Nonlinear problem - Unable to apply

[rgduarte]

Hi.

I'm trying to make a Nonlinear time history analysis of the Northridge NS earthquake.
I've used force based and displacement based elements but with no success. At the highest peak of acceleration this message appears, despite the element type i use:

Time= 3,62658, dt= (Iter: 50 => Prd_Ite)
Time= 3,62670, dt= 0,00012 (Iter: 1 => Converg)
Time= 3,62670, dt= (Iter: 50 => Prd_Ite)
Time= 3,62681, dt= 0,00012 (Iter: 1 => Converg)
Time= 3,62681, dt= (Iter: 50 => Prd_Ite)
Time= 3,62693, dt= 0,00012 (Iter: 1 => Converg)
Time= 3,62693, dt= (Iter: 50 => Prd_Ite)
Time= 3,62705, dt= 0,00012 (Iter: 1 => Converg)
Time= 3,62705, dt= (Iter: 50 => Prd_Ite)
Time= 3,62717, dt= 0,00012 (Iter: 1 => Converg)
Time= 3,62717, dt= (Iter: 50 => Prd_Ite)
Time= 3,62729, dt= 0,00012 (Iter: 1 => Converg)
Time= 3,62729, dt= (Iter: 50 => Prd_Ite)
-----------------------------------------
Unable to apply the next load step. Analysis terminated
Total Analysis Time: 0h:1min:45sec


I've read the topics on "unable to apply" and made the suggested changes (element type, convergence criteria, etc) But with no result.
It's a 2D Structure (XZ), 4 bays with 9,15m, 3 story with 3.96m and Rayleigh damping (2%). The 3rd bay is the weaker and the beams aren't moment resistant.
NOTE: I've made a analysis using element frame, and the program will run , but i want a nonlinear behavior.

What can i change more ?
The .spf file: https://www.dropbox.com/s/ztyb1dukmnd63 ... iz2%29.spf


Thank you

[seismosoft]

Have a look at the following video on SeismoSoft YouTube channel:
https://www.youtube.com/watch?v=p6tAyoofDog

Also note that the reason for divergence might simply be a structural failure.

SeismoSoft Support

[fakharifar.mostafa]

Dear rgduarte,

I went through your model. It seems that the structural failure (numerical instability, zero pivot) is the reason for analysis to stop. Your frame constitutes fairly slender columns with moment-released connections at the third bay.

I am not sure what you meant by "NOTE: I've made a analysis using element frame, and the program will run , but i want a nonlinear behavior." However, your frame at peak acceleration reaches the peak strength. If you remove the moment releases from the members, forces could be re-distributed and could continue analysis. However, at 4 seconds, almost all of your structural members were yielded and the top node displacement time history is permanently shifted (residual displacement typical of steel fracture behavior).

One note, as the current model Displacement Based elements are used, element discretization is required.

Hope this helps.

[rgduarte]

Thank you for your answer.
I'm trying to model the 3 story structure, developed in the paper: "Benchmark Control Problems for Seismically Excited Nonlinear Buildings".

With the following description of the 3 story structure:

The three-story (3-story) benchmark structure is 36.58 m (120 ft) by 54.87 m (180 ft) in plan, and 11.89 m (39 ft) in elevation. The bays are 9.15 m (30 ft) on center, in both directions, with four bays in the north-south (N-S) direction and six bays in the east-west (E-W) direction.
The building’s lateral load-resisting system is comprised of steel perimeter moment-resisting frames (MRFs) with simple framing between the two furthest south E-W frames. The interior bays of the structure contain simple framing with composite floors.
The columns are 345 MPa (50 ksi) steel. The columns of the MRF are wide-flange. The levels of the 3-story building are numbered with respect to the ground level (see Figure 1). The 3rd level is the roof. Typical floor-to-floor heights (for analysis purposes measured from center-of beam to center-of-beam) are 3.96 m (13 ft). The column bases are modeled as fixed (at the ground level) to the ground. The floors are composite construction (i.e., concrete and steel). The floor system is comprised of 248 MPa (36 ksi) steel wide-flange beams acting compositely with the floor slab. In accordance with common practice, the floor system, which provides diaphragm action, is assumed to be rigid in the horizontal plane. The inertial effects of each level are assumed to be carried evenly by the floor diaphragm to each perimeter MRF, hence each frame resists one half of the seismic mass associated with the entire structure.
The seismic mass of the structure is due to various components of the structure, including the steel framing, floor slabs, ceiling/flooring, mechanical/electrical, partitions, roofing and a penthouse located on the roof. The seismic mass of the first and second levels is 9.57´105 kg (65.5 kips-sec2/ft) and the third level is 1.04´106 kg (71.0 kips-sec2/ft). The seismic mass of the entire structure is 2.95´106 kg (202 kips-sec2/ft).

Therefor, i can't remove the frames with moment release neither the sections. Yet , the paper get consistent results with this NLTH using the El centro, Northridge, Kobe and Hachinohe accelerograms.

I've changed the displacement based to force based and I've made the adjustments suggested in the video (Thanks in advance,seismosoft). I've replaced the additional mass by mass elements, and the program runs , besides taking almost 3 hours. Yet the results are impossible, e.g. acceleration of the roof nodes take values around 4Km/sec^2. (using the same accelerograms.)
NOTE: I've run the wizard default model and it runs with no problems and the results are consistent.

Link containing the force based and mass elements model and the paper:
https://www.dropbox.com/sh/ssq8hq20prhu ... KpJVKP7q-a

Thank you. Hope i get this problem solved.

[huffte]

Hi rgduarte.

I don't believe either of the below issues would prevent the model from running, but they should certainly be addressed it seems to me.

1. There is a large ground drift in the applied accelerogram. Integrating the acceleration time history twice yields an end of record ground displacement of over 20 meters. I would recommend that you baseline adjust your input accelerogram using SeismoSognal or SeismoSpect before using it in a structural analysis.

2. You solution time step is more than twice the time step in the input accelerogram. I would recommend setting the number of steps equal to 50.02/0.02 = 2501. Certainly no fewer steps than this value, perhaps more than this value. In SeismoSignal for the generation of inelastic spectra, the time step of the input accelerogram is used initially, and if needed the time step is reduced to T/50 for the period being analyzed. Solving a MDOF in SeismoStruct is far different than solving a series of SDOF's in SeismoSignal so a time step of T/50 could be sluggish for MDOF, but could feasibly be necessary in rare circumstances.

Best of luck and kind regards.

[seismosoft]

Hi rgduarte,

I addition to the suggestions by other users, note that using inelastic frame elements with end-releases leads to serious convergence problems (hence it takes 3h ti run the analysis). You can either change the corresponding elements with releases to elfrm, or remove the releases (we tried both, the analysis runs in less than 2min).

SeismoSoft Support

[rgduarte]

Thank you very much, i was able to solve the problem using the elfrm on the moment released frames.

But when i try to insert bracing elements with moment releases, i get the Max_Tool message and that two nodes of the bracing element have zero diagonal elements, I understand that the element isn't stable when i insert them, but the model has moment released connections . Is there any way to model it on seismostruct?

For more details:
https://www.dropbox.com/s/c68lhopmc9haz ... RIDADE.spf

[huffte]

I wonder if bracing elements with no releases but with artificially low I22 and I33 values would be more numerically stable? Just a thought.

[rgduarte]

I've tried it before the comment, yet all i'v got was a Prd_ite on the first iterations instead of the Maxtool, I've tried to change the convergence and elements criteria on the option menu but with no success. I've made a eigenvalue analysis with the moment released elements and the results are good. But when i do a NLTH analysis the program doesn't run, it will only run whit the releases removes (which is not what i intend to model)
Any more suggestions ?

Thank you very much

[huffte]

I looked over the model NT-SMA-TENSEGRIDADE. It seems to me that you have 4 joints at which m2 is released at the only 2 elfrm members framing into the joint with a truss member the only other member framing into the joint. Is the parallelogram bracing scheme intended to model SMA damping systems? It seems to me that your modeling scheme may need to be adjusted.