Refine
Posted: 22 May 2018, 15:45
Hi,
I use Seismosignal to integrate acceleration time history to displacement which I then feed to my finite element model.
The FE models use a Newmark algorithm to solve the dynamic system so I need a baseline correction before I use it in the FE to avoid spurious acceleration from the Newmark algorithm.
My timesteps are defined at each 0.002 second. However, in one of my model, I fails to converge.
I would need to recude the time steps in the FE model. Problem is, when I do this, the newly assigned displacement (which is estimated without baseline correction) start to create spurious acceleration.
Is there a way that Seismosignal can provides me with more points when I make the baseline correction?
Say I have 4000 entry at 0.002 second which make a 8 second records. I would need, for example, 8000 entry at 0.001sec from the baseline correction?
The only option I see right now is to somewhat myself create the new value of acceleration from linear interpolation between existing value, then import in seismosignal then do a baseline correction. This process is time consuming... Any other way around?
Thanks for the help!
I use Seismosignal to integrate acceleration time history to displacement which I then feed to my finite element model.
The FE models use a Newmark algorithm to solve the dynamic system so I need a baseline correction before I use it in the FE to avoid spurious acceleration from the Newmark algorithm.
My timesteps are defined at each 0.002 second. However, in one of my model, I fails to converge.
I would need to recude the time steps in the FE model. Problem is, when I do this, the newly assigned displacement (which is estimated without baseline correction) start to create spurious acceleration.
Is there a way that Seismosignal can provides me with more points when I make the baseline correction?
Say I have 4000 entry at 0.002 second which make a 8 second records. I would need, for example, 8000 entry at 0.001sec from the baseline correction?
The only option I see right now is to somewhat myself create the new value of acceleration from linear interpolation between existing value, then import in seismosignal then do a baseline correction. This process is time consuming... Any other way around?
Thanks for the help!