Fuzzy Systems (EEM-724),

(PEE-112)

Dr. Subho Upadhyay

Suggested Book: T Ross: FUZZY LOGIC WITH ENGINEERING

APPLICATIONS
Apple Red in colour
 Bivalent System Multivalent System

In Boolean logic the In Boolean logic the truth

result will be either true or false or a degree of truth
or false (Degree of membership)

Fuzzy logic deals with

1) Partial i.e., a matter of degree
2) Imprecise (approximate) information
3) Granular (linguistic) information
4) Perception based on information
 Complexity Versus Precession

The closer one looks at a real world problem, the fuzzier becomes its solution”
Dr. L.A. Zadeh 1973
 Fuzzy Logic

As the complexity of a system increases, it becomes

more difficult and eventually impossible to make
precise statement about its behavior, eventually
arriving at a point of complexity where the fuzzy
logic method born humans is the only way to get at
the problem”

By Professor L.A. Zadeh

 Where we should use Fuzzy Logic

• Systems with uncertainties due to imprecision,

vagueness, ambiguous, partial truth and
approximation
• Black box or grey box systems (a part of
mathematical modeling).

Do not use Fuzzy Logic

• Mathematical modeling is known ( white box
modeling)
• Linear system
• System with moderate complexity and non linearity
 Fuzzy Logic Control and Decision Making

Input from the real world

measurements or Crisp output to be used
assessment of system as control signal or
conditions like decision
temperature, market data Fuzzy to Crisp output
Crisp to Fuzzy data (Defuzzification)
(Fuzzification)

Fuzzy Rule Base

Fuzzifier Fuzzy Inference Engine Defuzzifier

 Fuzzy Logic vs. Probability

• Operate over the same numeric range i.e, between 0

and 1.
• The semantic difference is significant
Probability  Degree of randomness
Eg. Head and Tail
6/10, 4/10
Fuzzy Logic  Degree of belongingness
Eg. Height of a person
Tall, Medium, Short
 Course Outcomes

1. Illustrate fuzzy concepts systems with understanding of basic mathematical elements of fuzzy set theory for representation of fuzzy systems.
2. Relate fuzzy sets and classical sets theories to understand the differences and similarities between them.
3. Build concepts of fuzzy logic and approximate reasoning by extending classical logic concepts (its operations, propositions, logical implications and certain classical inference mechanisms) for fuzzy rule based systems using natural language as expression of a knowledge form.
4. Explain methods of fuzzification, inference techniques for fuzzy systems based on linguistic rules and methods of defuzzification in order to provide an insight into fuzzy inference applications in the area of control.
5. Identify and analyze exciting applications of fuzzy knowledge processing in various domains such as fuzzy control system design by combining some of the traditional design approaches (P controller, PI controller, PD controller and PID controller) with fuzzy concepts, multi-objective decision
making, system identification, modeling and simulation.
Artificial Intelligence (Nature inspired Intelligence)

It is discipline which involves all sort of mechanisms/

algorithms that deal with mimicking the activities of
natural processes.

Eg. Movement of Swarm, Human Neural Network, Ant colony, Human

intelligence etc.
 Historical development of Fuzzy Logic
1936/37- Allen’s Universal Turing Machine was proposed

1942/43- Warren McCulloch and Walter Pitts created a

computational model for neural networks called threshold logic.

1950- Turing test was proposed

1955- John McCarthy Coined the term AI (Artificial Intelligence)

1957- Perceptron model was introduced

1960- Genetic Algorithm

1965- Fuzzy logic/ Deep learning

1970s- Evolutionary Computing

1980s- Neural Computing, Swarm intelligence

1990s- Hybrid Models, Neuro–Fuzzy Systems, Neuro-Fuzzy-

genetic etc.

2000s- Research Areas

(a) Adaptive Systems (b) Evolutionary Computing
(c) Data mining (d) Particle Swarm Optimization
(e) Simulated Annealing (f) Deep Neural Network
(g) Deep Fuzzy Network
 Applications of Fuzzy Systems
• Hitachi Turned over control of subway (Metro) Sendai, Japan to fuzzy system
(1988)
• Fuzzy auto controllers in Cars:- Nissan patented a fuzzy automatic transmission that
saves fuel 12 to 17 %.
• Fuzzy logic in automatic gear selection:- 1) Road conditions, 2) Driving Style
• Washing Machines:- Select 1) water supply-> Cold or Hot, 2) water level-> High,
Medium, Low, 3) Function-> Soap, Wash, Spin, 4) Course-> Wool, Blanket, Speedy
etc.
• Elevator monitoring and control
• Fuzzy in photo copy Machines:- Adjusted drum voltage based on picture density,
humidity and temperature.
• Fuzzy logic for fast decision making in Courts
• Fuzzy logic in Psychology ( Fuzzy logic based human Behaviour)
• Fuzzy logic in image processing-> 1) Contrast Enhancement, 2) Classification, 3)
Edge Detection, 4) Segmentation, 5) Filtering
• Fuzzy logic in Aerospace-> Altitude control of spacecraft
• Fuzzy logic based air conditioning system-> 1) Room temperature control, 2) Relative
humidity control, (Results->a) Efficient energy operation, b) comfortable environment
• Fuzzy logic based coronary artery disease-> Age, Gender, Cholesterol level, Obesity,
Smoking etc.
• Fuzzy logic in agriculture-> Soil moisture, other environmental conditions.
• Fuzzy logic for condition monitoring of machine-> Fault classification, Fault
detection, Remaining Life, Prediction

Journals
IEEE Transactions of fuzzy systems,
Fuzzy Sets and Systems Elsevier