Scribd
Upload
11 views

Fuzzy Logic 2

Uploaded by

jaguar721302
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
11 views

Fuzzy Logic 2

Uploaded by

jaguar721302
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 49

Topic

Applications of Fuzzy Sets


Applications of Fuzzy Sets

•Reasoning tool like Fuzzy Logic Controller

•Fuzzy Clustering

•Fuzzy Mathematical Programming

•Fuzzy Graph theory

•Others
Fuzzy Logic Controller
(FLC)
•Developed by Mamdani and Assilian around 1975

•Performance depends on Knowledge Base (KB), which consists of


Data Base (DB) and Rule Base (RB)

•The performance of an FLC largely depends on the RB and


optimizing the DB is a fine tuning process
Two Forms of FLC

•Linguistic Fuzzy Modeling: High interpretability and Low


accuracy.

Ex: Mamdani Approach

•Precise Fuzzy Modeling: Low interpretability and High


accuracy.

Ex: Takagi and Sugeno’s Approach


Mamdani Approach
•An FLC consists of four modules: Rule Base; Inference Engine;
Fuzzification; De-fuzzification
Steps Involved in the Working Cycle of
FLC
•Identify the condition ( also known as antecedent ) and action (also called
consequent) variables of the process

•Fuzzification: Measurements of input variables are converted into


appropriate fuzzy sets to express measurement uncertainties
•Inference Engine uses fuzzified measurements to evaluate the control
rules stored in the fuzzy rule base

•De-fuzzification: Fuzzified output is then converted into a single crisp


value
Fuzzy Reasoning Process
Rule1: IF S1 is A1 AND S2 is B1 THEN f’ is C1
Rule2: IF S1 is A2 AND S2 is B2 THEN f’ is C2
Let S1* and S2* are the inputs for fuzzy variables S1 and S2

Firing strengths of the first and second rules can be calculated as


    
follows:  min  S * ,  S *
1 A1 1 B1 2

2 min  S ,  S 
A2
*
1 B2
*
2
Membership value of the combined control action C is given by

 
C  f ' max C* 1  f ', C* 2  f '
De-fuzzification
•Methods
Center of Sums Method
p

 A   f
j 1
j j

U 'f '  p
,
 A  
j 1
j

where : firing area of -th rule

: center of the area

: No. of fired rules


Methods of De-fuzzificztion
• Centroid Method

 A f i j
U ' f '  i 1 N ,
A
i 1
i

where : No. of small areas or regions

: Area of -th small region

: Center of area of -th small region


• Mean of Maxima Method

From the membership function distribution of combined control action,


the range of the output variable corresponding to the maximum value of
membership is identified. The mid-value of this range is considered to be
the crisp output of the controller.
Numerical example (Mamdani Approach)
Let us consider a problem scenario related to navigation of a mobile
robot in the presence of four moving obstacles. Obstacle 2 is found to be
the most critical one. Our aim is to develop a fuzzy logic-based motion
planner that will be able to generate the collision-free path for the
robot. There are two inputs, namely the distance between the robot and
the obstacle (D) and angle () for the motion planner and it will generate
one output, that is, deviation. Distance is represented using four
linguistic terms, namely Very Near (VN), Near(NR), Far(FR) and Very Far
(VFR), whereas the input: angle and output: deviation are expressed
with the help of five linguistic terms, such as Ahead (A), Ahead Left (AL),
Left (LT) Ahead Right(ART) and Right (RT).
The data base (DB) and rule base (RB) of the FLC are shown below.
D=1.04 m and angle = , using Mamdani Approach. Use
Determine the output: deviation for the set of inputs: distance

different methods of defuzzification.


Solution:
Inputs: Distance = 1.04 m; Angle =
Distance of 1.04 m may be either NR or FR, Angle of
To determine the membership, corresponding to the distance = 1.04 m.
Distance of 1.04 m may be called NR with and FR with
Similarly, Angle of may be declared A with and ART with

Fired rules are as follows:

If Distance is NR AND Angle is A Then Deviation is RT


If Distance is NR AND Angle is ART Then Deviation is A
If Distance is FR AND Angle is A Then Deviation is ART
If Distance is FR AND Angle is ART Then Deviation is A
Strengths ( VALUES) of the fired rules:

429
Fuzzified outputs:
•Union of the fuzzified outputs:
Methods of Defuzzification
1. Center of Sums
Method
12.5 71.0  39.7089 0.0  25.0 45.0  25.5699 0.0
U
12.5  39.7089  25.0  25.5699

19.5809
2. Centroid
Method
A
U
B

A 9.7151  25.2860 20.2788 0.0  2.3588 20.2870 


24.8540 52.7153

B 9.7151  20.2788  2.3588  24.8540

 U 19.4450
3. Mean of
Maxima

U 0
Takagi and Sugeno’s Approach

•A rule is composed of fuzzy antecedent and functional


consequent parts

• i-th rule can be represented as follows:

If x1is Ai1 and x2 is Ai2 ….. and xn is Ain then

yi=ai0+ai1 x1+……….+ain xn
where a0, a1, …….., an are the coefficients

• A non-linear system is considered to be a combination of


several linear systems

•Weight of i-th rule can be determined as follows:

wi = μiA1(x1) μiA2(x2)……… μiAn(xn)


• Combined control action can be determined as follows:

k i i
 w y
y  i 1k
i
 w
i 1

where k indicates the total number of rules.


Numerical example (Takagi
and Sugeno’s Approach
•A fuzzy logic-based expert system is to be developed that
will work based on Takagi and Sugeno’s approach to
predict the output of a process. The DB of the FLC is
shown below. As there are two inputs: and and each
input is represented using three linguistic terms (for
example, LW, M, H for and NR, FR, VFR for ) there is a
maximum of 3 x 3 = 9 feasible solutions. The output of
the i-th rule, that is, i 1,2is,...,9
expressed as follows:
y i  f I 1 , I 2  a ij I 1  bki I 2 ,

i i i
•Where j , k 1,2,3; a1 1, a 2 2, and a 3 3,
If is found to be LW, M and H, respectively;
b1i 1, b2i 2, and, b3i 3, if is seen to be NR,
FR and VFR, respectively. Calculate the output of the
FLC for the inputs: =6.0, =2.2
Solution:

•Inputs: =6.0, = 2.2

•Inputs: of 6.0 unit may be called LW or M.

•Inputs: of 2.2 unit may be called FR or VFR.

•Determination of membership value:


•Using the principle of similar triangle

x 10  6

1.0 10  5

x 0.8
Input of 6.0 may be called LW with μLW =0.8

Input of 6.0 may be called M with μM =0.2

Similarly, Input of 2.2 may be called FR with μFR =0.8

Input of 2.2 may be called VFR with μVFR =0.2


Fired Set of Inputs

is LW and is FR

is LW and is VFR

is M and is FR

is M and is VFR
Weights
Functional consequent values

6
Output,
Hierarchical Fuzzy Logic Controller
Let us assume that a process having n inputs (I1,I2,……In) and
one output (that is, O) is to be modeled using an FLC
Let us also assume that m linguistic terms (fuzzy sets) are used to represent
each variable.

Total no. of rules = mn

Problem : curse of dimensionality


To overcome this problem, we go for Hierarchical Fuzzy Logic Controller

Total no. of rules = m2(n-1)


Sensitivity Analysis
•To determine the importance of input variable towards the output
Sensitivity

Methods to Determine Sensitivity

1.We carry out experiment by varying the input variables through different
amounts, say 0.1%, 1.0%, 10.0% etc. and recording the outputs of the
controller
2.O = f(I1, I2)

We change the input I1 by a small amount δI1 and determine the change
in output using Mean-Value theorem

f
O  f I1  I1 , I 2  f I1 , I 2   I 1
I1 I1 , I 2
Where I1 lies between I1 and I1+δI1

f
 1,
I1
If then δO > δI1

Thus, the contribution of I1 on the output O


can be determined.
Advantages of FLC

•Potential tool for dealing with imprecision and uncertainty

•Does not require an extensive mathematical formulation of the


problem
Disadvantages of FLC

•Designing a proper KB of an FLC is a difficult task

•May not be suitable for modeling a process involving many variables

You might also like