Virtual materials testing: come funziona e i benefici sulla produzioneCompositi
Webinar 29 settembre 2020
Connettendo il mondo reale con il virtuale, il processo di sviluppo di virtual materials testing permette di accelerare i tempi e ridurre i costi della filiera.
Design of novel heterogeneous thermomechanical tests using topology-based opt...Vform Xsteels
Nowadays, the development and design of new mechanical parts are performed using virtual manufacturing tools. For this purpose, numerical simulation software uses complex material constitutive models to reproduce real material behavior. However, the reliability of the simulation’s results is conditioned by the accuracy of the input data. The calibration of material models depends on a proper material characterization, which turns to be expensive due to the required large number of mechanical tests.
Digimat Virtual Allowables, to complement physical testing of composite couponseXDigimat
Integrated Solution to Compute Virtual Allowables. Winner of the 2015 JEC Innovation Award and the ACE Award for Composite Excellence.
Digimat Virtual Allowables is a Solution of Digimat software that integrate the right Digimat Tools and features to compute the behavior of composite coupons
Digimat-VA (“Virtual Allowables”) is an efficient solution that empowers engineers to virtually compare materials before going into the lengthy physical allowables. By generating virtual allowables, engineers can now start the component design in parallel to the physical allowable campaign.
Digimat-VA is a vertical solution developed to virtually compute the behavior of composite coupons (unnotched, open hole, filled hole). It allows engineers to screen, select and compute the allowables of composite materials in less time and at less cost.
On the inverse identification of sheet metal mechanical behaviour using a het...vformxsteels
ESAFORM 2023
João Henriquesa,*, A. Andrade-Camposa
, J. Xavierb
aTEMA, Department of Mechanical Engineering, University of Aveiro, Campus
Universitário de Santiago, 3810-193 Aveiro, Portugal
bUNIDEMI, Department of Mechanical and Industrial Engineering, NOVA School of
Science and Technology, NOVA University Lisbon, 2825-149 Lisbon, Portugal
Virtual materials testing: come funziona e i benefici sulla produzioneCompositi
Webinar 29 settembre 2020
Connettendo il mondo reale con il virtuale, il processo di sviluppo di virtual materials testing permette di accelerare i tempi e ridurre i costi della filiera.
Design of novel heterogeneous thermomechanical tests using topology-based opt...Vform Xsteels
Nowadays, the development and design of new mechanical parts are performed using virtual manufacturing tools. For this purpose, numerical simulation software uses complex material constitutive models to reproduce real material behavior. However, the reliability of the simulation’s results is conditioned by the accuracy of the input data. The calibration of material models depends on a proper material characterization, which turns to be expensive due to the required large number of mechanical tests.
Digimat Virtual Allowables, to complement physical testing of composite couponseXDigimat
Integrated Solution to Compute Virtual Allowables. Winner of the 2015 JEC Innovation Award and the ACE Award for Composite Excellence.
Digimat Virtual Allowables is a Solution of Digimat software that integrate the right Digimat Tools and features to compute the behavior of composite coupons
Digimat-VA (“Virtual Allowables”) is an efficient solution that empowers engineers to virtually compare materials before going into the lengthy physical allowables. By generating virtual allowables, engineers can now start the component design in parallel to the physical allowable campaign.
Digimat-VA is a vertical solution developed to virtually compute the behavior of composite coupons (unnotched, open hole, filled hole). It allows engineers to screen, select and compute the allowables of composite materials in less time and at less cost.
On the inverse identification of sheet metal mechanical behaviour using a het...vformxsteels
ESAFORM 2023
João Henriquesa,*, A. Andrade-Camposa
, J. Xavierb
aTEMA, Department of Mechanical Engineering, University of Aveiro, Campus
Universitário de Santiago, 3810-193 Aveiro, Portugal
bUNIDEMI, Department of Mechanical and Industrial Engineering, NOVA School of
Science and Technology, NOVA University Lisbon, 2825-149 Lisbon, Portugal
On the inverse identification of sheet metal mechanical behaviour using a het...vformxsteels
ESAFORM 2023
João Henriquesa,*, A. Andrade-Camposa
, J. Xavierb
aTEMA, Department of Mechanical Engineering, University of Aveiro, Campus
Universitário de Santiago, 3810-193 Aveiro, Portugal
bUNIDEMI, Department of Mechanical and Industrial Engineering, NOVA School of
Science and Technology, NOVA University Lisbon, 2825-149 Lisbon, Portugal
Shaped Metal deposition Based on Additive ManufacturingHassan Alwaely
A three-axis system was designed and manufactured to facilitate SMD operations. Initially, a feasible design of the proposed CADWD machine had been prepared. The design had been carried out according to the valid standards. The CADWDM is actuated with three stepper motors, which connected along with x-, y-, and z-directions. The maximum working area is (450 x 750 x 200) mm3. Machine design carried out using Auto CAD program and then it was simulated through ANSYS workbench program.
The COREMA system allow for non-destructive resistivity testing of semi-insulating wafers made with materials such as SiC, GaN, GaAs and CdZnTe. The range is 1E5- 1E12 ohm-cm.
University of Victoria Talk - Metocean analysis and Machine Learning for Impr...Aaron Barker
A presentation given on Metocean analysis and Machine Learning for improved estimates of energy production in WECs by Aaron Barker at the University of Victoria on the 7th of December 2017
Modeling of wedm process for complex shape using multilayer perceptron and re...eSAT Journals
Abstract
Wire cut electric discharge machining is the present days requirement in manufacturing the intricate and complex shape parts as required in the modern industrial products. The process being complex in nature for control over the machining process parameters. In the present study Multilayer perceptron (MLP) model is developed to predict the Material removal rate. The mathematical regression model is also developed by using Response Surface Methodology (RSM) to establish the relationshIp between the MRR and input process parameters like pulse on time (Ton), Pulse off time (Toff), Peak current (IP) and Servo voltage (SV). The predicted value by using MLP and RSM were compared with the experimental values. The average percentage errors are found to be 1.29 and -0.36527 for MLP and RSM respectively. It is observed that the predicted values with RSM are closer to experimental values as compared to MLP model. Key Words: WEDM, Response Surface Methodology, Neural Network, Multi Layer Perception, Material Removal Rate, Surface Roughness
Identification of anisotropic yield functions using an information-rich tensi...vformxsteels
Yi Zhang1, António Andrade-Campos2, Sam Coppieters1
1 Department of Materials Engineering, KU Leuven
2 Department of Mechanical Engineering, University of Aveiro, 3810-193
Aveiro, Portugal
More Related Content
Similar to On the power of virtual experimentation in MT2.0:a VFORM-xSteels outlook
On the inverse identification of sheet metal mechanical behaviour using a het...vformxsteels
ESAFORM 2023
João Henriquesa,*, A. Andrade-Camposa
, J. Xavierb
aTEMA, Department of Mechanical Engineering, University of Aveiro, Campus
Universitário de Santiago, 3810-193 Aveiro, Portugal
bUNIDEMI, Department of Mechanical and Industrial Engineering, NOVA School of
Science and Technology, NOVA University Lisbon, 2825-149 Lisbon, Portugal
Shaped Metal deposition Based on Additive ManufacturingHassan Alwaely
A three-axis system was designed and manufactured to facilitate SMD operations. Initially, a feasible design of the proposed CADWD machine had been prepared. The design had been carried out according to the valid standards. The CADWDM is actuated with three stepper motors, which connected along with x-, y-, and z-directions. The maximum working area is (450 x 750 x 200) mm3. Machine design carried out using Auto CAD program and then it was simulated through ANSYS workbench program.
The COREMA system allow for non-destructive resistivity testing of semi-insulating wafers made with materials such as SiC, GaN, GaAs and CdZnTe. The range is 1E5- 1E12 ohm-cm.
University of Victoria Talk - Metocean analysis and Machine Learning for Impr...Aaron Barker
A presentation given on Metocean analysis and Machine Learning for improved estimates of energy production in WECs by Aaron Barker at the University of Victoria on the 7th of December 2017
Modeling of wedm process for complex shape using multilayer perceptron and re...eSAT Journals
Abstract
Wire cut electric discharge machining is the present days requirement in manufacturing the intricate and complex shape parts as required in the modern industrial products. The process being complex in nature for control over the machining process parameters. In the present study Multilayer perceptron (MLP) model is developed to predict the Material removal rate. The mathematical regression model is also developed by using Response Surface Methodology (RSM) to establish the relationshIp between the MRR and input process parameters like pulse on time (Ton), Pulse off time (Toff), Peak current (IP) and Servo voltage (SV). The predicted value by using MLP and RSM were compared with the experimental values. The average percentage errors are found to be 1.29 and -0.36527 for MLP and RSM respectively. It is observed that the predicted values with RSM are closer to experimental values as compared to MLP model. Key Words: WEDM, Response Surface Methodology, Neural Network, Multi Layer Perception, Material Removal Rate, Surface Roughness
Identification of anisotropic yield functions using an information-rich tensi...vformxsteels
Yi Zhang1, António Andrade-Campos2, Sam Coppieters1
1 Department of Materials Engineering, KU Leuven
2 Department of Mechanical Engineering, University of Aveiro, 3810-193
Aveiro, Portugal
Independent Validation of Generic Specimen Design for Inverse Identification ...vformxsteels
Yi Zhanga , S. Coppietersa,*, S. Gothivarekara, A. Van de Veldea , D. Debruynea
a Department of Materials Engineering, KU Leuven, Technology Campus Ghent, 9000 Gent, Belgium
Belgium,19 March 2021
Identification of the large strain flow curve of high strength steel via the ...vformxsteels
Niels Vancraeynest1, Steven Cooreman2 and Sam Coppieters1
1Elooi lab, Department of Materials Engineering, KU Leuven
2 Applications & Solutions department, ArcelorMittal Global R&D Gent
On the topology design of an innovative heterogeneous mechanical test for mat...vformxsteels
RS 2022
On the topology design of an innovative
heterogeneous mechanical test for material
mechanical characterization
Centre for Mechanical Technology and Automation (TEMA), Department of Mechanical Engineering, University of Aveiro,
Portugal
Univ. Bretagne Sud, UMR CNRS 6027, IRDL, F-56100 Lorient, France
Mafalda Gonçalves, Sandrine Thuillier, António Andrade-Campos
Doctoral Program in Mechanical Engineering
On the comparison of heterogeneous mechanical tests for sheet metal character...vformxsteels
IDDRG 2022
M. Gonçalves, M.G. Oliveira, S. Thuillier, A. Andrade-Campos
Centre for Mechanical Technology and Automation (TEMA),
Department of Mechanical Engineering
University of Aveiro, Portugal
Univ. Bretagne Sud, UMR CNRS 6027, IRDL
F-56100 Lorient, France
1
Process-informed material model selectionvformxsteels
ESAFORM 2023
Mariana Conde1,*, Sam Coppieters2, António Andrade-Campos1
1Department of Mechanical Engineering, TEMA - Centre for Mechanical Technology and Automation, LASI – Intelligent Systems Associate Laboratory, University of Aveiro, Portugal
2Department of Materials Engineering, KU Leuven, Belgium
*marianaconde@ua.pt
On the selection of constitutive models for realistic numerical simulationsvformxsteels
TechMA 2022
M. Conde (a), *, S. Coppieters (b), A. Andrade-Campos (a)
(a) - Centre for Mechanical Technology and Automation (TEMA), Department of Mechanical Engineering, University of Aveiro; (b) - Department of Materials Engineering, KU Leuven(a) Aveiro, Portugal; (b) – Ghent, Belgium
* – marianaconde@ua.pt
Towards virtual forming and AI Implicit material modelling using AI technique...vformxsteels
RS 2022
Rúben Lourenço
Supervisory team: A. Andrade-Campos, Pétia Georgieva
Doctoral Programme in Mechanical Engineering
TEMA – Centre for Mechanical Technology and Automation
Parameter Identification of Swift law using a FEMU-based approach and an inno...vformxsteels
ESAFORM 2022
M. Conde1, J. Henriques1,
S. Coppieters2, A. Andrade-Campos1
1 TEMA, Department of Mechanical Engineering, University of Aveiro, Portugal
2 Department of Materials Engineering, Ghent Technology Campus, KU Leuven, Belgium
Uncertainty quantification of inversely identified plastic material behavior ...vformxsteels
IDDRG 2022
M Conde1,*, S Coppieters2, A Andrade-Campos1
* marianaconde@ua.pt
1 TEMA, Department of Mechanical Engineering, University of Aveiro, Portugal
2 Department of Materials Engineering, Ghent Technology Campus, KU Leuven, Belgium
A nonlinear topology-based optimization approach for the design of a heteroge...vformxsteels
TechMA 2022
Mafalda Gonçalves, Sandrine Thuillier, António Andrade-Campos
Centre for Mechanical Technology and Automation, Department of Mechanical Engineering, University of Aveiro, Portugal
Univ. Bretagne Sud, UMR CNRS 6027, IRDL, F-56100 Lorient, France
On the design of a heterogeneous mechanical test using a nonlinear topology o...vformxsteels
ESAFORM 2022
M. Gonçalves, A. Andrade-Campos and S. Thuillier
25th International Conference on Material Forming
27-29th April 2022
Braga, Portugal
Centre for Mechanical Tecnology and Automation, Department of Mechanical Engineering, University of Aveiro, Portugal
Univ. Bretagne Sud, UMR CNRS 6027, IRDL, F-56100 Lorient, France
IDENTIFICATION OF SWIFT LAW PARAMETERS USING FEMU BY A SYNTHETIC IMAGE DIC-BA...vformxsteels
ESAFORM 2022
J. Henriquesa,*, M. Condea
, A. Andrade-Camposa
, J. Xavierb
*Corresponding author: joaodiogofh@ua.pt
aTEMA, Department of Mechanical Engineering, University of Aveiro, Campus Universitário de Santiago,
3810-193 Aveiro, Portugal
bUNIDEMI, Department of Mechanical and Industrial Engineering, NOVA School of Science and Technology,
NOVA University Lisbon, 2829-516 Caparica, Portugal
On the comparison of heterogeneous mechanical tests for sheet metal character...vformxsteels
ESAFORM 2023
M. Gonçalves, M.G. Oliveira, S. Thuillier, A. Andrade-Campos
Centre for Mechanical Technology and Automation (TEMA),
Department of Mechanical Engineering
University of Aveiro, Portugal
Univ. Bretagne Sud, UMR CNRS 6027, IRDL
F-56100 Lorient, France
The Virtual Fields Method to indirectly train ANNs for implicit constitutive ...vformxsteels
ESAFORM 2022
25th International Conference on Material Forming
R. Lourenço1,a, A. Andrade-Campos1,b, P. Georgieva2,c
1Dept. of Mechanical Engineering
Centre for Mechanical Technology and Automation, University of Aveiro, Portugal
2Dept. of Electronics Telecommunications and Informatics
Institute of Electronics and Informatics Engineering, University of Aveiro, Portugal
a
rubenl@ua.pt, bgilac@ua.pt, cpetia@ua.pt
27-29 April, Braga, Portugal
On the power of virtual experimentation in MT2.0:a VFORM-xSteels outlookvformxsteels
Sam Coppieters, A. Gil Andrade-Campos et al.
MatchID Global User Meeting
On the power of virtual experimentation in MT2.0 : a VFORM outlook
22 February2023 | Southampton, UK
On the comparison of heterogeneous mechanical tests for sheet metal character...vformxsteels
ESAFORM 2023
M. Gonçalves, M.G. Oliveira, S. Thuillier, A. Andrade-Campos 1 1,2 2 1
Centre for Mechanical Technology and Automation (TEMA),
Department of Mechanical Engineering
University of Aveiro, Portugal
Univ. Bretagne Sud, UMR CNRS 6027, IRDL
F-56100 Lorient, France
1
Uncertainty quantification of inversely identified plastic material behavior ...vformxsteels
M Conde, S Coppieters, A Andrade-Campos
TEMA, Department of Mechanical Engineering, University of Aveiro, Portugal
Department of Materials Engineering, Ghent Technology Campus, KU Leuven, Belgium
NO1 Uk best vashikaran specialist in delhi vashikaran baba near me online vas...Amil Baba Dawood bangali
Contact with Dawood Bhai Just call on +92322-6382012 and we'll help you. We'll solve all your problems within 12 to 24 hours and with 101% guarantee and with astrology systematic. If you want to take any personal or professional advice then also you can call us on +92322-6382012 , ONLINE LOVE PROBLEM & Other all types of Daily Life Problem's.Then CALL or WHATSAPP us on +92322-6382012 and Get all these problems solutions here by Amil Baba DAWOOD BANGALI
#vashikaranspecialist #astrologer #palmistry #amliyaat #taweez #manpasandshadi #horoscope #spiritual #lovelife #lovespell #marriagespell#aamilbabainpakistan #amilbabainkarachi #powerfullblackmagicspell #kalajadumantarspecialist #realamilbaba #AmilbabainPakistan #astrologerincanada #astrologerindubai #lovespellsmaster #kalajaduspecialist #lovespellsthatwork #aamilbabainlahore#blackmagicformarriage #aamilbaba #kalajadu #kalailam #taweez #wazifaexpert #jadumantar #vashikaranspecialist #astrologer #palmistry #amliyaat #taweez #manpasandshadi #horoscope #spiritual #lovelife #lovespell #marriagespell#aamilbabainpakistan #amilbabainkarachi #powerfullblackmagicspell #kalajadumantarspecialist #realamilbaba #AmilbabainPakistan #astrologerincanada #astrologerindubai #lovespellsmaster #kalajaduspecialist #lovespellsthatwork #aamilbabainlahore #blackmagicforlove #blackmagicformarriage #aamilbaba #kalajadu #kalailam #taweez #wazifaexpert #jadumantar #vashikaranspecialist #astrologer #palmistry #amliyaat #taweez #manpasandshadi #horoscope #spiritual #lovelife #lovespell #marriagespell#aamilbabainpakistan #amilbabainkarachi #powerfullblackmagicspell #kalajadumantarspecialist #realamilbaba #AmilbabainPakistan #astrologerincanada #astrologerindubai #lovespellsmaster #kalajaduspecialist #lovespellsthatwork #aamilbabainlahore #Amilbabainuk #amilbabainspain #amilbabaindubai #Amilbabainnorway #amilbabainkrachi #amilbabainlahore #amilbabaingujranwalan #amilbabainislamabad
Forklift Classes Overview by Intella PartsIntella Parts
Discover the different forklift classes and their specific applications. Learn how to choose the right forklift for your needs to ensure safety, efficiency, and compliance in your operations.
For more technical information, visit our website https://intellaparts.com
Courier management system project report.pdfKamal Acharya
It is now-a-days very important for the people to send or receive articles like imported furniture, electronic items, gifts, business goods and the like. People depend vastly on different transport systems which mostly use the manual way of receiving and delivering the articles. There is no way to track the articles till they are received and there is no way to let the customer know what happened in transit, once he booked some articles. In such a situation, we need a system which completely computerizes the cargo activities including time to time tracking of the articles sent. This need is fulfilled by Courier Management System software which is online software for the cargo management people that enables them to receive the goods from a source and send them to a required destination and track their status from time to time.
Overview of the fundamental roles in Hydropower generation and the components involved in wider Electrical Engineering.
This paper presents the design and construction of hydroelectric dams from the hydrologist’s survey of the valley before construction, all aspects and involved disciplines, fluid dynamics, structural engineering, generation and mains frequency regulation to the very transmission of power through the network in the United Kingdom.
Author: Robbie Edward Sayers
Collaborators and co editors: Charlie Sims and Connor Healey.
(C) 2024 Robbie E. Sayers
Water scarcity is the lack of fresh water resources to meet the standard water demand. There are two type of water scarcity. One is physical. The other is economic water scarcity.
Final project report on grocery store management system..pdfKamal Acharya
In today’s fast-changing business environment, it’s extremely important to be able to respond to client needs in the most effective and timely manner. If your customers wish to see your business online and have instant access to your products or services.
Online Grocery Store is an e-commerce website, which retails various grocery products. This project allows viewing various products available enables registered users to purchase desired products instantly using Paytm, UPI payment processor (Instant Pay) and also can place order by using Cash on Delivery (Pay Later) option. This project provides an easy access to Administrators and Managers to view orders placed using Pay Later and Instant Pay options.
In order to develop an e-commerce website, a number of Technologies must be studied and understood. These include multi-tiered architecture, server and client-side scripting techniques, implementation technologies, programming language (such as PHP, HTML, CSS, JavaScript) and MySQL relational databases. This is a project with the objective to develop a basic website where a consumer is provided with a shopping cart website and also to know about the technologies used to develop such a website.
This document will discuss each of the underlying technologies to create and implement an e- commerce website.
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...Dr.Costas Sachpazis
Terzaghi's soil bearing capacity theory, developed by Karl Terzaghi, is a fundamental principle in geotechnical engineering used to determine the bearing capacity of shallow foundations. This theory provides a method to calculate the ultimate bearing capacity of soil, which is the maximum load per unit area that the soil can support without undergoing shear failure. The Calculation HTML Code included.
On the power of virtual experimentation in MT2.0:a VFORM-xSteels outlook
1. On the power of virtual
experimentation in MT2.0:
a VFORM-xSteels outlook
Sam Coppieters, A. Gil Andrade-Campos et al.
MatchID Global User Meeting
On the power of virtual experimentation in MT2.0 : a
VFORM outlook
22 February2023 | Southampton, UK
2. Outline
• The vForm-xSteels project;
• The power of virtual experimentation in MT2.0
Virtual experimentation (DVT)
DVT vs. Actual Experiment
CMAT: stress integration algorithms
• The power of virtual experimentation in ML approaches
The methodology
The case study of the Biaxial cruciform test
The virtual experiments data-base
Robustness analysis
5. Toward virtual forming and design:
Thermomechanical characterization of advanced
high strength steels through full-field
measurements and a single designed test
Call: RFCS-2019
Instrument: (RP / PDP / AM)
Start date: 01/06/2020
End date: 31/05/2024
Budget: €1,496,357.52
Type of Action: RFCS-RPJ
Number: 888153
Duration: 48 months
GA based on the: RFCS MGA — Multi - 2.null
Estimated Project Cost: €2,521,389.20
Requested EU Contribution: €1,496,357.52
This project has received funding from the Research Fund for Coal and Steel under grant agreement No 888153
8. Problem and need
Increase information &
reduce time in material testing
Realistic simulations and material
behavior reproduction
Faster and accurate
development & production
9. Problem
Limited quality and amount of data
"The data we have is provided by suppliers. In some
cases it is limited."
André Ferreira
Designer Projectist @ OLI Oliveira e Irmãos
The new paradigm of
Material Testing 2.0
(previous seminar from F.
Pierron)
10. Problem
Error in material behavior virtual
predictions
"It would be important to have a service that could identify and
provide these data to companies in due time."
Mário Marques
Country Manager Iberia @ Autoform
11. Project goal
Nowadays, the use of numerical simulation in general and particularly finite element analysis (FEA) has
become a mandatory step (…) However, todays methods to characterize the materials through constitutive
models, including damage, and their parameters are expensive and not robust.
The main goal of the project VForm-XSteels is to develop an
efficient and accurate methodology for material characterization
and determining the material parameters of thermomechanical
models, from a dedicated single test that involve non-homogeneous
temperature and strain fields. Indeed, (…) A database and online library with calibrated
material constitutive models, particularly for AHSS, is also developed.
The benefits of the proposed methodology and consequent implemented numerical tool developed within
this project are (…) cost and time reduction in the overall development process are also benefits of this
proposal.
13. Project goal and technology
IP test Optical
technology
Optimization
algorithms
Flexible
Accurate
Robust
Fast
Innovative experimental
test (1 000+ points)
Advanced Data analytics
to calibrate the material
models
14. WP3 - Design of a novel test using an integrated topology-shape
optimisation methodology and a thermo-mechanical indicator
• Mafalda Gonçalves, Thibault Barret, Antonio Andrade-Campos, Sandrine Thuillier
• Design by topology optimisation of a heterogeneous test to calibrate a viscoplastic model
• First from homogeneous tests (reference), then heterogenous tests Future work: thermal dependence to enhance the strain rate sensitivity
• KPI to quantify the
heterogeneity level: stress and
strain fields, stress triaxiality
and Lode parameter, rotation
angle, identifiability metrics,
DIC errors
• A heterogeneous test to
calibrate a viscoplastic model
• Synthetic images generated
with MatchID
• Investigation of the viscoplastic contribution
from hydraulic bulge test, DP600, thickness
0.8 mm
• Strain rate ranging from 10-3 up to 80 s-1
• Hydraulic bulge test, in quasi-static and
dynamic conditions (Hopkinson bars), strain
field measure using MatchID software
VISCOPLASTICITY HETEROGENEITY
15. Design of new heterogeneous specimens
Linear analysis; id = 0.0359 Nonlinear geometric
analysis; id = 0.0190
Nonlinear material and
geometric analysis; id = 0.0215
[6]
[7]
16. Thermomechanical heterogeneous test
Objective: developing a test with a combined
heterogenous temperature and strain field using a
Gleeble system
Multiphysics – coupled thermal-electrical-structural analysis
Select the best specimen and set-up
18. Status of UMAT structure in the VFM module and
integration with UMMDP/Fortran communication
D4.1 A large set of material user routines validated for commercial FEA software:
UMMDP has Fortran legacy (old Fortran)
Intermediate HUB
19. Status of Implementation of constitutive models in
UMAT routines
D4.1 A large set of material user routines validated for commercial FEA software
• Fully implemented for stress reconstruction
• Integrated within the Virtual Fields Module
• Youtube Tutorial XXXVII
• Stresses returned in global and material frame
• Fully parallelized
• Available in MatchID 2022.2
• Demo
20. Constitutive model: both plasticity and damage
behaviour depend on stress triaxiality and Lode angle!
Plasticity and damage behaviour
21. The power of virtual
experimentation (DVT) in MT2.0
23. Digital Virtual Twin (DVT)
Lava, P., Cooreman, S., Coppieters, S., De Strycker, M., Debruyne, D. (2009).
Assessment of measuring errors in DIC using deformation fields generated by plastic FEA.
Optics and lasers in engineering, 47 (7), 747-753.
25. 3D plastic anisotropy though FEMU and DIC
Denys, K., Coppieters, S., Seefeldt, M., Debruyne, D. (2016). Multi-DIC setup for the identification of a 3D
anisotropic yield surface of thick high strength steel using a double perforated specimen. Mechanics of Materials,
100, 96-108.
35. The power of experimentation
in data-driven approaches
Numerical data meets real data through virtual experimentation
36. What’s the fuss about AI & data-driven approaches?
• Inverse methodologies for full-field measurements
Data-driven and Machine Learning methods
characterization by Dall-E
37. What’s the fuss about AI & data-driven approaches?
• Inverse methodologies for full-field measurements
Dall-E:
Full-field deformation measurement tools
designed for a complete engineering
analysis
MatchID global
user meeting
pencil drawing of
a finite element
analysis
38. • Builds the inverse problem model as non-linear
regressor (using data)
• Build the inverse model LLM and finds directly the
parameters solving A = LLM(εexp,Fexp).
• The inverse model LLM is boundary conditions’
dependent
• High computational cost for the creation of a suitable
samples’ database (direct problem database)
• A simulation software is required
Inverse methodologies for full-field measurements
General goal
for parameter identification
Data-driven and Machine Learning (ML) methods
The objective function can be written as
where
39. The objective function can be written as
Procedure for material parameter identification
40. Constitutive parameters Input Space
σ0 [MPa] 120-300
n 0.1-0.3
K [MPa] 280-700
R0 0.6-2.5
R45 0.6-2.5
R90 0.6-2.5
The case of the biaxial cruciform test
ML approach for the Cruciform test with swift’s hardening and Hill’s anisotropy
Cruciform tensile test
Total displacement ux=uy = 2 mm
405 CPS4R elements (reduced integration)
Anisotropic plastic behaviour (Hill 1948)
Latin Hypercube Sampling – 2000 simul.
R = 7 mm
30
mm
0x RD
0y TD
R = 2.5 mm
15 mm
ux = 2 mm
Symmetry boundary conditons
(uy = 0 mm)
Symmetry
boundary
conditons
(u
x
=
0
mm)
uy = 2 mm
42. XGBoost
Model
σ0
n
K
Strain xy
(20 time steps *
405 features per time step)
Strain yy
(20 time steps *
405 features per time step)
Strain xx
(20 time steps *
405 features per time step)
Total force 0x
(20 time steps *
1 feature per time step)
Total force 0y
(20 time steps *
1 feature per time step)
R0
R45
R90
24340 features 6 Outputs
In/out features definition for the ML inverse model
However…
45. XGBoost
Model
σ0
n
K
Strain xy
(20 time steps *
114 features per time step)
Strain yy
(20 time steps *
114 features per time step)
Strain xx
(20 time steps *
114 features per time step)
Total force 0x
(20 time steps *
1 feature per time step)
Total force 0y
(20 time steps *
1 feature per time step)
R0
R45
R90
6280 features 6 Outputs
In/out features definition for the ML inverse model
Becoming …
46. Testing result
50
σ0 n K
R0 R45 R90
Test – 200 samples
2000 samples were generated, 1800 for training and 200 for testing
48. DIC levelling: the power of DVT
Comparison of the displacement
magnitude in the cruciform test,
between FEA (left) and virtual
experiments (right).
49. Digital virtual tests (DVT) database
ML database for training
ML architecture
Synthetic
image
generator
FEA program
(e.g. Abaqus)
?
?
?
?
?
Each sample was divided into 6455 subsets. In each subset, the location in
pixels and the strains εxx, εyy, and εxy were recorded.
2000 samples were generated, 1800 for training and 200 for testing.
51. … resulting in more than
400k features for each
sample.
To validate the models,
2000 images were
generated (1800 for training
and 200 for test).
only the regions of interest found with the SHAP values
were used to train the models (just 10% of the data)
number of
features was too
large
impossible to train
the model with
every feature
Digital virtual tests (DVT) database
52. Performance evaluation
σ0 n K
R0 R45 R90
Predictive
Simulated
R2=0.9980 R2=0.9610
R2=0.9779
R2=0.9779
R2=0.9923 R2=0.9925
Predicted (vertical axis) vs. real (horizontal axis) values of the constitutive
parameters
2000 samples were generated, 1800 for training and 200 for testing
53. DVT database results
• Results: virtual experiments database
• ML approach - Cruciform test
R2 and MAE for Abaqus vs MatchID database
The results are
still robust
even with DIC
noise/errors!
54. Robustness Analysis: material rotation
0
2
4
6
8
10
12
14
16
18
0 5 10 15 20 25 30 35 40 45
Relative
error
(%)
Rotation ( )
Y0
K
n
r0
r45
r90
Relative error in the
prediction of the
material parameters as
a function of the
rotation of the material
axes w.r.t. the global
axes system.
In this context, one of the samples
from the testing set was chosen
(Y0=172 MPa, n=0.16; K=486 MPa;
r0=2.38; r45=1.8; r90=1.06) and the
orientation of the material axes was
varied w.r.t. the global axes system
0xy
P.A. Prates, J. Henriques, J. Pinto, N. Bastos and A. Andrade-Campos, Coupling machine learning
and synthetic image DIC-based techniques for the calibration of elastoplastic constitutive
models, Esaform conference 2023
55. On the power of virtual
experimentation in MT2.0:
a VFORM-xSteels outlook
Sam Coppieters, A. Gil Andrade-Campos et al.
MatchID Global User Meeting
On the power of virtual experimentation in MT2.0 : a
VFORM outlook
22 February2023 | Southampton, UK
Editor's Notes
Decades ago, when designing engineering structures, such as bridges, planes and automobiles, the development required many engineers and time.
Today, everything is done on the computer, using simulation tools, reducing delivering time and resources. However, these simulations are only as good as the data & modelling they use.
We need to increase the speed and accuracy of development and production.
However, we need realistic engineering simulations, which needs accurate material behaviour reproduction that is only obtained with sophisticated and complex models!
However, these models required full material knowledge, that is cost and time consuming.
You know how these classical tests work, today.
There are many ways of testing the materials. As an example, the material is stretched until it breaks providing a single or few data points in time.
Here, the dots represent the data points. Several different tests are required to fairly characterize the material behaviour. It is not only a matter of quantity but also quality.
This confirmed limited data (obtained from such procedures) is not enough to make the sophisticated models work… Resulting….
Resulting… in errors in material behaviour predictions.
Therefore, the need for a solution was easily confirmed by software users in the field.
This vForm-xSteels project brings a solution…
The main goal of the project VForm-XSteels is to develop an accurate methodology for material characterization and determining the material parameters of thermomechanical models, from a dedicated single test that involve non-homogeneous temperature and strain fields.
A database and online library with calibrated material constitutive models, particularly for AHSS, is also developed.
we are designing a new test with a unique shape that, when stretched and using optical technology, generates more than 1 000 data points from a single test.
This data is processed using optimization algorithms and data analytics, providing a fast solution, accurate and robust technology, flexible for a wide range of steels.
This is something that no others have yet provided.
Cruciform test
Swift Law + Hill’48
Cruciform test
Swift Law + Hill’48
As observed in the previous case study, σ 0 and K depend almost exclusively on the initial and last force, respectively.
Only a few elements located at specific regions of the cruciform sample are important to predict R0, R45 and R90.
R0, R45 and R90 also use only a tiny region of the cruciform to obtain excellent results. Each parameter needs just up to 50 features to obtain a model with almost the same precision as the one with every parameter (6020).
Results with 100 best features
As good as obtained with every feature
Results keep getting better with more training samples
Simulated vs Predicted
Good results!
Since the number of features was too large, it was impossible to train the model with every feature. To solve this problem, only the regions of interest found with the SHAP values were used to train the models (just 10% of the information is used to train each parameter). Better results can be expected using more information, but …
Since the number of features was too large, it was impossible to train the model with every feature. To solve this problem, only the regions of interest found with the SHAP values were used to train the models (just 10% of the information is used to train each parameter). Better results can be expected using more information, but they are still great and can be compared with the Abaqus database using the same number of samples. Another way to improve the results would be to use the same number of samples used before to train the models presented above.
Since the number of features was too large, it was impossible to train the model with every feature. To solve this problem, only the regions of interest found with the SHAP values were used to train the models (just 10% of the information is used to train each parameter). Better results can be expected using more information, but they are still great and can be compared with the Abaqus database using the same number of samples. Another way to improve the results would be to use the same number of samples used before to train the models presented above.
shows the relative error in the prediction of each material parameter, for rotations of the material axes ranging between 0º and 10º w.r.t. the global axes system (the 0º case is taken as reference for calculating the relative error in predictions); this figure also includes an extreme case of 45º. In general, the parameter predictions remain robust regarding possible sample misalignments in-between 0º and 10º; in this interval, the maximum variation of the relative error occurs for parameter r0 (about 2.65%). This preliminary robustness analysis suggests that sample misalignments do not significantly influence the predictive performance of the ML model, regardless of the overall quality of predictions obtained for each parameter (poor quality for parameters n, r45 and Y0, which requires careful analysis to assess the cause).