Question about the weird variaiton of axial force at pier base during a nonlinear time history analysis

[EasonLu]

Hi, guys. Hope everyone is doing well.

When I conduct a nonlinear time history analysis on a single pier model (20m), I notice the axial force at the bottom section is not constant during the whole process. Even worse, at some increments/times, the axial force increases/decreases dramatically.

Since the motion record is only input in the horizontal direction, I am confused about the result. Can anyone throw light on this? Many thanks!

Eason

[EasonLu]

In order to increase the accuracy of evaluating seismic response of structures, it is critical to conduct dynamic analyses based upon precise nonlinear models being as consistent as possible with the real conditions of corresponding structures. The concentrated plasticity model including one elastic element and two nonlinear spring elements at both ends has been considered within the research community for simulating beams and columns, counting the effect of strength and stiffness degradation. In this type of simulation, the axial force ratio generated in each structural component, which is a major factor in introducing nonlinear springs, has always been considered constant in the literature. The main objective of the present research is, therefore, to modify the fundamental weakness in this type of modeling approach; indeed, any variation of element’s axial effort, owing to redistribution of axial forces during an earthquake, is applied in the calculation of parameters of the concentrated plasticity model as a decisive step toward the development of nonlinear dynamic analysis. Moreover, an algorithm is presented for implementing this approach in the OpenSees software. Verification is established and the efficiency of the proposed method is illustrated through a reinforced concrete moment frame subjected to a specific record, as a case study building. Regarding the results, it is confirmed that the proposed algorithm is an appropriate tool for achieving quite a realistic nonlinear model and estimating reasonably accurate responses of structural systems with cyclic degrading behavior under earthquake loading.

Thanks for your reply!
1. I used the "nonlinearBeamColumn" element, and I defined the nonlinear behavior of concrete and rebars;
2. As the model is built to simulate a single pier model, and, the earthquake record is only applied in the horizontal direction. I do not understand why the axial force varies during NTHA.
Bests,
Eason

[mhscott]

Did you restrain the vertical DOF at the top of the pier? It could be the neutral axis of the section moving away from the geometric center, leading to apparent axial forces.

[EasonLu]

Did you restrain the vertical DOF at the top of the pier? It could be the neutral axis of the section moving away from the geometric center, leading to apparent axial forces.

Dear Dr. Scott,

Thank you so much for your reply. By the way, I am a fan of you and following your blog at "https://portwooddigital.com". The posts there are very informative and helped me a lot . Thank you!

Sorry but I am confused about "restrain the vertical DOF at the top of the pier" ... I did not do that, and the pier top is free. Before running the NTHA, I applied the axial force at the top. Do you mean that I should restrain the vertical DOF after the axial loading? Looking forward to your reply, and thank you!

I think I may need to provide more information about the pier:
20m;
3D model;
bottom fixed and top free;
50 elements (to be consistent with other piers);
nonlinearBeamColumn;
section Fiber $SecTag -GJ $TorStif { ... (where concrete and steel fibers are assigned)};
mass lumped at each node and a big mass at the top node;
geomTransf PDelta;
constraints Transformation;
system UmfPack;
numberer Plain;
set TestType EnergyIncr;
set algorithmType NewtonLineSearch;
integrator Newmark 0.5 0.25;

Bests,
Eason

[Thethinker]

You have to adjust vertical DOF at the top of the pier, that will help you.FMWhatsapp APK

[EasonLu]

You have to adjust vertical DOF at the top of the pier, that will help you.FMWhatsapp APK

Okay, will try and thank you! As I am learning OpenSees, could you please advise me on the command that can restrain the vertical DOF? Thanks again!

Bests,
Eason

[fmk]

to original poster .. what convergence test are you using .. i suggest NormUnbalance if you are not using it .. NormDispIncr is pretty much useless in dynamic analysis if you want to ensure equilibrium

[rahsagroup]

You have to adjust vertical DOF at the top of the pier, that will help you.FMWhatsapp APK