I am currently working with a model-scale pile-soil-bridge model and trying to validate it using shake table experimental data for the same configuration. My dataset comprises accelerometer and displacement sensor readings.
Upon analyzing the acceleration response from my numerical model, I've observed a favorable match with the measured data obtained during the experiment. However, I've encountered an issue with the calculated displacements. These calculated values are consistently several orders of magnitude smaller than the displacements measured during the experimental test, as recorded by laser meters.
In my attempt to address this discrepancy, I've double integrated my acceleration data (while acknowledging the associated error). This process yields significantly larger values than the displacement readings noted in the file.
In my analyses, I have fixed the nodes at the base of the model and applied uniform excitation. My understanding is that in uniform excitation, the prescribed acceleration time history is applied to all nodes, generating "relative" results. As such, I've summed the calculated acceleration values with the seismic motion input.
However, I feel confused when it comes to handling displacement. Given that I'm inputting acceleration and have fixed the base of the model, I would anticipate that these displacement values should be "absolute" due to the nodes at the base are fixed.
When I compare:
- The summation of recorded accelerations with the input motion closely aligns with the input motion itself.
- However, upon inspecting the displacement data, I find that the values are on the order of e-20, essentially negligible. While my initial thought was that this might be due to fixing the base nodes in the loading direction, even when I release them, the resulting displacements at these nodes remain tiny.
I'm reaching out for assistance, as I've been stuck with this issue for a while and am unable to make progress. If you have any insights or suggestions, I would greatly appreciate your input.
For your reference, I have included the Eigenvalue modal analysis report below (from modal analysis procedure that STKO developed). Additionally, I've attached a clean copy of my model (I've made it specific to this topic, though I might have overlooked some parameters to delete—please disregard those). A sketch of my model has also been attached to provide a clearer understanding, in case you find the opportunity to review it.
I'd really appreciate any thought or helps that I could get to solve this issue.
Best Regards,
Burak O.
Sketch:
- Screenshot 2023-08-15 at 10.02.17 PM.png (222.11 KiB) Viewed 1761 times
- Screenshot 2023-08-30 at 12.34.22 AM.png (146.33 KiB) Viewed 1761 times