Output value problem

[Unknown]

Dear Juan,
I am using qtfuzzylite 5.0 version and it produces 0.844 ,0.844 and 0.72 as final output values for all YES input values when apply rules.But it should be 1.0,1.0 and 1.0 for all YES inputs.It is greatful if you could mention the correct way of obtaining the output.
Thank you

[Unknown]

This is my reseacch application and how can I send my FLL file?

[Juan Rada-Vilela (admin)]

Hi,

thanks for your post.

Please, copy and paste your FLL code here to take a look at.

Cheers.

[Unknown]

Engine:
InputVariable: Q1
enabled: true
range: 0.000 1.000
term: NO Trapezoid 0.000 0.000 0.200 0.400
term: AVERAGE Trapezoid 0.200 0.400 0.600 0.800
term: YES Trapezoid 0.600 0.800 1.000 1.000
OutputVariable: OUT
enabled: true
range: 0.000 1.000
accumulation: Maximum
defuzzifier: Centroid 200
default: nan
lock-previous: false
lock-range: false
term: POOR Trapezoid 0.000 0.000 0.200 0.400
term: AVERAGE Trapezoid 0.200 0.400 0.600 0.800
term: GOOD Trapezoid 0.600 0.800 1.000 1.000
RuleBlock:
enabled: true
conjunction: Minimum
disjunction: Maximum
activation: Minimum
rule: if Q1 is YES then OUT is GOOD
rule: if Q1 is AVERAGE then OUT is AVERAGE
rule: if Q1 is NO then OUT is POOR`

[Unknown]

when Q1 is YES (1.00) it generates 0.844 only not 1.00

[Juan Rada-Vilela (admin)]

Hi,

the result you obtain is the correct result. I understand your expectations, but the Centroid defuzzifier computes the centroid of the fuzzy output, which in in this case is at 0.844. If you want to obtain 1.0, you need to move the centroid of your linguistic term GOOD. For example,

OutputVariable: OUT
  enabled: true
  range: 0.000 1.400
  lock-range: false
  aggregation: Maximum
  defuzzifier: Centroid 100
  default: nan
  lock-previous: false
  term: POOR Trapezoid 0.000 0.000 0.200 0.400
  term: AVERAGE Trapezoid 0.200 0.400 0.600 0.800
  term: GOOD Trapezoid 0.600 0.800 1.200 1.400 #Extend centroid to 1.0

This can be better seen using QtFuzzyLite:

[Unknown]

THANK YOU VERY MUCH

[Author]

Posts