[Scilab-users] simp_mode odesn't seem to work

[Federico Miyara]

Dear All,

I need to have examples of rationals with very close poles and zeros, 
but the simp engine simplifies them. This is awkward, since the simp 
documentation says there is no threshold, so other than identical poles 
and zeros shouldn't be simplified. Actually this is the case with 
complicated rationals created from electrical network analysis of some 
filters, but not with the more transparent example below.

My second attempt to reproduce the problem in my simple numeric example 
has been to use the command simp_mode to disable any simplification. But 
it doesn't work as expected--and described. For instance, the code

N = poly([1 2 3 3.000009 5], "s")
D = poly([1.0000001 3.0000003 4 6], "s")
simp_mode = %f;
H = N/D

yields these results:

--> N
  N  =
   -90.00027 +213.00055s -184.00037s² +74.000099s³ -14.000009s⁴ +s⁵
--> D
  D  =
   72.000014 -126.00002s +67.000005s² -14s³ +s⁴
--> H
  H  =

    -30.000093 +31.000066s -10.00001s² +s³
    --------------------------------------
          24.000006 -10.000001s +s²

As can be seen, even if I disabled automatic simplification, the result 
is not the raw quotient N/D but a "simplified" version.

Any idea? Is it a bug?

Regards,

Federico Miyara


-- 
El software de antivirus Avast ha analizado este correo electrónico en busca de virus.
https://www.avast.com/antivirus
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20200827/e1b7c57f/attachment.htm>