[Octave-bug-tracker] [bug #45918] control: Nyquist plot is incorrect

[Lukas Reichlin]

Follow-up Comment #8, bug #45918 (project octave):

No, the original problem is not fixed.

The original problem is that the nyquist function doesn't show the interesting
part of the plot entirely. If the user doesn't specify a frequency range or
vector, nyquist tries to figure it out from the zeros and poles of the system.
This part of the code could be more sophisticated indeed. In the case of


sys = tf ([-10 10], [1 1 0])


the lower frequency bound is too high such that the part which goes to
infinity is not shown. On the other hand, if the lower frequency bound gets
too low, the values on the imaginary axis go quickly towards infinity such
that the part between -3 and 3 can't be seen anymore.

The task is to improve the code which computes the upper and lower frequency
bounds. Not only for the system above, but for a wide variety of systems.


    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?45918>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/