search

XuanyiLin / SWMM2PEST2.0

SWMM2PEST 2.0
6 stars 10 forks source link

Why are the output data after calibration always the same with the data before calibration? #11

Open LiamNeol opened 2 years ago

LiamNeol commented 2 years ago

Hi Xuanyi, I meet a confusing problem while I'm using SWMM2PEST2.1. It is the output data doesn't change after calibration, see the picture below. Hope that you can help me out of this problem, thanks for your time. The same results: Snipaste_2022-08-05_17-41-38

Part of the reports from pycharm:

List of selected pars: 
[]
in selected for estimation

 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 1.00 seconds.

   Sum of squared weighted residuals (ie phi) =   7.2604    

OPTIMISATION ITERATION NO.        : 1
   Model calls so far             : 1
   Starting phi for this iteration:   7.2604    

   Calculating Jacobian matrix: running model 1 times .....ϵͳ�Ҳ���ָ�����ļ���

 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 1.00 seconds.

   1 runs completed. 

       Lambda =   5.0000     ----->
       running model .....
 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 1.00 seconds.

          Phi =   6.2767      (  0.865 of starting phi)

       Lambda =   2.5000     ----->
       running model .....
 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 1.00 seconds.

          Phi =   6.2767      (  0.865 of starting phi)

   No more lambdas: relative phi reduction between lambdas less than 0.0100
   Lowest phi this iteration:   6.2767    
   Maximum   factor change:  1.010     ["f0prcnt_ntll"]
   Maximum relative change: 9.9187E-03 ["f0prcnt_ntll"]

OPTIMISATION ITERATION NO.        : 2
   Model calls so far             : 4
   Starting phi for this iteration:   6.2767    

   Calculating Jacobian matrix: running model 1 times .....ϵͳ�Ҳ���ָ�����ļ���

 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 1.00 seconds.

   1 runs completed. 
   phi gradient zero in non-frozen parameter space
   Switch to higher order derivatives calculation
   (restart from best parameters so far - these achieved at iteration    1)

OPTIMISATION ITERATION NO.        : 3
   Model calls so far             : 5
   Starting phi for this iteration:   6.2767    

   Calculating Jacobian matrix: running model 2 times .....ϵͳ�Ҳ���ָ�����ļ���

 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 0.00 seconds.

   1 runs completed.
 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 1.00 seconds.
   2 runs completed. 

       Lambda =   2.5000     ----->
       running model .....
 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 0.00 seconds.

          Phi =   6.2767      (  1.000 times starting phi)

       Lambda =   1.2500     ----->
       running model .....
 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 0.00 seconds.

          Phi =   6.2767      (  1.000 times starting phi)

   No more lambdas: relative phi reduction between lambdas less than 0.0100
   Lowest phi this iteration:   6.2767    
   Maximum   factor change:  1.000     ["f0prcnt_ntll"]
   Maximum relative change: 2.9164E-04 ["f0prcnt_ntll"]
ϵͳ�Ҳ���ָ�����ļ���

OPTIMISATION ITERATION NO.        : 4
   Model calls so far             : 9
   Starting phi for this iteration:   6.2767    

   Calculating Jacobian matrix: running model 2 times .....
 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 0.00 seconds.

   1 runs completed.
 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 0.00 seconds.
   2 runs completed. 

       Lambda =   1.2500     ----->
       running model .....
 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 0.00 seconds.

          Phi =   6.2767      (  1.000 times starting phi)

       Lambda =  0.62500     ----->
       running model .....
 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 1.00 seconds.

          Phi =   6.2767      (  1.000 times starting phi)

   No more lambdas: relative phi reduction between lambdas less than 0.0100
   Lowest phi this iteration:   6.2767    
   Maximum   factor change:  1.000     ["f0prcnt_ntll"]
   Maximum relative change: 3.2538E-04 ["f0prcnt_ntll"]

   Optimisation complete:   3 optimisation iterations have elapsed since lowest
                          phi was achieved.
   Total model calls:     13

Running model one last time with best parameters.....
 o  Retrieving project data
 o  Simulation complete           
... EPA-SWMM 5.1 (Build 5.1.13)

... EPA-SWMM completed in 1.00 seconds.

Part of groof09q1.rec :

 OPTIMISATION RECORD

INITIAL CONDITIONS: 
   Sum of squared weighted residuals (ie phi) =   7.2604    

      Current parameter values
      f0prcnt_ntll     70.0000    

OPTIMISATION ITERATION NO.        : 1
   Model calls so far             : 1
   Starting phi for this iteration:   7.2604    

       Lambda =   5.0000     ----->
          Phi =   6.2767      (  0.865 of starting phi)

       Lambda =   2.5000     ----->
          Phi =   6.2767      (  0.865 of starting phi)

   No more lambdas: relative phi reduction between lambdas less than 0.0100
   Lowest phi this iteration:   6.2767    

      Current parameter values                 Previous parameter values
      f0prcnt_ntll     70.6943                 f0prcnt_ntll      70.0000    
   Maximum   factor change:  1.010     ["f0prcnt_ntll"]
   Maximum relative change: 9.9187E-03 ["f0prcnt_ntll"]

OPTIMISATION ITERATION NO.        : 2
   Model calls so far             : 4
   Starting phi for this iteration:   6.2767    
   phi gradient zero in non-frozen parameter space
   Switch to higher order derivatives calculation
   (restart from best parameters so far - these achieved at iteration    1)

      Current parameter values
      f0prcnt_ntll     70.6943    

OPTIMISATION ITERATION NO.        : 3
   Model calls so far             : 5
   Starting phi for this iteration:   6.2767    

       Lambda =   2.5000     ----->
          Phi =   6.2767      (  1.000 times starting phi)

       Lambda =   1.2500     ----->
          Phi =   6.2767      (  1.000 times starting phi)

   No more lambdas: relative phi reduction between lambdas less than 0.0100
   Lowest phi this iteration:   6.2767    

      Current parameter values                 Previous parameter values
      f0prcnt_ntll     70.6737                 f0prcnt_ntll      70.6943    
   Maximum   factor change:  1.000     ["f0prcnt_ntll"]
   Maximum relative change: 2.9164E-04 ["f0prcnt_ntll"]

OPTIMISATION ITERATION NO.        : 4
   Model calls so far             : 9
   Starting phi for this iteration:   6.2767    

       Lambda =   1.2500     ----->
          Phi =   6.2767      (  1.000 times starting phi)

       Lambda =  0.62500     ----->
          Phi =   6.2767      (  1.000 times starting phi)

   No more lambdas: relative phi reduction between lambdas less than 0.0100
   Lowest phi this iteration:   6.2767    

      Current parameter values                 Previous parameter values
      f0prcnt_ntll     70.6507                 f0prcnt_ntll      70.6737    
   Maximum   factor change:  1.000     ["f0prcnt_ntll"]
   Maximum relative change: 3.2538E-04 ["f0prcnt_ntll"]

   Optimisation complete:   3 optimisation iterations have elapsed since lowest
                          phi was achieved.
   Total model calls:     13

   The model has been run one final time using best parameters. 
   Thus all model input files contain best parameter values, and model 
   output files contain model results based on these parameters.
LiamNeol commented 2 years ago

Still the same results even if I open the SWMM2PEST with the SWMM2PEST.exe instead of Pycharm.