Skip to main content
ppal_25568718-Opal
18-Opal
July 5, 2024
Solved

Ignore non solution in Solve Block

I have a list of data. Some do not give a solution within the tolerance. What is the best way to ignore and move onto the next set of data? Is there a way to set a more coarse tolerance value?

Prime 10 attached.

ppal_0-1720152305169.png

 

Werner_E
Werner_EAnswer
25-Diamond I
July 5, 2024

I would suggest to use "try ... on error ..." to catch the errors and return a NaN in that case.

If you like you may use "filterNaN" to get rid of the NaN's

Werner_E_0-1720156098858.png

Prime 10 file attached

View in thread

4 replies

Werner_E
Werner_E25-Diamond IAnswer
25-Diamond I
July 5, 2024

I would suggest to use "try ... on error ..." to catch the errors and return a NaN in that case.

If you like you may use "filterNaN" to get rid of the NaN's

Werner_E_0-1720156098858.png

Prime 10 file attached

Translate
    ppal_25568718-OpalAuthor
    18-Opal
    July 7, 2024

    Used minerr which worked for me in this case.

    Translate
    Werner_E
    Werner_E25-Diamond I
    25-Diamond I
    July 8, 2024

    If using "minerr" works OK for you, that's fine.

    But you have to be aware that you may get inaccurate or even wrong (depending on your application) results without you noticing it.

     

    In your case, the vector calculated in the pic ideally should consist of all zeros.
    Apart from the first and last entry the values are way off!

    Werner_E_0-1720409847962.png

     

    Translate
    ppal_25568718-OpalAuthor
    18-Opal
    July 8, 2024

     Yes there are wild fluctuations with some clustering. It is essentially locating faults on a transmission line.

     

    On a good day it should look like    :

    ppal_0-1720412921035.png

    Thank you for your help.

     

    Translate
    Terms & ConditionsAccessibility statement