first of all I want to thank you all and especially lululxvi for your amazing work.
And here is what I want to do with deepxde:
I want to use deepxde for an inverse advection-diffusion problem where I want to find the advective velocity as a function of time based on measured data at different locations in the model domain.
Here is an example of the dataset: enter image description here
The plot is showing temperature measurements in time for different depths. I want to use the lowest and highest depth as boundary conditions and the inner depths as observables for the inverse problem. I am not sure how i can do this. For the two boundary conditions do I need to use dde.icbc.DirichletBC? In most of the examples this is used to implement the boundaries but ususually there is a function passed where the BC are evaluated. However in my case I have an array with BC values for different timesteps which i cannot pass to dde.icbc.DirichletBC.
In another example dde.icbc.PointSetBC(X,observe_T) is used to pass both, the boundary conditions + internal observations for the inverse problem. Here i guess X is a vector with two columns where the first column contains the x-location and the second column the timestamp, and observe_T the corresponding observed temperatures.
Can anyone confirm this is correct and/or has an example script for me where something similar was done?
Thank you
I tried to use the dde.icbc.DirichletBC