About absolute acceleration

[khunglongkhelua]

I am doing some simple model used the Opensees software and I think I need your help about absolute acceleration.

This is the code to achieve the absolute acceleration

recorder Node -file output.out -timeSeries $IDloadTag -time -node 1 -dof 1 accel

I followed the instruction to come up with my codes like this :

set outfile2 NWH090_AT2.g3
source ReadSMDFile.tcl
# source Smd file to obtain dt
ReadSMDFile NWH090_AT2.txt $outfile2 dt2
timeSeries Path 4 -dt $dt2 -filePath $outfile2 -factor $g
pattern UniformExcitation 2 1 -accel 4; # create Uniform excitation
recorder Node -file 1stModelbeta11/Accel1st11.out -timeSeries 4 -time -node 2 4 5 6 -dof 1 accel

However when I run my files, I have the warning like this

TimeSeries - none found with tag 4

I don't know where I did wrong, can anyone help me?

When I omit the -timeSeries and replace by the -time everything is fine so I think the problem is from the tag timeSeries 4

Thanks a lot.

[vesna]

I looked at the part of the code you posted and it looks right, but since you are getting the error message something in the definition of the time series must be wrong.

Look at the file NWH090_AT2.g3 to make sure it contains only numbers. Check what are the values of dt2 and g.

[khunglongkhelua]

Dear Vesna
The error message is this :

TimeSeries *getTimeSeries<int Tag> - none found with tag: 4

Do you have other idea because I checked what you suggested and everything seems right.

[vesna]

Make sure that somewhere else in the code before you have defined your time series 4 you do not have a recorder for absolute acceleration.

[khunglongkhelua]

Dear Vesna

I can run my program now. Thanks a lot for your help. However I found that the result may seems wrong. I will illustrate by the example below. I chose $dt of my earthquake record and my analyze step equal 0.02 s

Times------------ Relative Acceleration------Absolute Acceleration---------Earthquake record should be
0.02----------------------1--------------------------------3---------------------------------2
0.04----------------------2--------------------------------5---------------------------------3
0.06----------------------3--------------------------------6---------------------------------4

However, this is my earthquake record
Times-------Earthquake
0 --------------0 ( does not show in the record file but I understand that) )
0.02----------- 100
0.04----------- 2
0.06----------- 3
0.08----------- 4

I ran the gravity analysis first and then I set -time 0.0 and then run the uniform excitation analysis.

[vesna]

I do not understand your new question. Can you try to explain it again?

[khunglongkhelua]

I mean to get the absolute acceleration:

AAccel(t) = ReAccel(t) + GAccel (t)

However, the result I got is this:

Aaccel(t) = ReAccel(t) + GAccel (t+dt)
I just edited my previous post, hope you can understand my question now.

[vesna]

that is because the 1st value in the ground motion file is the one that corresponds to time 0.0 and the 2nd one corresponds to 0.02. If you would compare the accelerations at the same time you would see that AAccel(t) = ReAccel(t) + GAccel (t)

[khunglongkhelua]

I am sorry I don't understand what you mean.

According to the Chopra book, I think the first value of the ground motion is at t=0.02, acceleration at t = 0 is zero ( does not show in the ground motion file).

[vesna]

The way OpenSees manipulates the date is like what I explained in my previous message. At time 0.0 the relative acceleration is zero and absolute acceleration is equal to ground motion but it is not recorded since recorder is not created until the first step of analysis was completed.

[khunglongkhelua]

Thanks vesna.

[khunglongkhelua]

I have one more question vesna. I encountered the problem like this when I try to do the eigen analysis

ArpackSolver::Error with _saupd info = -9999 Could not build an Arnoldi factorization.IPARAM(5) the size of the current Arnoldi factorization: is 9factorization. The user is advised to check thatenough workspace and array storage has been allocated.

Can you tell me what this warning means and how to fix this.

[vesna]

If your system is really big you may need to use one of the sparse solvers (e.g., UmfPack) for eigen analysis.

[successor]

khunglongkhelua wrote:
> Dear Vesna
>
> I can run my program now. Thanks a lot for your help. However I found that
> the result may seems wrong. I will illustrate by the example below. I chose
> $dt of my earthquake record and my analyze step equal 0.02 s
>
> Times------------ Relative Acceleration------Absolute
> Acceleration---------Earthquake record should be
>
> 0.02----------------------1--------------------------------3---------------------------------2
>
> 0.04----------------------2--------------------------------5---------------------------------3
>
> 0.06----------------------3--------------------------------6---------------------------------4
>
> However, this is my earthquake record
> Times-------Earthquake
> 0 --------------0 ( does not show in the record file but I understand
> that) )
> 0.02----------- 100
> 0.04----------- 2
> 0.06----------- 3
> 0.08----------- 4
>
> I ran the gravity analysis first and then I set -time 0.0 and then run the
> uniform excitation analysis.

Hi,

How did you solve the problem of TimeSeries *getTimeSeries(int tag) - none found with tag :4? I'm confronted with this problem right now.

Thank you very much,
Yenny

[skamalzare]

Dear Yenny,

I am working with this command all the time, and so far I have not encountered any problem with it. Here is the part that I use in my code, you can check it out with your code and see if you find any problems:

set accInputFile InputMotions/EQ-YBI.txt
set motionDT 0.01
set accMul 9.807
timeSeries Path 1 -filePath $accInputFile -dt $motionDT -factor $accMul
...
recorder Node -file Output/acceleration.out -time -timeSeries 1 0 -nodeRange 1 42 -dof 1 2 accel

This adds above defined timeSeries to first DOF of my node range.

Bests,
Soheil