inverse goes into loop (II)

A discussion forum for users of the new HYDRUS 2D/3D. HYDRUS is a software package for simulating water, heat and solute movement in two- and three-dimensional variably saturated media. Happy Posting!
Post Reply
tomgreen
Posts: 11
Joined: Sun Dec 03, 2017 9:13 am
Location: Israel

inverse goes into loop (II)

Post by tomgreen » Wed Aug 22, 2018 10:17 am

hi,
when running the inverse solution to determine solute tranpsort parameters the model goes into a loop of unreadable numbers at the end of the simulation. a year ago i had a very similar problem which was solved- now something else is amiss.
hope you can take a look,
thanks!
tom
Attachments
to forum.zip
(313.76 KiB) Downloaded 56 times

Jirka
Posts: 4759
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: inverse goes into loop (II)

Post by Jirka » Thu Aug 23, 2018 9:57 am

You cannot have Beta (the Freundlich exponent) equal to zero. It should be equal to 1. J.

tomgreen
Posts: 11
Joined: Sun Dec 03, 2017 9:13 am
Location: Israel

Re: inverse goes into loop (II)

Post by tomgreen » Thu Aug 23, 2018 1:49 pm

hi, put beta to 1, but it still goes into that funny loop at the end.
tom

tomgreen
Posts: 11
Joined: Sun Dec 03, 2017 9:13 am
Location: Israel

Re: inverse goes into loop (II)

Post by tomgreen » Sun Aug 26, 2018 1:06 pm

if i put zero for the max number of iterations in the inverse solution window it also goes into this funny loop at the end,
but there are results. if the number of iterations is more than zero there are no results.
tom

Jirka
Posts: 4759
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: inverse goes into loop (II)

Post by Jirka » Wed Aug 29, 2018 1:24 pm

I have cleaned up your project and had no trouble running it for more than zero inverse iterations. However, your initial guess of dispersivities was pretty good and I have to admit that I do not see much reason for trying to improve the fit further. See the comparison between results for initial parameters (left) and after 2 iterations (right). J.
Attachments
Obs_node_Conc.png
Obs_node_Conc.png (41.57 KiB) Viewed 1287 times

tomgreen
Posts: 11
Joined: Sun Dec 03, 2017 9:13 am
Location: Israel

Re: inverse goes into loop (II)

Post by tomgreen » Wed Aug 29, 2018 3:27 pm

thanks Jirka!

i made a list of the 'clean-up' you did so that anyone that has the same issue can be helped by it:

time information window:
lowered the minimum time step in the from 1.44e-017 to 1.44e-005
output information window:
turned on T-level info to 25 in the (it was turned off)
set the interval output to 10 min (it was 1)
turned off the screen output (it was on)
iteration criteria:
put the number of iterations down to 10 (from 25)
solute transport window:
put the absolute concentration tolerance down to 0 (from 0.1)
reactions parameters for solute material one window:
put higher values for disp T and disp L (probably to see if they would change from what i had originally put, did not put anything in the minimum or maximum values)
put beta to 1

Post Reply