What migh be producing these huge amount of Warnings in my FL control?

3 views (last 30 days)
Dear all,
I have a battery control using Fuzzy Logic, it seems to be working properly, but I am receiving a tons of warnings like these:
> In fuzzy.internal.utility.throwWarning (line 14)
In PVGeo_based_LG (line 123)
Warning: In 'Full_PED_FL/FL_Battery_Control/Fuzzy Logic
Controller', no rules fired for Output 2. Defuzzified output value set to its mean range value 0.5.
> In fuzzy.internal.utility.throwWarning (line 14)
In PVGeo_based_LG (line 123)
Warning: In 'Full_PED_FL/FL_Battery_Control/Fuzzy Logic
Controller', no rules fired for Output 1. Defuzzified output value set to its mean range value 0.5.
What does this mean ? how can I overcome this misbehaviour?
Best regards
  4 Comments
Sam Chak
Sam Chak on 22 Feb 2022
I see. When your Power Error = 0, none of rules is fired, because Power Error only has Negative MF (< 0) and Positive MF (> 0).
It is similar to the common issue in low-level collision avoidance systems. If the Obstacle is on the left, then STEER right (+ve angle). If the Obstacle is on the right, then STEER left (–ve angle). When the Obstacle is exactly in the middle, it creates a "confusion" because no rules fired for the STEER output. And if the defuzzified output value set to its mean range value, which is 0° (no steering occurs), then disastrous collision may happen.
The same like playing in Badminton Doubles. If opponents are in a defensive position, and the center of their court is open, then smashing to the middle will cause confusion becaise the opponents have to mentally decide in a split second who must return.
As a first-line rectification, try modifying the membership functions in Power Error to look like in the figure below where the adjacent triangular fuzzy sets cross at 0.5, in which the middle truth-value 0.5 can be interpreted as "undecided". If you fix that, then Rules #5 , #8, #11, and #14 will be fired.
Tony Castillo
Tony Castillo on 23 Feb 2022
Thank you so much for your guidance and explanation, you were right in your workaround. It is working properly.

Sign in to comment.

Accepted Answer

Sam Chak
Sam Chak on 23 Feb 2022
It is good to hear that my suggestion described above works for your case.

More Answers (0)

Categories

Find more on Fuzzy Logic in Simulink in Help Center and File Exchange

Products


Release

R2021a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!