How to introduce a lot of functions into Time history analysis without having to click one by one?

[eperez3W9AN]

Hello,

I am trying to model a pedestrian vibration analysis so I have create a lot of load cases (128) simulating the pedestrian moving on:

eperez3W9AN_0-1743091099729.png

eperez3W9AN_1-1743091122343.png

eperez3W9AN_2-1743091147022.png

 Then, in Time history analysis window i have to introduce the time-function of every load case:

eperez3W9AN_3-1743091258384.png

But  is there any way to enter them other than clicking one by one? With tables, excel,... I have to do differents cases :(.

Thank you sou much.

[mdesjardinsXJ286]

Hi,

I don't have the answer for your particular request. But about the analysis of the vibration on a pedestrian bridge I found that link below on youtube. I am actually doing the evaluation of a suspended pedestrian bridge and I am facing the same problem as you are. The amount of work is to do a complete evaluation is huge amd require a lot of manual work in the software.

On the video they show how to do the vibration verification with robot, according to french regulation(wich si the same as canadian one(my country)). The process they use is first doing a modal analysis and then for every mode that is within the critical range of pedestrian induced vibration(stated in the standard) they create a FRF analysis in wich they applied a uniform load(defined by the standard in N/m²) that is applied in direction of the deformation of the particular mode(screen shot below). The output is the acceleration that need to be below the critera fixed by the standard. With about 50 different configuration to check it makes it a huge work. As you I am looking for an simple solution and fast solution.

mdesjardinsXJ286_0-1743688207340.png

[mdesjardinsXJ286]

Hi,

I am struggling with a problem similar to you doing the vibrations verification on a pedestrian bridge. According to the CSA standard I should do a FRF analysis in every mode while applying the load in the opposition direction  of the mode deformation.

mdesjardinsXJ286_0-1743690997519.png

It's make it huge amount fo work to model every case since mode shapes varies a lot and the load is deffirent for every mode. I am struggling to find an simpler way to evaluate the acceleration.

I attached the french guide below wich is the a very similar procedure to the canadian one.

[mdesjardinsXJ286]

Also found that simplified procedure to approximate accéleration but I am struggling finding the value that I should apply to Pn because it is the only value not given by ROBOT.

mdesjardinsXJ286_8-1743693015620.png

mdesjardinsXJ286_3-1743691357734.png

mdesjardinsXJ286_5-1743691378301.png

mdesjardinsXJ286_6-1743691392916.png

I tryed using the harmonic load multiplied by the total bridge deck surface but It was not releveant.

mdesjardinsXJ286_7-1743692836571.png

If anyone have an idea let my know

[Romanich]

Hi @mdesjardinsXJ286,

I think Pn (generalized load for nth mode) is shown on the picture you attached to the previous post.

[Romanich]

You can also find more useful information in this book:

European Commission
EUR 23984 EN – Joint Research Centre – Institute for the Protection and Security of the Citizen
Title: Design of Lightweight Footbridges for Human Induced Vibrations

SBN 978-92-79-13387-9
DOI 10.2788/33846

[mdesjardinsXJ286]

Hi,

It is my tought to but I get some very high acceleration values witch make me think I am making something wrong. Here is my procedure.

-Modal masse is found adding  mX, mY and mZ from ROBOT modal analysis
-Damping is also found from robot analysis

mdesjardinsXJ286_0-1743696534626.png

-Pn I multiplied the unit force obtained from the standard by the total area of the bridge (Pn=pm(t)*S)

I got those result that are very very high and seems unreallistic

mdesjardinsXJ286_2-1743697848170.png

[mdesjardinsXJ286]

Thanks a lot! I will give a look!

[mdesjardinsXJ286]

Do you think I should try one of the following approach:

1) Adding the forces relative to the direction Pn=(pm(t)*(S2-S1-S3)

mdesjardinsXJ286_3-1743698384195.png

2) Considering only one vibrating lenght Pn=pm(t)*S2

Thanks for your help

[Romanich]

I think the first approach will be correct.

[mdesjardinsXJ286]

Hi,

I found this on the document you sent me.

EUR 23984 EN – Joint Research Centre – Institute for the Protection and Security of the Citizen
Title: Design of Lightweight Footbridges for Human Induced Vibrations

mdesjardinsXJ286_0-1743709217116.png

So I used the following n being the number of half waves.

mdesjardinsXJ286_1-1743709274489.png

For more accurancy I think I will need to do the FRF on the specific case I want to investigate