search

org-arl / arlpy

ARL Python Tools
BSD 3-Clause "New" or "Revised" License
119 stars 37 forks source link

Eigen rays are not correct #11

Closed ZHANGGU closed 5 years ago

ZHANGGU commented 5 years ago

Hei

Great python package!

I tried scenarios as shown by the figures, the eigen rays are not correct somehow.

  1. It is wrong in env setup, as the Rx is below the bottom, but it gives all rays.

rays_37km

  1. It should not give all rays. rays_46km

By the way, using Bokeh plot is resource consuming, and it takes much time to give a figure of low resolution in saving it. While it is much faster in zoom-in/out

Thanks

ZHANGGU commented 5 years ago

2018-10-26_08-14-35

mchitre commented 5 years ago

Thanks for the feedback.

I don't know the exact setup you used, but it looks like the environment parameters you are using is generating too many rays. You may wish to only generate eigenrays rather than all rays. In any case, please refer to Bellhop documentation on how the modeling is controlled by the parameters, as all arlpy does is wrap Bellhop, and does not do the ray computations itself. Do note that eigenray computation in Bellhop gives a set of rays that pass close to the receiver (not necessarily exact).

As for use of Bokeh, I take your point on it being resource intensive for graphs with lots of data. I am evaluating use of Plotly instead of Bokeh, but haven't made that decision yet.

ZHANGGU commented 5 years ago

Hei

Thanks for your reply.

Actually, I set up the env very simple according to your documentation.

I have no idea why so many rays are computed, and will have a look about the manual of course.

Have a nice day

Regards

Guosong

ZHANGGU commented 5 years ago

Hi Do you have detailed Bellhop manual? I can see one manual of 2011 on the web.

Thanks.

mchitre commented 5 years ago

Yes, I believe the 2011 one is the most recent one that I know of.

ZHANGGU commented 5 years ago

Hi, I did read the 2011 manual which is an intro really. I am still wondering at you would have the intro about output data structures. If yes, could you please share? As my simulations as shown are not correct at all. I am stucked in getting the correct rays. Appreciate your help!

mchitre commented 5 years ago

The original acoustics toolbox has a number of MATLAB scripts to deal with the output. Those are the best source of information on the output formats for me, as they are current and work. My Python implementations were based on that, and as far as I can tell from the use cases I use it for, that seems to work fine.

ZHANGGU commented 5 years ago

Thanks. I will take look about the m-codes. while I would not expect people enjoy such eigen ray plot as shown :-)