Eigen Value Analysis

[maryammf]

Hi everyone,

I am doing eigen value analysis for the soil medium with quadUP elements and PDMY2 material model. The rate for period in simulation is higher than what I am expecting by the formulation of 4H/vs (about 10 times). I checked the steps but do not know where the problem is. I will appreciate your suggestions in this respect.

Bests,

[Jhno]

Are you doing your analysis in the elastic or inelastic phase of the PDMY material?
Also make sure you have correct effective stress and correct small strain G and K.
If you still get discrepancy, perhaps you should try with a 1D column using elastic isotropic material first to get the 4H/Vs value.

[maryammf]

Hi,

I am using elastic analysis. You mean effective stress values in the output?
I am using these values for PDMY2 material"

# Soil -material parameters definition
set massDen 1.99; #(ton/m3)
set refG 75.71e3; # (kPa)
set refB 202.14e3;# (kPa)
set frinctionAng 31.33;
set peakShearStrain 0.1;
set refPress 100; # (kPa)
set pressDependCoe 0.5;
set phaseTransAng 26.33;

set contractionParam1 0.009;
set contractionParam2 2.0;
set contractionParam3 0.2;

set dilationParam1 0.15;
set dilationParam2 3.0;
set dilationParam3 0.0;

set liqParam1 1.0;
set liqParam2 0.0;

set noYieldSurf 20;
set void 0.658;
set n [expr $void/(1+$void)]
set cs1 0.9;
set cs2 0.02;
set cs3 0.0;
set pa 100; # (kPa)
set c 0.1; # (kPa)

# Some variables for the ELEMENT
set fluidDen 1.0; # Fluid mass density
set waterbulk 2.2e6; # kPa
set kdrain 1.e1; # permeability for drained loading
set kundrain 1.e-8; # permeability for undrained loading
set gravY -9.81;
set gravX 0.0;

puts "Finished defining some element properties..."

I really appreciate your suggestions in this respect!
Many thanks

[Jhno]

May I ask what type of soil you are modelling? It's very hard to say anything without carefully looking the code.
In any cases, did you check the OS example of a soil column analysis? http://opensees.berkeley.edu/wiki/index ... _Analysis) and http://opensees.berkeley.edu/wiki/index ... oil_Column
Hope it helps, GL

[maryammf]

It is a sandy soil, using PDMY2 model and quadUP elements.
I am modeling a soil medium of 70m*30m and wanted to do the eigen value analysis for this. Shall I just model it as a column to see the natural frequencies?
Many thanks,

[Jhno]

Shall I just model it as a column to see the natural frequencies? -- I would tend to say it depend on why you want to model this soil deposit.

You say you want to " do the eigen value analysis ". If this is your only objectif, use the relation 4H/(Vs*(2n-1)) to obtain the n-th periods value.
If you want to conduct a wave passage analysis then you can sure do it with a soil column. Just use consisten lateral boundary (equalDOF). You should then cross-check your result using proShake.

[maryammf]

My total model is a pile inside a soil medium. First I decided to make the eigen value analysis to take the natural frequencies for just the soil medium wo the pile and check the natural frequencies from OpenSees with that equation to see if I am doing right in modeling. That is why I wanna do eigen value analysis. But the results I got are higher than the one from the equation and I do not know where I am doing wrong! I fixed the base and the lateral boundaries are defined by equalDOF in x and y directions.
It will be great if you can help me with this problem, many thanks

[Jhno]

Check your model against proposed example and you will eventually find the problem. I would suggest you use simpler element for a start (don't start with a complex fluid-coupled model as it introduce more difficulty.. you can add it later when you have a solid model).
GL

[maryammf]

Ok, many thanks for your explanations.
Bests,