Jump to content
GMS, SMS, and WMS User Forum

Jonas

Members
  • Posts

    36
  • Joined

  • Last visited

Posts posted by Jonas

  1. Hi, 

    I still have the same problem but with another model this time. The problem happens when I add additional parameters for the PEST-estimation. No difference between a PEST calibration with SVD or SVD-assist.  Any ideas on how to solve this issue would be appreciated. I run GMS 10.3.5.

    Jonas

  2. This error still seems to be present when running Stochastic Modflow. I still get the same error message as described above. There is no difference if all solutions converges or if there are a few solution that are not converging. The model works perfectly fine in a forward run. I use the NWT solver. Any ideas on how to solve this issue?

  3. Hi, stupid question but how do I change from estimation mode to regularization mode in GMS? I have tried to use the PARREP utility to make a change in the pst (PEST control file) file but GMS keeps to run in estimation mode.

    6 hours ago, Michal said:

    To your new problem. Try turning SVD-Assist and Regularization off and set NOPTMAX to 0.

    I tried this and then the model was able to run once (really fast). I have however not examined the results with PESTCHEK.

    As I mentioned earlier, I have been able to make this work without fixed pilot points.

  4. I just tested the model again without parallel PEST and got the same error message: "pest has stopped working". I get this when I make a parameter estimation run with noptmax = -2.

    I also got this error message when from this run:

     

    The whole process before seems to work fine

    image.png

  5. Often when I initiate a PEST run with SVD-assist and NOPTMAX = -2 or -1 the process ends with the message "cannot find Jacobian array file".

    The process starts with a calculation of the Jacobian matrix:

    image.png.cc60d353054f7b7d9cd90376de7b6a72.png

    But ends like this:

    image.thumb.png.ab35c98b9465a8c4ed764b34b655b032.png

    It always works when I run the model without pilot points. A forward run with the pilot points also works.

    I have managed to run both a parameter estimation and NSCM with SVD-assists before with the same model constraints but when experience this problem when I attempt to make changes in the model. 

    Ideas on how to solve this issue?

  6. An update, my own model was possible to run after using the SUBREG1 utility in PEST. A problem is that my own model with a fixed pilot point is only possible to run for log-uniform distributions with the RANDPAR utility. When I try to create log-normal distributions in RANDPAR i get the error message seen in the screenshot below. Ideas on how to solve this issue are appreciated.  

    randpar_error.png

  7. An update on this matter. As suggested by Michal, I was able to run the PARREP and RANDPAR utility in PEST to create the *.par files for the different runs. Before running PARREP, I ran a parameter estimation run in GMS with NOPTMAX = -1 to create a *.bpa file. After running RANDPAR I was able to perform the normal procedure for a NSMC inverse calibration run in GMS. This worked out well for the NSMC tutorial with "sc1v56" as a fixed pilot points (keeping value 400 in screenshot below). I cross my fingers that it also works for my own model. However, I would appreciate if this procedure could work directly in GMS. 

    nsmcII_PEST_fixedPP.png.3d39957c7cd364531c47aac4707d42ff.pngnsmcII_PEST_fixedPP_GMS.thumb.png.ba7bccafea99a33fed13eacb558b8749.png

  8. I have problems running PEST NSMC with fixed pilot points. Both in my own model and in the NSMC tutorial files, I get the same error message "Cannot open parameter value file nsmcII_200_random_nul.par." and the simulation finishes without completing any iteration. Is this a bug i in GMS?

    Kind regards,

    Jonas

    Error fixed pilot point.JPG

  9. Thanks Michal and Julian for your replies. Eventually I might end up running PEST outside of GMS if it does not work.

    I managed to get a parameter estimation run with svd-assist to work. I also managed to get a PEST NSMC run to work. But, when I want to repeat the procedure with fixed pilot points I get a message: "Cannot find sm20s20000_random_nul1.par". I cannot find the corresponding file in the folder of the model file neither. As I wrote in my first post I find it quite arbitrary when PEST works together with GMS. I would be happy with a response from someone from GMS's developing team on this issue. 

  10. Thanks for your quick reply Michal, 

    I agree on that it seems like PEST is doing its job but GMS does not read in the result properly. But I can't find the reason for this since the same modelling procedure has worked for other models. Quite frustrating.

    How do use the PARREP utility? Should I use it outside of GMS interface?

    Kind regards,

    Jonas

  11. This is what typically happens when running with SVD-assist. As can be seen from the blue marked line, PEST is able to reduce the error to 98.885 from the initial error of 155.76. But, then something goes wrong since this solution never comes up as a iteration before the parameter estimation is finished.

    SM13t7.PNG

  12. I often experience problems when running a PEST parameter estimation model with SVD-assist. I find it quite arbitrary when it work and when it does not work. For some occasions it seems to work just by moving the modelling files to another computer and run the model on that workstation. For my current model, SVD works but SVD-assist does not. To reach reasonable results with NSMC is even more problematic, in particular for transient models. I run the latest nightly build and NWT as solver. 

    Someone with similar experiences or suggestions to a more streamlined and stable model procedure? 

  13. Unfortunately I still experience the stress-periods as quite buggy. The images below show stress periods as date/times and as seconds respectively. The result is from a PEST NSMC simulation with a transient pump-test as calibration target. During the initial forward runs and PEST calibration without NSMC I did not experience the same issue with the date-time series. The simulation itself seems to work fine but the problem arises when creating graphs and comparing the simulated results with my observations. Since I think there is a bug behind the issue I would appreciate if someone from the GMS development team could look into it.

    Kind regards

    Jonas 

    Stress periods date-times.PNG

    Stress periods seconds.PNG

  14. Hi,

    I experience the tool for defining stress periods as dates/times as quite buggy. Please refer to screenshot below.

    First, there is a different format between the first row and the second two rows (month as number "10" instead of letters "Oct"). But the most significant issue is that the "length" changes to weird numbers when the date and time is changed. In the example below the first stress period has a negative number but I have also experienced zero as length. My workaround has been to not make changes as date/times by unmarking the "use dates/times" box, calculate the corresponding length between and then change back to "use dates/times". This works but not so convenient.

    Another issue is that I have seconds as time format in my model but to get the length right I have to divide the seconds by 2 in the "length" box to get the date/times right.

    Finally, the model terminates with an error when I use date/times but works when I am un-tickning the box mentioned above (not using date/times).

    Any ideas on how to get these issues sorted out?

    Kind regards,

    Jonas

     

     

    Stress beriods bug.PNG

×
×
  • Create New...