LS-Opt curve matchi...
 
Notifications
Clear all

[Solved] LS-Opt curve matching

8 Posts
2 Users
3 Reactions
53 Views
Posts: 10
Topic starter
(@richardp)
Student
Joined: 3 weeks ago

Hi,

i am getting this lsopt error:

image

 I am using a curve like this as target:

0.21833666380167788, 0.0
0.2800215564644908, 0.007227182223262746
0.34170644912730375, 0.013495961247278466
0.4033913417901167, 0.019354151757712462
0.4650762344529296, 0.02534956844023005
0.5267611271157426, 0.03203002598049655
0.5884460197785555, 0.03994333906417723
0.6501309124413683, 0.04903401943102264
0.7118158051041814, 0.056785852301871005
0.7735006977669943, 0.06005532963270841
0.8351855904298073, 0.056888433205581374
0.8968704830926202, 0.05210871707795717
0.958555375755433, 0.05293671143712328
1.020240268418246, 0.06659571446303642
1.081925161081059, 0.10017821744732189
1.1436100537438718, 0.15496694383371676
1.2052949464066849, 0.22424929733715748
1.2669798390694977, 0.30073083184337956
1.3286647317323108, 0.3777936370812482
1.3903496243951237, 0.45112864426132177
1.4520345170579367, 0.5169188916869559
1.5137194097207496, 0.5714166388595009
1.5754043023835624, 0.6134678047666188
1.6370891950463755, 0.6452323225471168
1.6987740877091884, 0.6690861358189248
1.7604589803720012, 0.6874985926304731
1.8221438730348143, 0.704693704663318
1.8838287656976271, 0.7264558615149429
1.94551365836044, 0.7586238322519033
2.007198551023253, 0.8057387112527168
2.0688834436860657, 0.8652832218046606
2.1305683363488788, 0.9323518011926345
2.192253229011692, 1.0009968673641032
2.253938121674505, 1.0611372461754351
2.3156230143373175, 1.101665051562172
2.3773079070001306, 1.1143568077723236
2.4389927996629437, 1.1025641866887619
2.5006776923257563, 1.0728860814116237
2.5623625849885694, 1.0334684712252842
2.6240474776513825, 0.993735326875853
2.6857323703141955, 0.975664345763689
2.747417262977008, 1.0108642161258201
2.8091021556398212, 1.1222393818877028
2.8707870483026343, 1.2871180455098623
2.932471940965447, 1.468357595412725
2.99415683362826, 1.6344729337654547
3.055841726291073, 1.7831041466107345
3.1175266189538857, 1.9212276664058257
3.179211511616699, 2.0558226536967767
3.240896404279512, 2.1933099511286365
3.3025812969423245, 2.3387255007553502
3.3642661896051376, 2.4968941900381245
3.4259510822679506, 2.6739790539644783
3.4876359749307633, 2.8885912182502484
3.5493208675935763, 3.1612358030105776
3.6110057602563894, 3.4608130748322132
3.672690652919202, 3.7303446683801904
3.734375545582015, 3.9761891977604256
3.796060438244828, 4.2472131763602725
3.8577453309076413, 4.581520516918969
3.919430223570454, 4.919580454199814
3.981115116233267, 5.168982883297691
4.04280000889608, 5.393169841101814
4.104484901558893, 5.742607841536437
4.166169794221705, 6.286697689020696
4.227854686884518, 6.883054101637552
4.2895395795473314, 7.4411329468305665
4.351224472210144, 8.007486224455315
4.412909364872957, 8.614510108473805
4.47459425753577, 9.212542492674828
4.536279150198583, 9.832681290312351
4.5979640428613955, 10.610251375542873
4.659648935524209, 11.54350164324941
4.721333828187022, 12.459053674666974
4.783018720849834, 13.282250981546419
4.844703613512648, 14.085886266575118
4.90638850617546, 14.952519712947788
4.968073398838273, 15.902122327309515
5.0297582915010866, 16.842462450756198
5.091443184163899, 17.739585648673376
5.153128076826713, 18.665034665074668
5.214812969489525, 19.658010846279264
5.276497862152338, 20.66591705759585
5.3381827548151515, 21.623709057757807
5.399867647477964, 22.518384030638735
5.461552540140777, 23.63967363949115
5.52323743280359, 24.687717544011807
5.584922325466403, 24.846560137859722
5.6466072181292155, 25.58724252192522
5.708292110792029, 26.58911994378675
5.769977003454842, 27.2636268385939
5.831661896117654, 27.921235826698595
5.893346788780468, 28.630604743585604
5.95503168144328, 29.276725136180552
6.016716574106093, 29.761493028801326
6.078401466768907, 30.030716535583075
6.140086359431719, 30.124338196109598
6.201771252094532, 30.1029615307222
6.263456144757345, 30.02719005976219
6.325141037420158, 29.95762730357086

 

What am i doing wrong?

some advice would be much appreciated.

7 Replies




Posts: 10
Topic starter
(@richardp)
Student
Joined: 3 weeks ago

If i switch to curve_mapping it works.... but i do not understand why not with RMSE.

image

Now i get the next error.

 

Bild

and

13:01:04  Number of points used for metamodel = 7
13:01:04  
13:01:04   ALL THE METAMODELS HAVE BEEN ANALYZED: continuing to next step
13:01:04  
13:01:04  
13:01:04  
13:01:04  None of the experiments (designs) has history displacement (Iteration 1)

But if i run the LS-RUN the created files manually, i get a binout and it holds a nodout for the node 2000434 i specified in the displacement history.

 

image

 

 

image

 

Reply




Posts: 13
(@joedoe)
Student
Joined: 6 years ago

This error occurs when all points in your curve matching comparison are outside the specified range. This typically happens due to one of these common reasons:

  1. The ranges specified in your curve matching setup are too narrow
  2. There's a scale mismatch between the curves being compared
  3. The computed/simulated curve is significantly offset from the target curve

To fix this, you could try:

  1. Check and possibly increase your curve matching range settings
  2. Verify that both curves are using the same units/scale
  3. Plot both curves to visually inspect their alignment
  4. Review the scale factors in your curve matching definition

The difference between curve_mapping and RMSE (Root Mean Square Error/Mean Square Error) behavior in LS-OPT comes down to how they handle curve comparisons:

  1. Curve_mapping is more forgiving because:
    • It creates a mapping between points on both curves
    • Can handle curves with different x-coordinates more flexibly
    • Often works better when curves are significantly offset
  2. RMSE/MeanSqErr is more strict because:
    • It compares curves at exact x-coordinates
    • Requires better initial alignment between curves
    • Is more sensitive to range settings
    • Points outside the specified range are completely omitted

Since curve_mapping works but RMSE doesn't, this suggests your curves might be:

  • Significantly offset from each other
  • Have different x-coordinate distributions
  • Have points falling outside your specified comparison range

To make RMSE work, you could try:

  1. Widening your comparison range
  2. Pre-aligning your curves better
  3. Adjusting your initial design parameters to get curves closer together
Reply




Posts: 10
Topic starter
(@richardp)
Student
Joined: 3 weeks ago

 

Hello Joe Doe,

thank you very much for your thorough explanation.

However, I am unsure where the problem lies. LS-OPT brings up the message before any simulation is being conducted. There is no binout, nothing could have been compared to the target curve I guess. In the objective function i have not defined any constraints.
Therefor probably something is generally amiss with my target curve. It should depict a measured force-displacement curve, however, it is just taken with plotdigitizer for me to start with an example. The curve i generated with plotdigitizer was then interpolated with python to have evenly spaced x-values. 

I attached the lsopt files, the key file, the output and the target curve.

 

Regards,

Richard

Reply




Posts: 10
Topic starter
(@richardp)
Student
Joined: 3 weeks ago

If I run the keyfile manually, extract from binout hopefully the same curves as defined in lsopt, and compare to the target, i get this:

image
Reply




Page 1 / 2




Share: