[Scilab-users] Help us define what Scilab can do in control system

[Samuel Gougeon]

Hello Yann,

Why not starting from Serge Steer's proposal and discussing it:
http://bugzilla.scilab.org/14239
Actual proposal: http://bugzilla.scilab.org/attachment.cgi?id=4184

In a general case, the only thing i would say about it (as about any 
other one), is that
no subsection should have less that 4-5 entries. For proposed 
sub(-sub)sections with
less than 4 entries, these ones should be transfered at the root of the 
parent section.
E.g.:

Control Design
   /Linear quadratic
     gcare
     ... ~10 entries : OK
   /Pole  placement
     kpure
     krac2
     ppol
     stabil
   /Tracking
     gfrancis              NOK
   /Disturbance decoupling
    ddp                  NOK
   /H-infinity
     ccontrg
     dhinf
     dhnorm
     ... > 5 entries : OK

*should rather be*

Control Design
   /Linear quadratic
     gcare
     ... ~10 entries : OK
   /Pole  placement
     kpure
     krac2
     ppol
     stabil
   /H-infinity
     ccontrg
     dhinf
     dhnorm
     ... > 5 entries : OK
   gfrancis ddp


For my part, i hate loosing my time in exploring deep levels having 
almost nothing.
 From having for years not a single subsection with 140 functions on the 
same common box, to having too many subsections poorly populated, i 
think there is a wise middle way.
If further functions are implemented and may feed a new common 
subsection, then it will never be too late to create it later when 
really needed and meaningful.

Best regards
Samuel

Le 18/04/2016 20:08, Yann DEBRAY a écrit :
> Hello Scilab user,
>
> We would like to develop Scilab further in specific fields such as 
> *Control System* and that starts by making Scilab features clearer. 
> For now the Scilab online help 
> <https://help.scilab.org/docs/5.5.2/en_US/section_46d43fa43588512dc7dc2c3d190b775b.html> 
> is quite hard to operate. So, we are beginning to organize it starting 
> with the control system (CACSD 
> <https://help.scilab.org/docs/5.5.2/en_US/section_64a8529216e858b335b0e6c058385350.html>) 
> features. Here is a first shot:
>
> *LTI model generation and manipulation*
> # Discrete and continuous time
> # State-space and transfer function representations
> *Analyzing model*
> # Evans for root locus
> # Steady state error
> # Bode, Nyquist and Black
> # Time, frequency response
> *Designing and tuning control systems*
> #PID, LQG tuning
>
> It is incomplete and obviously superficial. That is why we need a 
> competent opinion.
> --> Please give us feedback about how you would like it to be organized ;)
>
> Yann Debray
> For the Scilab Team
>
>
>
>
> _______________________________________________
> users mailing list
> users at lists.scilab.org
> http://lists.scilab.org/mailman/listinfo/users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20160418/332cf16a/attachment.htm>