I want to implement a fuzzy control algorithm that has the following rule
base:
1. if A=low and B=high then C=medium
else
2. if A=high and B=low then C=high
else
3.
else
4.
..
else
n.
The fuzzy process is as follows according to me:(Please correct me if I 'm
wrong)
- The rules have the following form:
1. (if A=low then C=medium) and (if B=high then C=medium)
- I use the relation method to obtain a relation between an input and an
output, thus for rule 1 I obtain two relations, namely a relation for
if A then C and for if B then C. I now have 2 relation matrices per rule.
- Now I AND these two relation matrices with the intersection method to
obtain a final relationship matrix for rule 1 that relates both A and B to C.
- Now I use the min operator on all the final relationship matrices( one for
each rule), to obtain a final resultant relationship matrix for the system.
Now my problem is to obtain the control set if I have the input values for A
and B. I can get it from the two relationship matrices per rule, but what do
I do with it? I need to know if I can obtain a control set from the
resultant relationship matrix?
Am I reasoning wrong or not?
If anynone can refer me to an article that describes an algorithm for these
type of rules, please let me know.
Thanks
Deon Roelofse
University of Pretoria
South Africa
e-mail: s9110305@mella.ee.up.ac.za
------------------------------