Composite Modelling of different RVE sizes
ฝัง
- เผยแพร่เมื่อ 4 ก.ค. 2024
- This video is an #ABAQUS tutorial investigating the effect of Representative Volume Element (RVE) size on composite modelling. The composite material studied is a unidirectional composite and three RVE sizes were investigated. The study revealed that RVE size does matter especially in the post-yield or damage evolution part of the stress-strain plot.
Virtual domain of the randomly distributed fibres was created using #MontCarlGen2D which is a software that is developed by CM Videos. Also, periodic boundary conditions were imposed on the domain using another CM videos developed software called #PBCGenLite. These software (where available) can be developed using the link below.
Contour plots were generated from the study which showed different crack evolution patterns as you change from one RVE size to another. Also, stress-strain plots from the study showed the quantitative evidence that RVE size does matter both in elastic and onset of yield (though minimal) and also post-yield response (quite significant). The implication of this study is that modellers must objective determine the appropriate RVE size that is suitable for a given simulation not just for elastic properties but also post-yield response of a test material. #cmvideos #abaqus #rvemodelling #udcomposites
Time stamps
--------------------------------------
You can quickly navigate through the videos using these chapters.
00:00 Intro
01:01 Virtual domain
01:21 Analysis Framework
01:24 MonteCalGen2D v1.0
01:54 PBCGenLite v1.0
02:27 Material: GFRP
02:50 Case studies investigated
03:08 ABAQUS + MontCalGen: Virtual Domain creation
08:34 ABAQUS: Extracting History variables from Corner Nodes
11:28 PBCGenLite: PBC application
13:10 MATLAB: Submitting ABAQUS jobs within MATLAB
17:00 ABAQUS: Visualize Results
19:02 Principle of finding stress and strain from simulations
21:04 ABAQUS: Extracting stress and strain from simulations
25:58 Comparison of stress-strain plots
26:30 Answer: "Does RVE size matter?"
27:50 Outro
--------------------------------------
--------------------------------------
📥* DOWNLOADABLE RESOURCES
Please click on this link for Python Scripts, Excel spreadsheet and ABAQUS model used in the video. tinyurl.com/fwapked5
--------------------------------------
📥* DOWNLOAD LINK TO SOFTWARE
*PBCGenLite: A software for implementing Periodic Boundary Conditions on 2D RVEs (cmvideos.ck.page/products/pbc...)
*MontCarlGen2D: A software for creating random inclusions in 2D RVEs. (COMING SOON)
✔️*RELATED VIDEOS
💻: RVE Modelling of Unidirectional Composites in ABAQUS: • RVE Modelling of Unidi...
💻: How to manually apply Periodic Boundary Conditions in ABAQUS • How to manually apply ...
💻: How to AUTOMATICALLY apply PBCs in ABAQUS using PBCGenLite • How to AUTOMATICALLY a...
💻: MontCarlGen2D: A software for implementing the Monte Carlo Methods for Micromechanics. (Video coming soon)
✔️*Link to Michael Okereke's Finite Element Applications textbook
--------------------------------------
Okereke, M. and Keates, S., 2018. Finite element applications. Cham: Springer International Publishing AG. link.springer.com/book/10.100...
✔️*REFERENCE JOURNALS CITED IN THIS VIDEO
--------------------------------------
* Okereke, M.I. and Akpoyomare, A.I., 2013. A virtual framework for prediction of full-field elastic response of unidirectional composites. Computational materials science, 70, pp.82-99. (tinyurl.com/4bjz6tyj)
👨👩👦 CM Video Insider
--------------------------------------
Become a CM Video Insider (100% free) to access modelling resources, advanced workshops, bonus trainings, behind the scene contents and regular CM Video updates: cmvideos.ck.page/16a99bd0e3
🔔 Subscribe
--------------------------------------
If this is the kind of content you like, please do subscribe to this channel and click the notification button so when contents like this are made, you will be the first to see it. If you also have ideas of videos you want me to make, do suggest those in the comment section and I will consider them. - แนวปฏิบัติและการใช้ชีวิต
Great job, thanks lot Dr. Michael.
Thanks Nasser.
Hi I'd like to ask if it is possible to carry out high velocity impact on the RVE model using Hashin damage?
Yes, you can always do a high velocity impact test on an appropriate RVE. I am not sure if this is the right type of RVE you should be using. I suppose you are suggesting a high velocity plate bending test, in which case this will not be the right RVE to use. With respect to Hashin damage, no, this will not be the right RVE since hashin damage works at a macroscale representation of the laminate. What we have here is a microscale representation, so due the the length scale difference, it therefore means that this is not the right or appropriate RVE for such study.
Hi sir. can you do video on Directionally distributed steel fibre reinforced concrete
Will try
Thank you for this fantastic video. I have one question. How to build in Material Library (4:42 clip video) in the ABAQUS? I hope you demonstrate this in the video. Can you introduce how to use MontCarlGen2D?
Hello @Enidentictok, thanks for the comment. I actually do not have a video about creating a material library in ABAQUS but there is a good video by another academic content creator (see Dr Ahmed Elkady channel) and here is the link to the video about creating material library: th-cam.com/video/x0GDmGqKDdk/w-d-xo.html
With respect to a video introducing MontCarlGen2D, I already did and here is the link: th-cam.com/video/Zb6ytX6oRJc/w-d-xo.html
Let me know if there is any more information you want.
Great job professor Michael, it's valuable. I would like to ask you if is it available to do the fiber matrix interface I think it's interesting issue
Hello Eng Thanks for your interest in the channel. I actually am working on a video with interface so keep coming back for that.
@@MichaelOkereke happy to hear this! keeping an eye
Very interesting video but MontCarlGen2D software isn’t available yet
I know John, I have been promising the MontCarlGen2D software for a few weeks now. I am actually fine-tuning the code to make it more user friendly before launching it to a wider audience hence the delay. I am almost done with it. I am also recording two-sets of
videos that will go with the software - so these should be out soon on the channel. Please keep watching out for those. Thanks for your patience and in the next few weeks, the software will be there for you to download and use.
Very interesting video Dr Micheal
I would like to ask about the weight percentage (wt%) of inclusion in 2D . suppose i have 1000x1000 nm rectangle and size of inclusion is 80 nm, is there any formula or description which we allow me to include 3wt% of inclusion?
Hello @Muhammad Adkan Pke183037, thanks for your interest in the channel. I had this query two weeks ago from someone in my network. I am going to make a video about it so please do keep coming back to the channel to see when the video is released. There is definitely a way of calculating volume fractions from weight fractions - which I believe is the essence of your query.
Question: It is interesting that the values of PE,Max.In-plane Principal are used for calculating Young's modulus. Is it because of 2D model and is it really to represent Young's modulus or shear modulus in this video?
Hello @eslingena, no I have not calculated the Young's Modulus using the maximum plastic strain, PE. It was shown simply for the contour plot. The actual calculation is based on reaction forces at the retained corner nodes.
@@MichaelOkereke Thank you for your kindness! I am still not clear of your answer. In your 2D RVE model, how to get an effective Young's modulus?
Dear sir, I have a question , when you determine stress, i think we should use 'V' and 'x" in the current deformed configuration beacause of principle of virtual work. but why you use 'V' and 'x' which is in initially undeformed configuration.
Dear Chao, thanks for the comment. I think you might have misunderstood what happened here. The x and F terms were tracked using 'history outputs' in ABAQUS throughout the duration of the test. That means that the different stages of deformation (or as you describe it) configuration of the model is noted. It is not enough simply to evaluate the initial or final configuration, for such simulation. All stages of the deformation are essential.
As regards the V term, this is a volume term, and it remains constant even at the start and end of the simulation. This is because there is no material loss in the model (no damage) hence we can assume that the initial volume resulted in the same final volume.
The equation of volume-averaged stress used in the model depends more on the coordinate positions of the corner nodes and their reaction forces, and it is not the traditional force divided by area term that we know always to be the case with nominal or true stresses.
The above makes the arguments of the video correct. You can learn a bit more about this sort of derivation of stresses in a PBC-enabled domain from this video: th-cam.com/video/qHt-Mn3nuiQ/w-d-xo.html
Dear sir, will this PBC also be available for a 3D model for complex structures, and what if the case is aperiodic, thank you sir.
Hello Vivek, thanks for your interest in the channel. I am thinking making a video on this issue. Please keep watching out for it.
@@MichaelOkereke Ok Dear sir thank you.