels. (2) Deep learning on the pixel sets over the changed superpixels only, obtained in the first phase, to discriminate real changes from false changes. SLIC is employed again to achieve new superpixels in the second phase. Low rank and sparse decomposition are applied to these new superpixels to suppress speckle noise significantly. A further clustering step is applied to these new superpixels via FCM. A new PCANet is then trained to classify two kinds of changed superpixels to achieve the final change maps. Numerical experiments demonstrate that, compared with benchmark methods, the proposed approach can distinguish real changes from false changes effectively with significantly reduced false alarm rates, and achieve up to 99.71% change detection accuracy using multi-temporal SAR imagery.

Sign in / Sign up

Export Citation Format

Share Document