How to manually apply Periodic Boundary Conditions in ABAQUS

[@gideonlyngdoh3538]

Great video! Thank you, Dr. Okereke. I have few questions: 1. If I tie the internal nodes of one side to a reference point and then define the equation on these reference points, will that solve the problem instead of selecting individual nodes? 2. Is it necessary that the stress or strain should be periodic in nature after deformation? From the video, it is seen that the stress is not identical between the front face and the back face, and similarly with the top and bottom face.

[@MichaelOkereke]

Hello Gideon, thanks for your continued interest in the channel. I will attempt to answer your questions: 1. If you tie the internal nodes to one reference point and define the canonical equations, this will violate the periodicity argument. What you are doing with the tie is you are forcing all tied nodes to take their behaviour from the reference node. This is not what periodicity require. The beauty of PBC is that individual nodes have freedom to behave as they should in the macroscale bulk provided it has a linear correspondence between that node and the one directly opposite it on the other edge. So, simple answer is No! 2. Similar to the same explanation as 1 above, the stress and strains will not necessary be the same but will be periodic. Think about it this way, the stiffness between the two nodes allows for equivalent stress-and-strain values between the two PBC-connected nodes. Away from them, a different stiffness exists, and so the stress can be different. So, although there is a periodicity of stress between the two nodes, there is still no evidence of that when you study the model as a whole since the stiffness across a heterogeneous model will be different across the microstructure. I am not sure if I have confused you further but the answer is Yes in a local node-to-node (or element-to-element) comparison but globally, it is not.

[@johnsonezenwankwo3497]

@@MichaelOkereke Thank you, Dr, for this particular answer. Some had tried to sell me this idea as PBC, although I tried it and it didn't work - though it looks more like what you had done in Dirichlet BC (in a later video) or is it somewhat different?

[@yifanzhi2648]

Thank you very much! Until now, I start to know what is RVE and PBC and have some direct experience by your tutorials.

[@MichaelOkereke]

Thanks @Yifan. Glad it's helpful.

[@carlosmedina2711]

Very useful video! You will help me a lot with my research. Please keep posting videos!

[@MichaelOkereke]

Thanks Carlos, glad you found the video helpful. I will in deed keep posting. Is there any video ideas you would want me to cover?

[@carlosmedina2711]

​@@MichaelOkereke Thanks for your response, Dr. Michael! I have a question, if we have a cube with some inclusions for a coupled temp-displacement analysis, we restrict three of its faces and then apply temperature, ¿how could we couple or tie the nodes from one face so this nodes have the same displacement for the direction normal to that face? Considering this for the remaining three faces.

[@muratanaka9547]

Thank you fo sharing the application of periodic Boundary Condition and for me your videos is the best and awesome in the field of multiscale modeling please keep doing Dr Michael Okereke

[@MichaelOkereke]

Glad you like them!

[@musbauyusuff]

Thank you Dr sir

[@MichaelOkereke]

You are welcome

[@awfsea]

Thx a lot. You help me a lot in this video !

[@MichaelOkereke]

Glad it helped!

[@minerva5435]

thank you for your video Dr. Okereke! i did not understand one thing : why did you use nodes? is it for forming the structure by the layers hidden under that node? what is the relationship between periodic boundary conditions and using nodes?

[@MichaelOkereke]

Hello Minerva, I am confused with your question. It is not clear to me what you mean with nodes and PBCs. I think maybe the video need to be rewatched to see if you understand better.

[@minerva5435]

@@MichaelOkereke I was asking what are the nodes used for?

[@MichaelOkereke]

Right @minerva, the nodes are the key anchor points for applying the loads and boundary conditions on the model. When a system has periodic boundary conditions imposed on it, nearly all the boundary nodes have the PBC constraint applied on them excluding the retained nodes. These are the nodes you refer to. They are the 4 corner nodes for this 2D system. With those, I can then decide on how to apply a fixed support or a tensile load on the structure. They are also the nodes that are used as reference nodes when deploying the canonical equations that enforce periodic boundary condition on the model. I discussed these more in my Finite Element Applications textbook, see link in the description, and then read those to understand more about this.

[@minerva5435]

@@MichaelOkereke thank youu :)

[@NeerajSharmaResScholarCivilEng]

heartiest thanks for this valuable video, I am struggling with 3D lattice structures. If it is possible for you please make a video on 3D lattice structures with PBCs so that I can get some idea of how to apply PBC in 3D lattice structures.

[@MichaelOkereke]

Hello @Neeraj Sharma, thanks for the comment. I am of course interested in making such a video but that will be in the future. Treating PBCs in 3D is not so straightforward so I will make a series of videos abou this sometime in the future and hope that will help. Please keep watching this space.

[@MultiSaded]

Thank you Dr Okereke for the videos you provide regarding the study of composite materials, for the case of elastoplastic material, how can periodic boundary conditions be applied? Is this the same as the elastic case? it is a matter of plotting the stress-strain curve.

[@MichaelOkereke]

Hello @Saad bouta, thanks for your kind comments. Yes, indeed, if you want to model an elastoplastic material, it will still be the same approach as the elastic case. In fact, you will even get better predictions from your simulations (in terms of distortion of the domain) for the elastoplastic case in comparison with the elastic case. One thing that will definitely help is to have an RVE with different constituents in it, for you to see the excellent periodicity of deformation that typically shows a model with PBC imposed on it. I hope this helps.

[@MultiSaded]

@@MichaelOkereke Thanks for the quick reply. in this case, how do I draw the stress-strain curve, how do I choose the elastoplastic model (like ex: voce, ludwik), and do I need to know the elastic limit.

[@MichaelOkereke]

Hello @Saad, I have a few video with the answers you need. Check out my videos about strain-stress plots.

[@johnsonezenwankwo3497]

Thank you Doc. Is it possible to do this with a 3D unit cell consisting of 4 yarns in a simple weave?

[@MichaelOkereke]

Hello @Johnson, I believe this is possible but not with the exact method that I showed here. Dealing with 3D RVEs and PBCs involves a different approach. In future, I will like to make videos to show how to do so. Thanks for your query.

[@johnsonezenwankwo3497]

@@MichaelOkereke So your model in this case is a planar surface?

[@MichaelOkereke]

No @Johnson, it is a solid continuum element model.

[@johnsonezenwankwo3497]

@@MichaelOkereke Thank you, Sir. I'll be waiting for the video. I'm a final PhD student in Czech Republic

[@쳬이소톤졸업생]

Hello, Dr. Will I obtain the same values in strain as well as in displacement, too, after applying these PBCs?

[@MichaelOkereke]

Yes you should.

[@쳬이소톤졸업생]

@@MichaelOkereke Hello Dr. But why, after applying these PBCs, I obtained the same value of each pair of nodes only in displacement, not the same value in strain and stress as well?

[@nuraqilakadirhussein4490]

Great video! Hi, Dr. Michael, can you share on how to simulate porous elastic material that consists two type material?

[@MichaelOkereke]

Hello @Nur Aquila Kadir Hussein, thanks for your comments. I have some videos on the channel about modelling of porous media. Have you seen them? I am not sure what is unique about the two-type material you reference here? Please give me a reference to a paper or information about this so I can understand what you mean and think about how to build such a model. Thanks.

[@ioannisalakiozidis2518]

Great video and very clear explanation Dr. Okereke. I want to do a similar thing with my model, which is a homogeneous material instead, however, I want to model a small region with few pentagonal grains and apply PBC so that they remain in-plane while tensile stress is applied. Do you think that this is something that can be done in the same way? And another question, I didn't understand why for the x-direction tensile test you applied displacement just in the bottom right corner node - why not at the top right as well?

[@MichaelOkereke]

Hello @Ioannis Alakiozidis, thanks for your kind comment. I believe you can still use this approach for your material. If I understand correctly, it is a homogeneous material, but the sub-region you want to model is heterogeneous. This is still possible with the PBC approach I describe here. The only condition I think you should worry about it the pentagonal grains that you want to remain in-plane (in otherwords), you do not want them to rotate during the deformation. If you do not constrain them, following the periodic boundary condition deformation, they can rotate (especially if the matrix that holds them together) shows finite deformation. It does sound an interesting problem. I will suggest you contact me via my CM Videos Insider Group, (you receive a welcome email when you subscribe) using this link: cmvig.cmvideos.org/. We can then discuss specifics of your problem.

[@ioannisalakiozidis2518]

@@MichaelOkereke Great, I subscribed and sent you an email with more information - I hope this is the right way to contact you. If not, please let me know how to get in touch with you to discuss some more details

[@joonabil]

Very interesting video but I would like to know is the only matter thing in order to apply PBC that the number of nodes for both sides shall be the same without caring about having symmetrical mesh which is somehow impossible especially when we have random inclusions like your model. Thanks in advance

[@MichaelOkereke]

Hi John, thanks for your continued interest in the channel. Yes, you are correct, we need to focus on having same mesh on both sides of the domain. If not, then we cannot apply the current method I showed in the video. This is what I referred previously in the **Theory of Periodic Boundary Conditions video* ,* as periodic mesh. For where random inclusions exist, like ours, you have to be careful how you mesh as well to make sure you have same number on both sides. The trick is to use the edge seed function in ABAQUS to make sure you have same number of seeds on parallel edges/faces. This is the principle.

[@alifallahi8323]

Great video! Hi, Dr. Michael,can I use this code for tensile Sample?

[@MichaelOkereke]

Yes you can @ali fallahi! You just have to identify the correspondeding edges: Left, Right, Bottom and Top and apply the PBCs on them as shown in this video. However, the method is better suited for analysis of micromechanical systems where an RVE can be isolated for tthe study. You will better use a Dirichlet Boundary contition in this instance rather than using PBCs.

[@johnsonezenwankwo3497]

Is the PBC kinematic linking to be performed on corresponding corner nodes only or other face nodes?

[@MichaelOkereke]

Hi, the corner nodes are not usually kinematically tied as they are used to specify the required loading type that will enforce/initiate the PBC. The face/surface nodes are the ones with PBC and are kinematically tied.

[@likk-k9y]

Great video! Hi, Dr. Michael,I have a difficulty. I want to do a compression test on a periodically arranged lattice, and the whole lattice will undergo great deformation. Can I take a single cell out of it and apply periodic boundary conditions to it to reduce the amount of computation? If so, how can periodic boundary conditions be imposed on this cell to satisfy the same deformation in the lattice?

[@MichaelOkereke]

I think you can do that. If you want to do it in 2D, then here is a video that can help from my channel: kzbin.info/www/bejne/aoqnn36lnL-ShMksi=XmUkCzyKVQ3aK4tG In 3D, that is different and difficult. I will hope in future to make a series of videos explaining how this can be done.

[@lucassanchezfellay4344]

Hi Dr. Okereke, first I want to thank you for your videos, they are very helpfull. I have a question about periodic boundary conditions, Why don't you apply the "second group" (base and top) of constraints to the nodes 8 and 5? You did apply it to first group (back and front) for 8 and 9 nodes. Probably is a simple question, thanks in advance. Lucas

[@MichaelOkereke]

Hello Lucas that node is a slave node so it follows either the behaviour of the first group or the second. That's why I didn't use it for the second group since I did for group one.

[@lucassanchezfellay4344]

@@MichaelOkereke Thanks for your response Michel. But now I need to ask you another question, what does slave node mean? Is a node with no boundary conditions? Sorry again for the basic questions, I'm newby. Regards

[@alibehbahani4656]

Thank you so much Doctor. I checked this video. But still I have my question, How can Extract homogeneous mechanical property after implantation of this PBCs. Best, Ali

[@MichaelOkereke]

Please explain what you Mean? Are you looking to get stiffness and strengthen?

[@alibehbahani4656]

@@MichaelOkereke Hi Dr. Okereke, thanks for your kind response. In exact words, I had made a random field by assigning random material properties to each element, now I want to find the homogeneous Young modulus and Poisson's ratio of this random field. What should I do?

[@farzanehabedini5723]

Thank you for the great video. I was searching a lot about periodic boundaries and I found some related papers. They used kinematic coupling for lateral boundaries and called it periodic boundary. What is the difference between these 2 methods? Can you tell me how can I use kinematic coupling for sides of my model? Thanks

[@MichaelOkereke]

Hello @Farzaneh, thanks for the query. I do not know much about lateral boundary conditions but from my quick research into this, I think it is used in ocean studies, in fluid systems where you need to match the response of a given face/edge of a domain to a corresponding, directly opposite edge. If this is the case, then the approach of using a kinematic coupling to link the faces is about right, or sort of acceptable. This is because for a fluid domain, the is really no heterogeneous material (I hope) between the edges, it is simply a fluid flow. In that case, it is okay to use a kinematic coupling of the Tie constraint category to link the faces. For the Periodic Boundary Condition as described in my video, this is different. It does not require that there is homogeneity between the edges/faces. It is a none-to-node coupling rather that a face/edge set coupling as I will expect is the case for lateral boundary conditions. They are two different things. It really depends on the problem you are solving. If you are in the heterogeneous microstructure domain of material science, then my type of PBC will be applicable. If you are in ocean studies, then a lateral boundary condition, with the kinematic coupling (tie or MPC constraint variant) will suffice. Let me know if this helps.

[@pu7358]

Hello sir, do you know if this method is also applicable for heat transfer simulation, i.e., I am trying to impose PBC for heat transfer problem instead of a mechanical problem. Thank you!

[@MichaelOkereke]

I have not tried it on thermal problems but I think it should work too.

[@RajnandiniDasedd]

First of all hank you for such wonderful video. I have few questions. I am working with lattice structure. So the PBC which you have applied here, will this be same in case of lattice structure?

[@MichaelOkereke]

Yes indeed. I will make a video about this in future.

[@RajnandiniDasedd]

@@MichaelOkereke It would be really helpful sir if You could make a video on applying PBC to lattice structure and extracting the homogenized elastic properties as soon as possible. It's a request.

[@MichaelOkereke]

Hello @Rajnandini, I have made this video. I did not get to extract elastic properties but I suggested a video you can view to see how to extract properties from results in which PBCs have been implemented. I hope it works for you.

[@RajnandiniDasedd]

@@MichaelOkereke Thank you so much sir. I have seen your video and liked it. It's really helpful.

[@Aman-bj9he]

Thank you very much sir . It was really helpful.

[@MichaelOkereke]

You are most welcome

[@subratkumarmaharana213]

Thank you verymuch sir, it is very helpful

[@MichaelOkereke]

You are most welcome. Watch out for the automatic way of doing the same .. a video coming up soon.

[@jyothilekshmijayan2815]

Sir, how to provide viscous absorbent boundary condition for seismic analysis of soil medium

[@MichaelOkereke]

Hello, thanks for your comment. I unfortunately do not know anything about "viscous absorbent boundary conditions" and so unable to help with this query. Good luck with the research.

[@ammarvanank]

Good Morning Sir, I have a query about non absorbing boundary condition for Dam- Reservoir- Foundation Interaction... how to create Lysmer - Kuhlemeyer’s viscous boundary for foundation in Abaqus software for foundation? Please help me out...

[@MichaelOkereke]

Hello, thanks for your query. I do not have any idea about this type of boundary condition so unfortunately unable to help you. Good luck with the research.

[@alokdwivedi1949]

Hello sir this one is really very useful content to understand how to apply the periodic boundary conditions manually I want to learn for the case of 3-d but manually it is next to impossible by selecting each node and by creating a set after that applying constraints. Can you please let me know the source where I can learn or can you help me by creating a video on how to apply periodic boundary conditions for 3-D models?

[@MichaelOkereke]

Hello @Alok dwivedi, PBCs for 3D domains is not trivial and clearly not possible using the manual methods described in this video. I published about this with my student awhile ago so you may be able to learn 3D PBCs implementation from this publication: doi.org/10.1016/j.compstruct.2016.10.114

[@singingfingers3128]

Hi Dr. May I know why you use nodes instead of edges (2d case)? If you were to apply PBCs in 3d, will it be the same? I’m really sorry to ask these questions as I’m not familiar with Abaqus. When I use Ansys, I can just select the edges (2d) or faces (3d) in order to setup the constraint equations for the PBCs. However, your tutorial is really helpful. I really hope you can do it in 3d as well. I believe I’m not the only one who request that tutorial. Thanks

[@MichaelOkereke]

Hello @Singing Fingers, the reason I use nodes instead of edges is because of the requirement of Node-to-node coupling expected in classic definition of the Periodic Boundary conditions. If you use the edge-to-edge coupling, what you are doing is that you are forcing on the nodes on that edge to be kinematically tied to the corresponding directly opposite edge. This can be a problem for heterogeneous systems where you might have a fibre deforming less than a matrix, if you constriant them to deform by edge to edge coupling, then the deformation might be incorrect. There may be instance where the results will be correct (but it is intrisically flawed in my view and you should be careful how you use it). For the 3D case, this is something I am thinking about. It is not trivial and something I want to teach extensively about first before showing how this is done in ABAQUS. It is somethign I am keen to do but I just need to find the time to do it. I know certainly that it will benefit many but it is not a trivial exercise. I do not think there is a video anywhere on KZbin abou the 3D implementation of PBCs.

[@DeepikaKushwah-h2u]

Nice video sir. I have followed the same procedure considering the same problem but in case of tensile loading in x-direction I am not getting same displacement (U2) for top and bottom edges. Also, there is non-zero strain in y-direction, with the implementation of PBC but it should be zero, isn't it?

[@MichaelOkereke]

That means your kinematic linking is wrong. If you follow every step properly, you'd certainly get the same results. Please check your solutions and implementations again. On zero y-strain, that's not true. You should unless you mean something different. The stresses across parallel boundaries are mutually opposite, but there is no case when zero is expected.

[@jiyadferoke2600]

sir, i want to apply apply pbc on rve of concrete mesoscopic model. how i can mesh the system to get identicl node ponts on opposite edges sothat ican couple the node. provided that my opposite edges are different as my inclusions are randomly placed

[@MichaelOkereke]

Hello @Jiyad, This is a real issue for heterogeneous media where periodicity of material is not enforced. I will say my method here might not be effective, although you could get an answer, but you need to do some parametric studies to assess the effect. If you have to use my method, I suggest you use the seeding-by-edge function (under mesh module) in ABAQUS, to make sure the parallel edges have same number of seeds (divisions) to allow my method to be used. Good luck with the research.

[@jiyadferoke2600]

thank you sir

[@muhammetcuneytsakonder1361]

Thank you for your valuable videos. Is it possible to apply Periodic Boundary Conditions in 3D ? How can we do that ?

[@MichaelOkereke]

Hi @Muhammet, it is certainly possible to apply PBC to 3D virtual domains. I intend to make a video about this in future.

[@abdallarselbana3394]

Dear Dr Michael. I have applied the canonical equation to my 3D RVE. I have encountered an error when submitting a job. Each node on the edges (12 edges of the cube) will have two different equations for the two perpendicular directions of that node, the corner nodes would have more. Any suggestion on how to fix this?

[@abdallarselbana3394]

This is the error message " ***ERROR: DEGREE OF FREEDOM 2 DOES NOT EXIST FOR NODE 1 INSTANCE MATRIX-1. IT HAS ALREADY BEEN ELIMINATED BY ANOTHER EQUATION, MPC, RIGID BODY, KINEMATIC COUPLING CONSTRAINT, TIE CONSTRAINT OR EMBEDDED ELEMENT CONSTRAINT. THE REQUIRED EQUATION CANNOT BE FORMED."

[@MichaelOkereke]

Hello @Abdalla, thanks for the comment. This is quite normal, the error with edges. This is because an edge is shared by two faces (being intersection of two faces). So, using the classic form of the canonical equations (as a 2D case) will prove problematic as we do not really have such intersecting faces in 2D problems. I did encounter the problem you describe when I was learning about this. I will be doing a video soon on how to apply PBC for 3D RVEs as that will help you understand how to deal with edges, vertices as well as internal surface nodes. These need to be clearly distinguished and appropriate canonical equations will be implemented for them. Please keep coming back to the channel so you see the video when I launch it.

[@abdallarselbana3394]

@@MichaelOkereke. Thank you very much. Your book, papers and videos were quite helpful. I truly appreciate you.

[@adarshshah4309]

How to do the same in 3D model? What are the extra sets we have to generate? Did we have to generate Right face and Left face also?

[@MichaelOkereke]

Hello @Adarsh Shah, thanks for your interest in the channel. You are right about 3D RVEs, you will need to specify extra planar constraint equations matching one face to another. This has to be done for the six faces i.e. three pairs of faces. Also, edges will need to be accounted for too. However, that is about where the similarity between this and the 2D RVEs lie. The real challenge is how to make the connection between the faces as you will probably have non-periodic meshes. As a result, node-to-node kinematic coupling might not be possible. A different strategy is required. I have published on this here: doi.org/10.1016/j.compstruct.2016.10.114. I intend to make a video about this in future. So, please keep watching this space.

[@adarshshah4309]

@@MichaelOkereke Thank you very much for your reference and explanation. Best Wishes!

[@tonyyang8209]

Hey, what is the Periodic Boundary Conditions for 3D parts? thanks

[@MichaelOkereke]

Hello @Tony Yang, I will make a video in future about implementing PBCs on 3D domains. It is not trivial so requires careful teaching and demonstration. I will do this so please keep watching this space.

[@pavithrareddy3975]

Sir, can you please explain on how the PBC can be applied to a 3D model manually in abacus.

[@MichaelOkereke]

Noted. Thanks for the suggestion

[@josftxn]

How can work with perturbation displacement ínstead prescribed displacement that you do?

[@MichaelOkereke]

Hello @josftx, I am not sure I understand the question. It seems your question is around principle of virtual work that i have used in deducing volume averaged stresses in the 2D RVEs. If this is so, then all you will need is displacements at the corner nodes (prescribed) as I have done here. I understand theretically, the principle of virtual work requires displacements (incrementally perturbing the domain). However, since we are doing a numerical setup here, the prescribed displacements will be incrementally act on the model across the time steps of the simulation, thus agreeing with what the theory says. In my mind, the two are one and the same. I am not sure if this is what you were asking about.

[@josftxn]

@@MichaelOkereke some like that. My question is if always i need a prescribed displacement ? Why i need that?? can i use another way or iterative way for find the material response.

[@hosseinmalekinejad5950]

I get this error " 2 nodes are missing degree of freedoms. The MPC/Equation/kinematic coupling constraints can not be formed. The nodes have been identified in node set ErrNodeMissingDofConstrDef." but I exactly follow you . why the corner node are missing DOF

[@MichaelOkereke]

You must have missed a node in your coupling... without seeing the input file I will not know. Check the errNodeMissing set to find out which.

[@hosseinmalekinejad5950]

@@MichaelOkereke thank you for your prompt response I check the error nodes and when I use quad mesh two node has error one in the load applying place and the second on top left corner. but when I use triangular element one node has error and the one that load applied there

[@hosseinmalekinejad5950]

@@MichaelOkereke 🙄🙄🙄🙄🙄🙄😔

[@MichaelOkereke]

Hello Hossein, I am not sure what is going on here.

[@awfsea]

Hello I met the same problem. Have you solved it now?@@hosseinmalekinejad5950

[@hossemeddinesaidi2601]

Thank's

[@MichaelOkereke]

You are welcome @Hossemeddine.

[@Jerry-gx8ey]

Thanks for your video! I have learned how to apply PBC in Abaqus a weak, and I have funded a Python script to create the equation. But I want to calculate the lattice model, not the composite material. As you know, the lattice model always doesn't have the vertex. So do you know how to apply the PBC to the lattice?

[@Aman-bj9he]

Can you please help me to get that script to create equation Please reply @ [e-mail].

[@MichaelOkereke]

Hi Jerry, glad you found the video helpful. With regards to the lattice model, the approach shown here might not be adequate for it. I was working on a quadrilateral RVE with clear defined boundaries. For your lattice model, this is not the case. I suppose the better was to deal with it is to use same approach but for a non-periodic mesh. It is not something I have covered but maybe in future I could look into showing viewers how to do this. I will recommend you use the good old Dirichlet approach for your lattice model problem, for now!

[@MichaelOkereke]

Hello Aman, there script Jerry referred to in this video is simply a python script for recreating the virtual domain. It does not form the equations, unless Jerry was referring to another video/script. I will be making a new video where I can show how to automate the applications of PBCs. Please keep coming back for that video which will land soon.

[@Jerry-gx8ey]

@@MichaelOkereke OK Dr. Michael Okereke, thanks for your replay. I'll be looking forward to your next video！You have helped me a lot, thanks!

[@Aman-bj9he]

@@MichaelOkerekeyes sir got it thank you very much, looking forward to the upcoming video.

[@akshayr4348]

How do we define the boundary conditions for multiload direction? If I want to include all three loads in one job, then how do I define the boundary conditions?

[@MichaelOkereke]

Hello @Akshay R, thanks for your comment. I believe your query relates to multi-loading on the RVE. I will direct you to the video I made about how to generate failure envelopes here (kzbin.info/www/bejne/rl6sZpd8iad6a68) where I showed how to apply multiple loading using the PBC approach. This might be a solution to your query. Let me know if that helps.

[@akshayr4348]

@@MichaelOkereke thank you for the information.

[@tonyyang8209]

nice

[@MichaelOkereke]

Thanks

[@sushantsaurabh0611]

if you could upload a video regarding the application of PBC in 3D RVE, it would be of great help to the modelling community.

[@MichaelOkereke]

Hello @Sushant, thanks for the comment. The PBC in 3D RVE video is something planned for the future. I have this in mind but it is not available now. It is not a trivial exercise so I will have to find the best way to present the approach in order to be able to teach it. Watch this space.

[@subratkumarmaharana213]

Sir , Is there any way to extract the python script from the input file for abaqus??

[@MichaelOkereke]

Hello Subrat, I am not sure there is a direct way to extract Python scripts from an ABAQUS *.INP file. The input file is a bunch of texts that tells ABAQUS how to execute the FEA investigation. It is not possible, as I see it to extract a python script from that. Having said that, I do think you can generate a python script using the Macro function in ABAQUS CAE as you build the model. This is definitely something worth looking into. If there is appetite for this in the channel, I can make an ABAQUS tips and tricks video about this. I hope this answers you query Subrat.

[@subratkumarmaharana213]

@@MichaelOkereke Thank you sir for your response....I am pretty comfortable with macro....It will be great to see your videos...on tips and tricks

[@abaqususer2810]

Hi, there. Have you ever tried using PBCs in an abaqus explicit analysis? I encountered a problem that abaqus explicit would report an error when there were too many linear constraint equations applied. The error basically reads "there are over 1000 independent DOFs in the kinematic constraints, and the explicit can not solve the system efficiently". PS: when my meshes were coarse enough that there weren't too many constraint equations defined, the dynamic explicit can get the job done well. Could you please lend me a hand on this issue?

[@MichaelOkereke]

I have not seen this problem but will look into it and see if I can recreate what you describe. I do not think though this should be the case as I cannot see why.

[@abaqususer2810]

@@MichaelOkereke Appreciate your reply. If you are interested in discussing this problem further, I can send you my abaqus input file. It’s a unit cell model of 3d braided composite shafts. I’ve tried a simple cuboid model as well, but the problem remains. I could send you both if you will. By the way, the abaqus version I am using is 2018.