the parameter
Brought to you by:
ampuero
Menu
▾
▴
dear ampuero,can you help me check the parameters? another problem is that the stability of the simulation is not bad in the case that I have ensured the Gaussian point> 5 in the minimal wave length. why is it ?
Dear user,
Your mesh has a problem. In my reply to your previous support request #26 (https://sourceforge.net/p/sem2d/support-requests/26/) I already explained the problem and suggested a meshing strategy that should work.
Best regards,
OH ， I am sorry , I Uploaded a error picture ,the right is as following.
Please describe the symptoms of your problem with more detail.
I study the seismic response of sedimentary basins beside the fault which is rupturing. But the response in the same point is quicker than no sedimentary basins. I want you help me to check my paraters in the par if someone is fatal error .
second I find that different meshing size( value of h change ) for the same model can cause different results especially for the acceleration below the Gaussian point of the minimal wave length > 5 .it is to say the stability of the simulation is bad ,why do this occur ?
Thank you for your attention and hope to receive your reply.
Last edit: hanjianbin 2015-01-05
For your simulation without basin, what mesh and what input file did you use? What is the position of the receiver shown in your figure?
Did you set the Kelvin Voigt damping "eta" according to the instructions in section 2.2.2 of the manual? Did you set the "Courant" parameter according to the note in the "MAT_KV" input block in the manual?
1.I use 'MESH2d' method,the the input file is as the attachments .the position of the receiver is (3000 0),which is same in the containing basin model- the center (3000 0).
2.I set it MAT_KV eta=0.2d0 ,set the courant = 0.4d0 ,but I did not find their relation ,setting the courant<0.5 casually .I study the relation carefully. The "ex0" is not satisfied to the equation 'eta/dt=0.1~0.3' ?
Last edit: hanjianbin 2015-01-07
How are the parameters 'eta''counter' ascertained generally? According to experience or calculation ? my head gets a little confused ? they are not calculated seemingly in the examples that you appled in the sempack2.3.8.
Last edit: hanjianbin 2015-01-07
In the same strike-slip fault model ,different number of the GLL point can make big difference to the acceleration recodded in time travel curve at the same place of the face of earth under satisfing the GLL precision. why do they make the difference ?
Last edit: hanjianbin 2015-01-14
I ran your two simulations. It did not reproduce your problem. As shown in the figure attached, I found that without basin the wave arrives earlier than with basin. This is the expected behavior, and it is the opposite of the figure you posted on 2015-01-05. My figure shows that the code is working as expected.
Please verify that you were plotting seismograms from the same receiver (#37).
yes ,thank you I am checking my error .I am very grateful