You are on page 1of 25

1

Fuzzy Logic for Most of US

Norm Dingle – Senior Systems Engineer


EMP Technical Group – Noblesville, IN

2
What is Fuzzy Logic?

• Wikipedia definition: “Fuzzy logic is a form of


many-valued logic; it deals with reasoning that is
fixed or approximate rather than fixed and exact.
In contrast with traditional logic theory, where
binary sets have two-valued logic: true or false,
fuzzy logic variables may have a truth value that
ranges in degree between 0 and 1”.

3
What is Fuzzy Logic?

• Rule based system of IF X THEN Y


• Relies on the operator's experience

4
How Did it Get Started

• Prof. Lotfi Zedeh University of California 1965


• Prof. E. H. Mamdani London University

5
Where is it Used?

• Cameras
• Washing machines
• Directional drilling control
• Barcode scanners
• Medical diagnosis systems
• Stock portfolio management

6
Why use it

• Tolerates imprecise inputs


• Easily modified with rules
• Not limited to a few inputs or outputs
• Useful in non-linear systems
• Blendable with conventional control systems

7
When NOT to Use it

• Not a “silver bullet”


• Simpler solution already exists
• Not convenient
• Not solvable

8
Fuzzy Logic Terminology

• Fuzzy Sets
• Linguistic Variables
• Membership functions
• “Defuzzify”

9
Non Fuzzy Set

10
Fuzzy Set

11
Fuzzy Logic Method

• Receive Inputs
• Process these inputs
• Average and weight the result

12
Fuzzy Logic build steps

• Determine Inputs
• Build membership functions
• Set up Fuzzy rules
• “Defuzzify” to produce “crisp” output

13
Inputs

• Current Temperature
• Motor Speed

14
Input Membership Function

15
Output Membership Function

16
Rules

• If COLD THEN motor speed SLOW


• If NORMAL THEN motor speed MEDIUM
• IF HOT THEN motor speed FAST

17
“Defuzzification”

18
“Defuzzification”

19
“Defuzzification”

20
LabVIEW Tools

• PID and Fuzzy Logic Toolkit


• Fuzzy System Designer
• Fuzzy Logic VI’s

21
Using the Fuzzy System Designer

22
Demo

23
Further information

• PID & Fuzzy Logic Toolkit user manual.


• Fuzzy Thinking - Bart Kosko
• Fuzzy Engineering - Bart Kosko
• Web

24
25

You might also like