Concrete Microplane Model - FEA using ANSYS

Concrete Microplane Model - FEA using ANSYS - Lesson 10

Рет қаралды 15,193

2 жыл бұрын

This tutorial shows how to use the Microplane model to simulate a concrete cube and reinforced concrete beam. This implementation incorporates coupled plasticity-damage and nonlocal behavior. This is the first tutorial in this series incorporating Mechanical APDL commands into the analysis.
Learning objectives:
1. Sketch the Drucker-Prager yield surface and locate key points and features.
2. Explain the various parameters for the Microplane model and their typical values.
3. Add APDL commands to model complex material behavior and change element types.
=== Downloads ===
Drucker-Prager Yield Surface (Excel):
docs.google.com/uc?id=1CmO5iN...
(direct download link of Excel file, both psi and MPa units)
Microplane APDL Commands for psi (text file):
drive.google.com/file/d/1PToS...
Microplane APDL Commands for MPa (text file):
drive.google.com/file/d/1XC4u...
=== References ===
This particular implementation in ANSYS is based on the following references:
Zreid, I. & Kaliske, M. (2018). A gradient enhanced plasticity-damage microplane model for concrete. Computational Mechanics. 10(1007), s00466-018-1561-1. doi.org/10.1007/s00466-018-15...
Zreid, I. & Kaliske, M. (2016). An implicit gradient formulation for microplane Drucker-Prager plasticity. International Journal of Plasticity. 83, 252-272. doi.org/10.1016/j.ijplas.2016...
Zreid, I. & Kaliske, M. (2014). Regularization of microplane damage models using an implicit gradient enhancement. International Journal of Solids and Structures. 51(19), 3480-3489. doi.org/10.1016/j.ijsolstr.20...
The original microplane model was developed by Z.P. Bazant and others:
Bažant, Z. P., & Gambarova, P. G. (1984). Crack shear in concrete: Crack band microplane model. Journal of structural engineering, 110(9), 2015-2035. doi.org/10.1061/(ASCE)0733-94...)
Bazant, Z. P. & Oh, B. H. (1985). Microplane model for progressive fracture of concrete and rock. Journal for Engineering Mechanics. 111, 559-582. doi.org/10.1061/(ASCE)0733-93...)
Carol I, Jirásek M, & Bažant Z (2001). A thermodynamically consistent approach to microplane theory. Part I. Free energy and consistent microplane stresses. International Journal of Solids and Structures. 38:2921-2931. doi.org/10.1016/S0020-7683(00...

Пікірлер: 65

@Jmurr420 2 жыл бұрын

Cool channel! I appreciate you using a platform to share your knowledge. I took your Structural Analysis course at UW-Madison way-back-when, now I have my P.E. license and I'm working on a structural Master's at UMN-Twin Cities. Thanks!

@HamOn-ps1ym 4 ай бұрын

God bless you for bringing us this perfect video. This is de rosetta stone between ANSYS and the real world. Please more videos on Microplane Model and brittle materials.

@llloomm 2 жыл бұрын

This is great, thank you.

@StructuresProfH 2 жыл бұрын

You are welcome!

@amrmasoud5580 2 жыл бұрын

Dear Dr. Thanks Please tell me what are the advantages of MP model versus solid 65? regarding static, dynamic (as impact and blast )and cyclic analysis . thanks

@hanxiao3246 2 жыл бұрын

Nice video sir!

@StructuresProfH 2 жыл бұрын

Thank you, Han!

@xcs05 2 жыл бұрын

Thank you for this, a fantastic tutorial! I tried it and the only reason it didn't work for me is that I don't have the ANSYS Enterprise license. Unfortunately this will not work with the Premium license, which I believe does not support CPT215/216 elements. For now I'll have to stick with the Menetrey-William method (previous video). May I ask, does the inbuilt concrete NL work? It includes "Drucker Prager Strength Piecewise", "Tensile Pressure Failure" and "Crack Softening Failure". I tried this and I can't get it to behave any differently to the standard concrete (linear) material. Could it be that this also needs the CPT215/216 elements? Please keep the videos coming, they are very insightful and helping a lot of people.

@aneeshkumar432 2 жыл бұрын

Sir, is there any way to obtain SN Curve for concrete? i need to do fatigue analysis of RCC and SN curve of steel is available already but concrete's SN curve is not available.

@ankitagupta3296 4 ай бұрын

Can we do strain probe for unit cube instead of deformation probe, because both are given different graphs. Why did you choose deformation probe if strain probe is available ?

@mohamadmasri9583 11 ай бұрын

Dear Dr. Brock, i need the curves that u explained to obtain it in my scientific paper how can i get these regards

@aydnaydin9109 8 ай бұрын

We are waiting for more video about Ansys Prof. H..

@israelcrawford8734 2 жыл бұрын

Hello can you make a video doing post-tension flat slabs ?

@jiyadferoke2600 Жыл бұрын

Thank you sir for such a wonderfull and informative video. Sir iam doing microplane model of concrete composite so that i need to apply nonlinear properties for two components. How can i do that

@Pinky_Gaming123 Жыл бұрын

Very very great video. Before this my references was DrDalyO and Tanapon, now this is the best to refer. Thank you

@StructuresProfH Жыл бұрын

Glad it was helpful!

@ahmedalshekaili9720 2 жыл бұрын

Thank you very much for your fabulous presentation. I am just wondering whether Concrete MP Model can be a good model to simulate the rock-like material such as Sandstone? Is there any way to use APDL commands in explicit analysis instead to static structural analysis? I appreciate your great effort.

@StructuresProfH 2 жыл бұрын

You can use the microplane model for sandstone and other geo-materials, but I don't believe the microplane model is supported in the Explicit Dynamics solver. However, there are other material options in Explicit. When defining materials in the Engineering Data tab, there are options for "Brittle/Granular" materials. One of these is called RHT Concrete Strength, which can be used for concrete-like materials, and can probably work decently well for sandstone. There are also a few different options for Drucker-Prager Strength, which might be more straightforward. There is no need to define any APDL commands for any of these materials. There is an example Concrete material defined in the ANSYS "Engineering Data Sources" in the Engineering Data tab. It is hidden under Explicit Materials --> Conc-35MPA. This has parameters for RHT Concrete calibrated to a typical 35 MPa concrete mix. It also has some porosity expressions (P-alpha) and equations of state (Polynomial EOS defines the pressure), but not being an expert on sandstone I'm not sure if you need those or what you'd set these parameters equal to if you do need them. There is also the Explicit Materials --> Concrete-L predefined material, which uses Drucker-Prager strength. Obviously you'll want to have some calibration data handy to make sure whatever you do works OK.

@user-rm4uc8rb7p Жыл бұрын

It seems like the damage outputs do not work in workbench when there is more than 1 step，how to fix it??

@babaloluwapopoola654 Жыл бұрын

Thank you so much for the work you put into making these videos and explaining the fundamentals behind the concepts. I utilized the material model you provided for a uni-axial and flexural analysis using a simple beam geometry and while the uni-axial system gave the expected results, flexural loading conditions gave a tensile peak stress which was significantly higher than the tensile strength defined. Could this be due to tri-axial stresses, the inner workings of the micro-plane model or some other consideration?,

@StructuresProfH Жыл бұрын

It's tough to say. How much is significantly higher? In real-world tests, the modulus of rupture, which is the stress at which flexural cracking initiates, can be 2 to 3 times higher than the direct tensile capacity. That's just a rough estimate, but the different stress states for uniaxial tension versus flexure do make a difference. If this is still a problem, you can try lowering the Rt parameter, which controls the tension cap hardening of the Drucker-Prager yield surface. It can go down to zero, but that might cause numerical problems, so maybe try 0.1 or 0.5 and see what happens?

@babaloluwapopoola654 Жыл бұрын

@@StructuresProfH Thank you for the swift response. I initially had peak stresses which were almost 10 times the defined tensile strength, but after adjusting the Rt parameter I was able to arrive at values approximately 3-4 times the tensile strength. Thank you for your help

@user-pl2nn4fk4r Жыл бұрын

Hi, thank you for this video, it's very clear. I was wondering if this model is the one which is approximatively the same than Concrete Damage Plasticity in Abaqus ?

@StructuresProfH Жыл бұрын

Hi Justine, There are subtle differences of course, but yes, this is similar to the Concrete Damaged Plasticity model in Abaqus.

@mohammadhajsadeghi1724 Жыл бұрын

Very useful video! Thanks for the sharing. Just have a question. The straight line having the angle of 60 degrees with negative X, passes through the origin? I did draw the line but didn't intersect at fc!

@StructuresProfH Жыл бұрын

Thank you for your comment! First point of clarification - the X-axis is the equivalent volumetric stress and the Y-axis is the equivalent deviatoric stress. For a uniaxial applied stress f, the volumetric stress is equal to f/3 and the deviatoric stress is equal to f*sqrt(3)/3, which corresponds to a 60-degree line. Therefore, this 60-degree line should intersect the yield surface at x = -fc/3 and y = fc*sqrt(3)/3, where fc is the uniaxial compressive strength. Second point - the described intersection only holds true if the 60-degree line intersects with the linear part of the yield surface. If the 60-degree line intersects with the rounded tension cap at the right end of the yield surface, this intersection will happen at a smaller absolute value of x and y. However, using the proposed value of sigmaVT = -1/3*fc, the intersection will always occur on the linear part and never in the tension cap (note that sigmaVT defines the tension cap, and is assumed at this particular value for this implementation of the microplane model).

@mohammadhajsadeghi1724 Жыл бұрын

@@StructuresProfH Thank you so much for your reply.

@asaritas 2 жыл бұрын

Thank you very much for your helpful video. It was really very informative. If you have any experience of the robustness (accuracy, run time performance) of ANSYS microplane model v.s. ABAQUS plastic damage model for concrete, would you provide your feedback? For example, in monotonic/cyclic static or dynamic analysis of plain/reinforced concrete members. Thanks.

@StructuresProfH 2 жыл бұрын

You are welcome! I do have experience with both ANSYS microplane and ABAQUS damaged-plasticity, but I've never done a side-by-side comparison of the two. Both work well, but in my completely subjective opinion, I feel the nonlocal version of the microplane model in this video tends to have fewer convergence issues.

@asaritas 2 жыл бұрын

@@StructuresProfH Thank you for your feedback. It is great to follow your videos.

@asaritas 2 жыл бұрын

Do you have any knowledge on the use of microplane model with any of the shell elements in ANSYS instead of solid (215/216) elements?

@StructuresProfH 2 жыл бұрын

@@asaritas No, the coupled microplane model shown here does not work with shell elements. It does, however, work with some 2-D elements (for example, for plane stress or plane strain applications), which are given as (212/213) for linear/quadratic element order.

@asaritas 2 жыл бұрын

@@StructuresProfH Thank you for your reply.

@sahinsozen3946 2 жыл бұрын

Thank you very much for the impressive presentation. Do you think that this model can be used for masonry? If possible, how can the material model be calibrated? Best regards.

@StructuresProfH 2 жыл бұрын

That is an excellent question. I'll preface this comment by stating that I've never personally modeled masonry. However, I'd give it a tentative "yes, it will work" on the notion that any strain-softening geomaterial could be modeled using the microplane model. As for calibration, it depends on exactly how you are doing the modeling. If you are modeling individual blocks/bricks with mortar, then you'd need calibration data for both materials independently. If you are just treating the masonry structure as a single homogenous material, I assume you'd need to account for some sort of anisotropy - calibration would be done at a structural level in this case.

@sahinsozen3946 2 жыл бұрын

@@StructuresProfH I have been studying historical buildings and I use the macro modeling technique due to its simplicity. So I model the whole structure as a homogeneous material. I would like to use the microplane model you described, actually in masonry structures. Where do you think I should start exactly? Do I need stress-strain data of stone? Thank you in advance, Kind regards.

@pyaesone2845 Жыл бұрын

Dear Prof. H, Why is the nonlocal range parameter (c =1) not changed for the beam model ?? If the nonlocal behavior is not considered, will that affect the results? Thank you Sir..

@StructuresProfH Жыл бұрын

The main purpose of the nonlocal behavior to is help the model converge. Modifying the nonlocal parameter will alter the results, but it's not a "physical" parameter per se, insofar that it doesn't represent an actual physical property of the material. In my practice, I occasionally adjust this parameter if I have issues with convergence, but it's not something I change unless I'm forced to do so.

@aaaengineering8892 2 жыл бұрын

My name is Ayodele Akin-Adamu. Thank you so much for the guidance you provided on this topic. I am working on validation of the experimental results of small beams made with Ultrahigh performance concrete(UHPC) found in literatures for my Masters thesis project. Without your channel, I am sure I would never have stand a chance of completing my project. The procedure you highlighted for the micro plane model and menetrey-William model, do the models apply to UHPC material? and what extra factors do I need to consider when running the analysis on Ansys for UHPC?

@StructuresProfH 2 жыл бұрын

I'm glad this was helpful! You should be able to use either Microplane or Menetrey-Willam for modeling UHPC, though the "typical ranges" for the material parameters that I have listed may not apply. I would first start by generating stress-strain curves with a cube in compression (as I did in this video) or tension (I didn't do this, but similar idea) over a range of parameters. Get a feel for how parameters affect the response and what parameter values make sense. Select some values that more-or-less match the stress-strain behavior, then model a beam. Your first attempt probably won't fit the experimental data exactly - that's OK. You can then decide to iterate or accept the difference. Validation can be a tricky thing. If you just blindly adjust your parameters to exactly match some experimental data, that's merely calibrating a model. You haven't really proved that anything is "valid" because your selection of parameters might make no sense; it is possible to accidently fit results. That's why you need some sort of baseline understanding of the model, how it works, and what parameters represent. Once you have that understanding, then you can calibrate your model parameters on one set of experiments and then use that model (unchanged) to predict the response of some other set of experiments - that's a much more convincing validation. Good luck on your thesis!

@aaaengineering8892 2 жыл бұрын

@@StructuresProfH Thanks so much for the insights on how to use the model. I am so grateful. I will like to ask one more question. I have problem with force convergence when I tried to run the analysis on a modeled beam. What can I do to avoid this convergence program?

@aaaengineering8892 2 жыл бұрын

I have already overcame the force convergence challenge. I am very happy I now have a tool that can model the behavior of UHPC beams. I have 4 type of similar experimental beam data but with different percentage of tension reinforcement provided. As you have advised, I will do more studies on the micro plane model, i will calibrate 1 of the 4 type of beam that I have and validate the remaining 3 beams with the calibrated data.

@MathKM-pv6mm Жыл бұрын

Many thanks Prof. H , it was very helpful , but i have one question , in the excel sheet file for drawing Drucker-Prager graph there is Ratio , what's that ratio ?

@StructuresProfH Жыл бұрын

I assume you mean the "Ratio" column that varies from 0 to 1? This is a just a dummy variable so that I can plot the yield function from purely compressive volumetric stress (Ratio = 0) to purely tensile volumetric stress T0 (Ratio = 1). For example, Ratio = 0.5 gives the yield surface value at the volumetric stress halfway between the compressive and tensile limits.

@MathKM-pv6mm Жыл бұрын

@@StructuresProfH Thank you so much for your answer , it was very helpful.

@designsoft1765 Жыл бұрын

1) Can we get crack pattern of beam using this model? 2) I tried solid65 element and link180 and modelled the beam and used the command snippet for solid65 and link180. It analysed the beam for M20 grade concrete but when i changed the values for M25 grade concrete it is still showing me the m20 grade results. And also i want to change the diameter of the rebar for m25 grade beam but I'm unable to do that. Can you please help me out sir 😊

@StructuresProfH Жыл бұрын

1) Unfortunately, no, this type of model will not give you a crack pattern. Sometimes we call this type of analysis "smeared crack" modeling, as we attempt to model the overall cracked behavior of the structure but do not explicitly form discrete cracks. 2) I'm afraid I don't know the answer to this question. It sounds like you might need to clear your results before changing and running the analysis again? Tough to say though.

@sebastiankrug8355 2 жыл бұрын

Nice Video! Is there any literature about the microplane model and drucker-prager?

@StructuresProfH 2 жыл бұрын

This particular implementation in ANSYS is based on the following references: Zreid, I. & Kaliske, M. (2018). A gradient enhanced plasticity-damage microplane model for concrete. Computational Mechanics. 10(1007), s00466-018-1561-1. doi.org/10.1007/s00466-018-1561-1 Zreid, I. & Kaliske, M. (2016). An implicit gradient formulation for microplane Drucker-Prager plasticity. International Journal of Plasticity. 83, 252-272. doi.org/10.1016/j.ijplas.2016.04.013 Zreid, I. & Kaliske, M. (2014). Regularization of microplane damage models using an implicit gradient enhancement. International Journal of Solids and Structures. 51(19), 3480-3489. doi.org/10.1016/j.ijsolstr.2014.06.020 The first reference is the best for understanding how the Drucker-Prager condition fits into the picture. The original microplane model was developed by Z.P. Bazant and others - here are some key references for that: Bažant, Z. P., & Gambarova, P. G. (1984). Crack shear in concrete: Crack band microplane model. Journal of structural engineering, 110(9), 2015-2035. doi.org/10.1061/(ASCE)0733-9445(1984)110:9(2015) Bazant, Z. P. & Oh, B. H. (1985). Microplane model for progressive fracture of concrete and rock. Journal for Engineering Mechanics. 111, 559-582. doi.org/10.1061/(ASCE)0733-9399(1985)111:4(559) Carol I, Jirásek M, & Bažant Z (2001). A thermodynamically consistent approach to microplane theory. Part I. Free energy and consistent microplane stresses. International Journal of Solids and Structures. 38:2921-2931. doi.org/10.1016/S0020-7683(00)00212-2

@Pinky_Gaming123 Жыл бұрын

@@StructuresProfH thank you

@adrianulza8778 2 жыл бұрын

Thank you for this wonderful presentation. I wonder if I could use this microplane model for a kind of cyclic test simulation? Btw, could you please kindly give any suggestion for me to begin with, whether to use static structural or transient dynamic for RC cyclic simulation? Thank you in advance.

@StructuresProfH 2 жыл бұрын

Adrian, You should be able to use the microplane model for cyclic test simulations. I would employ static structural, not transient dynamics. Transient dynamics solves equations of motion (F=ma) that include displacement, velocity, and acceleration. Static structural solves equations of equilibrium (F=0) that use only displacement. Most laboratory tests are slow enough such that velocity and acceleration effects (like damping or inertia) don't matter, in which case static structural is far more efficient.

@adrianulza8778 2 жыл бұрын

@@StructuresProfH Hi, thank you for your intuitive explanation. Could I am asking one more further question? Supposed that I would like to run a cyclic simulation, how can I control the hysteresis behavior for both concrete and steel (pinching, strength degradation, etc)? There are some of of hysteresis model that I know, such as takeda, pivot, or kinematic in a conventional structural analysis. However, I am not sure how a heavy FEM platform such as ANSYS might handle this. Could you please enlighten me what kind of material properties should I modify, and could you please recommend me any relevant literatures? Thank you in advance, and I am greatly appreciate your help, sir.

@StructuresProfH 2 жыл бұрын

@@adrianulza8778 Hi Adrian, Now that's a tough question. I don't have a great answer, but perhaps I'll try to steer you on the right track. Things like the Takeda model are specifically for cross-section behavior, not material behavior, so that's not going to work for a detailed FEM. To get the best representation of pinching and other cyclic behavior, you will need some way of handling bond slip and the cyclic material behavior of the rebar. The Menegotto-Pinto model is probably the most commonly used for rebar, but I don't think it is included by default in ANSYS. I've seen some papers modify the rebar stiffness to account for bond slip (otherwise, you need to include additional DOFs connecting the concrete and steel, which can be a hassle); that seems like it would be easier to me. This publication... www.researchgate.net/publication/334029651_CYCLIC_NONLINEAR_MODELING_OF_SEVERELY_DAMAGED_AND_RETROFITTED_REINFORCED_CONCRETE_STRUCTURES ... by Markou et al. (2019) might point you in a productive direction. So might this one... www.researchgate.net/publication/278710287_Seismic_Analysis_of_Steel-Concrete_Composite_Buildings_Numerical_Modeling Good luck on this! I may do some research on my own in the meantime, as this would be a great topic for a new video.

@adrianulza8778 2 жыл бұрын

@@StructuresProfH Thank you very much! I am really appreciate your help. Keep doing great with this cool channel! 🤟

@hkc6615 2 жыл бұрын

Hey Sir, I'm graduating and my project consist on modelling an interface of steel base plate with a concrete, and I would like to discuss with you using a video call please answer me, I really need you, we won't take a lot of time

@pradeepsinghkanyal8922 2 жыл бұрын

Hello sir.. For CFST column..i have to calculate ultimate load carrying capacity along with stress-strain behaviour of CFST column..how can i do this in ansys mechanical..i do not work in apdl..plz help

@StructuresProfH 2 жыл бұрын

For a model that doesn't require APDL, check out my previous video on using a Menetrey-Willam model (kzfaq.info/get/bejne/qNd4iqRqrrDSYJs.html). That said, the APDL code posted in this video description is meant to be copied-pasted and then easily modified to your needs (as explained in the video), so you don't need to be an expert. Give it a try! For concrete-filled steel tubes, the triaxial and dilatancy behavior of the concrete are rather important, so it may take a bit of calibration to make sure that your results are realistic.

@mohammadhajsadeghi1724 Жыл бұрын

One more question: Is it possible to use this method directly in ANSYS APDL environment? Writing an apdl makes parametric studies a lot easier...

@StructuresProfH Жыл бұрын

Absolutely!

@mohammadhajsadeghi1724 Жыл бұрын

@@StructuresProfH I tried but I face warning saying it is not supported. I assume that I am making mistakes. I really appreciate if you can make a video on how to use it in ANSYS Mechanical APDL.

@StructuresProfH Жыл бұрын

I’m quite certain it should work unless perhaps you are using some different version of ANSYS. In the video, the commands to define the microplane model are APDL, and in general APDL can do everything the GUI can do (it may just be more cumbersome). So it’s tough to say what the issue might be. I’m afraid I won’t have the chance to make such a video for some time, but if you solve your issue I’d be interested to see how it turns out.

@mohammadhajsadeghi1724 Жыл бұрын

@@StructuresProfH I am using ANSYS 18. Perhapd I should try latest version. I put this on my do-list and hope I get some free time soon to try. For sure I will share it with you. Thanks for your reply and helpful channel!