function maximize/minimize in Smath Studio

function maximize/minimize in Smath Studio - Help! - Messages

#21 Posted: 5/4/2021 8:33:27 PM
gallyon1

gallyon1

0 likes in 6 posts.

Group: User

hello. I have the same problem with the transfer from mathkad to smath and the maximize function . Could you help me? I would really appreciate it . Sorry English is not perfect either here or in the pdf fileMathcad - mag. 2020 Vershinina Ehvelina Optimizacija slivochnogo krema1.pdf (113 KiB) downloaded 42 time(s).
#22 Posted: 5/4/2021 8:53:18 PM
Alvaro Diaz Falconi

Alvaro Diaz Falconi

992 likes in 1674 posts.

Group: User

Wrote

hello. I have the same problem with the transfer from mathkad to smath and the maximize function ...



Hi. If you upload the SMath file maybe someone can help you.

Best regards.
Alvaro.
#23 Posted: 5/5/2021 4:35:10 AM
gallyon1

gallyon1

0 likes in 6 posts.

Group: User

yeah totally forgot about it sbalancirovannost_39__organoleptika_zefira_-_f-ii_Ketle_new.sm (19 KiB) downloaded 37 time(s).
#24 Posted: 5/5/2021 7:27:06 AM
Alvaro Diaz Falconi

Alvaro Diaz Falconi

992 likes in 1674 posts.

Group: User

Wrote

yeah totally forgot about it sbalancirovannost_39__organoleptika_zefira_-_f-ii_Ketle_new.sm (19 KiB) downloaded 37 time(s).



Hi. You have an issue here, and I don't know if it's because SMath can't calculate very well the exponential function or why, but the value for Ko(x) aren't the same in Mathcad and SMath, and thus the values for the target function K(x).

In the attached, the procedure for the "ad hoc" minimizer.

sbalancirovannost_39__organoleptika_zefira_-_f-ii_Ketle_new.sm (53 KiB) downloaded 43 time(s).

Best regards.
Alvaro.
1 users liked this post
gallyon1 5/7/2021 11:49:00 AM
#25 Posted: 5/5/2021 8:04:26 AM
gallyon1

gallyon1

0 likes in 6 posts.

Group: User

Oh I see thank you But final answer doesnt look like in MathCad sbalancirovannost_39__organoleptika_zefira_-_f-ii_Ketle_new (3).sm (53 KiB) downloaded 34 time(s).
#26 Posted: 5/5/2021 10:30:31 AM
Alvaro Diaz Falconi

Alvaro Diaz Falconi

992 likes in 1674 posts.

Group: User

Wrote

Oh I see thank you But final answer doesnt look like in MathCad



Hi. You must to choose the adequate parameter "percent" p for see if you can raise up the maximum. For instance this other seems to hold all the restrictions with a maximum only just 8 orders of magnitude bigger than Mathcad answer. Try to put the SMath xo value in Mathcad as new guess for the solve block and see if you get the same maximum for K(x) in Mathcad.

sbalancirovannost_39__organoleptika_zefira_-_f-ii_Ketle_new.sm (90 KiB) downloaded 42 time(s).

Best regards.
Alvaro.
1 users liked this post
gallyon1 5/7/2021 11:49:00 AM
#27 Posted: 5/7/2021 10:35:31 AM
gallyon1

gallyon1

0 likes in 6 posts.

Group: User

If i use this x0 in MathCad I get the same XFile not found.. x=X. I tried a lot of p values but I didnt get close result. Results are always too big.File not found.
#28 Posted: 5/7/2021 1:19:13 PM
Alvaro Diaz Falconi

Alvaro Diaz Falconi

992 likes in 1674 posts.

Group: User

Wrote

If i use this x0 in MathCad I get the same XFile not found.. x=X. I tried a lot of p values but I didnt get close result. Results are always too big.File not found.



Hi gallyon1. You are looking for a maximum. So, for your problem, bigger is better. SMath returns a bigger result, but that's not all: the results verify all the problem constraints, which make it valid answer for the problem.

Maybe I don't explain well: try to check what happen with mathcad if you use the SMath answer as the guess, this is substitute the x value in Mathcad for the xo value from SMath here:

Clipboard01.jpg

But actually that's for compare. SMath solution checks all the constraints, then it's a candidate for the problem solution. You "didn't get close result", but I don't see a problem there, where I see a problem is if you want a close result, this is a result near to 8.8*10^8 you need to review your physical system and check if there are some additional constrictions for some variable or function which make your result "close" to that value. If not, SMath solution it's just a good solution, and Mathcad's isn't.

Best regards.
Alvaro.
2 users liked this post
gallyon1 5/7/2021 1:37:00 PM, Denis 5/10/2021 7:40:00 AM
#29 Posted: 5/7/2021 1:39:04 PM
gallyon1

gallyon1

0 likes in 6 posts.

Group: User

Ok I got it. Thank you all Maybe I was too attached to MathCad solution.
  • New Posts New Posts
  • No New Posts No New Posts