1) Compare Mamdani and Sugeno FIS

2) What are hybrid system? Explain its advantages with example

A hybrid system combines two or more different technologies to create a whole system that
offers the advantages of each individual technology. Here are some key benefits of hybrid
systems, along with examples:
1. Increased Efficiency and Performance:
 Example: Hybrid Cars: A hybrid car combines a gasoline engine with an electric motor. The
electric motor can provide power during low-speed driving or act as a regenera ve brake to
capture energy when stopping, improving fuel efficiency compared to tradi onal gasoline
cars.
2. Enhanced Reliability and Backup Power:
 Example: Hybrid Solar Power System: This system combines solar panels with a ba ery bank
and connec on to the u lity grid. During the day, solar panels generate electricity. Excess
energy can be stored in the ba eries for use at night or during power outages, providing
backup power and increased reliability compared to just solar panels.
3. Improved Sustainability and Reduced Emissions:
 Example: Hybrid Hea ng System: This system might combine a geothermal heat pump with
a natural gas furnace. The geothermal heat pump uses renewable geothermal energy for
efficient hea ng and cooling most of the year. The natural gas furnace acts as a backup for
extremely cold weather, reducing reliance on fossil fuels compared to a tradi onal gas
furnace.
4. Greater Flexibility and Adaptability:
 Example: Hybrid Manufacturing System: This system might combine human workers with
robots for specific tasks. Humans can handle complex or variable tasks, while robots excel at
repe ve tasks with high precision. This combina on allows for greater flexibility and
adaptability in the produc on process compared to a fully automated or manual system.

3) What are Fuzzy membership func on. Give two techniques by which
fuzzy membership can be assigned
A fuzzy membership func on is a mathema cal func on used in fuzzy logic to represent the
degree of membership of an element in a fuzzy set. In a fuzzy set, an element can belong to
 the set with a degree of membership ranging between 0 and 1 where O means no
membership and 1 means full membership.
Two common techniques for assigning membership values are:
1. Triangular Membership Func on: This is a simple and commonly used method
where the membership func on is represented by a triangle. It has three
parameters: the lower limit (a), the upper limit (b), and the peak (c). The degree of
membership increases linearly from a to c, reaches its maximum value of 1 at c, and
then decreases linearly from c to b. This func on is o en used when there is a clear
peak value of membership.
2. Gaussian Membership Func on: This func on is based on the Gaussian distribu on
curve. It is defined by its mean (u) and standard devia on (o). The shape of the
Gaussian func on is bell-shaped, with the highest membership value at the mean,
and it decreases symmetrically on both sides. Gaussian func ons are par cularly
useful when dealing with uncertain es or when the distribu on of membership is
expected to follow a normal distribu on.

4) Explain fuzzy rule based system

A fuzzy rule-based system (FRBS) is a type of expert system that uses fuzzy logic to represent and
process knowledge in a form similar to human reasoning. It consists of a set of fuzzy rules, a fuzzifier,
an inference engine, a defuzzifier, and a database of membership func ons.

Here's how it generally works

1. Fuzzifier: The fuzzifier converts crisp input data (non-fuzzy data) into fuzzy sets by assigning
membership degrees to each element of the input data based on predefined membership func ons.
This step allows the system to handle imprecise or uncertain inputs.

2. Fuzzy Rule Base: The heart of the fuzzy rule-based system is the fuzzy rule base, which consists of
a set of IF-THEN rules. Each rule describes a rela onship between the fuzzy input variables and the
fuzzy output variables. The antecedent (IF part of the rule contains fuzzy sets defined over the input
variables, and the consequent (THEN part) contains fuzzy sets defined over the output variables.

3. Inference Engine: The inference engine evaluates the fuzzy rules using fuzzy logic opera ons (such
as fuzzy AND, fuzzy OR, and fuzzy implica on) to derive fuzzy outputs based on the fuzzy inputs. It
combines the informa on from mul ple rules to generate fuzzy output sets

4. Aggrega on. In this step, the outputs from all the ac vated rules are aggregated to produce a
single fuzzy output set that represents the overall output of the system. This aggrega on process can
involve fuzzy AND or fuzzy OR opera ons, depending on the system's design.

5. Defuzzifier. The defuzzifier converts the aggregated fuzzy output set into a crisp output value by
calcula ng the centroid or other appropriate method. This process provides a single, interpretable
output value from the fuzzy output set.

6. Decision Making: Finally, the crisp output value obtained from the defuzzifier is used to make
decisions or control ac ons based on the applica on requirements.
6) What is defuzzifica on. 3 techniques of defuzzifica on
Defuzzification is the process of conversion of fuzzy quantity into a crisp quantity.

Defuzzifica on methods include:

1 Centroid method

2 Center of sums.

3 Mean of Maxima

4 Smallest of Maxima

5 Largest of Maxima

6 Bisector of Area

The centroid method of defuzzification is a widely used technique in fuzzy logic

systems to convert fuzzy output sets into crisp values. It calculates the center of
gravity of the fuzzy output set to determine the crisp output value.

Here's how the centroid method works with an example:

Let's consider a fuzzy temperature control system where the linguistic variable
"temperature" is represented by three fuzzy sets: "cold," "warm," and "hot." Each
fuzzy set has a membership function that represents the degree of membership of
the current temperature in that set.

1. Fuzzy Output Sets: Suppose after processing input data, the fuzzy output
sets for adjusting the air conditioner are as follows:

 Cold: Membership = 0.2

 Warm: Membership = 0.7
 Hot: Membership = 0.5
 2. Defuzzification: To obtain a crisp output value from these fuzzy sets, we'll use
the centroid method.

The centroid method calculates the center of gravity of the fuzzy output set
using the following formula:

Crisp Output=∑(Membership×Output Value)∑MembershipCrisp Outp

ut=∑Membership∑(Membership×Output Value)

 For the "Cold" set:

 Membership = 0.2
 Output Value (representative value for "Cold," e.g., 10°C)
 0.2×10=20.2×10=2

 For the "Warm" set:

 Membership = 0.7
 Output Value (representative value for "Warm," e.g., 25°C)
 0.7×25=17.50.7×25=17.5

 For the "Hot" set:

 Membership = 0.5
 Output Value (representative value for "Hot," e.g., 35°C)
 0.5×35=17.50.5×35=17.5

 Calculate Crisp Output:

Crisp Output=(2+17.5+17.5)/(0.2+0.7+0.5)
Crisp Output=371.4≈26.43Crisp Output=1.437≈26.43

CENTRE OF SUMS
It calculates the center of area under the fuzzy output set curve, weighted by the degree of
membership of each fuzzy set element

Suppose we have the following fuzzy output set for temperature control:

 Linguis c term "cold" with a triangular membership func on centered at 10°C and spanning
from 0°C to 20°C.
  Linguis c term "cool" with a triangular membership func on centered at 25°C and spanning
from 15°C to 35°C.

 Linguis c term "warm" with a triangular membership func on centered at 35°C and
spanning from 25°C to 45°C.

 Linguis c term "hot" with a triangular membership func on centered at 50°C and spanning
from 40°C to 60°C.

Let's say the membership values for each linguis c term are:

 "cold": 0.2

 "cool": 0.7

 "warm": 0.8

 "hot": 0.4

Using the CoS method, we calculate the center of area for each linguis c term:

 Center of area for "cold": (0+20+0)3=6.673(0+20+0)=6.67

 Center of area for "cool": (15+25+35)3=253(15+25+35)=25

 Center of area for "warm": (25+35+45)3=353(25+35+45)=35

 Center of area for "hot": (40+50+60)3=503(40+50+60)=50

Now, we calculate the weighted sum:

(0.2×6.67)+(0.7×25)+(0.8×35)+(0.4×50)(0.2×6.67)+(0.7×25)+(0.8×35)+(0.4×50)

=(1.334)+(17.5)+(28)+(20)=(1.334)+(17.5)+(28)+(20)

=66.834=66.834

Next, we normalize this value by dividing by the total sum of membership values:

Total sum of membership values=0.2+0.7+0.8+0.4=2.1Total sum of membership values=0.2+0.7+0.8+

0.4=2.1

Normalized value=66.8342.1=31.82Normalized value=2.166.834=31.82

MEAN OF MAXIMA
This method calculates the centroid of the highest membership values in the fuzzy output.

Consider a fuzzy logic system for controlling the speed of a ceiling fan based on the room
temperature. The linguis c variable "temperature" is defined with three fuzzy sets: "cold," "warm,"
and "hot." We have the following membership func ons:

 "Cold" has a triangular membership func on with points at (0,1), (10,0), and (20,0).

 "Warm" has a triangular membership func on with points at (10,0), (20,1), and (30,0).

 "Hot" has a triangular membership func on with points at (20,0), (30,1), and (40,0).
Suppose the fuzzy output of the system, a er applying fuzzy inference, is as follows:

 "Cold": 0.2

 "Warm": 0.8

 "Hot": 0.4

To find the crisp output using the Mean of Maxima method:

1. Iden fy the fuzzy set(s) with the highest membership value: "Warm" (0.8).

2. Calculate the centroid of the membership func on of "Warm":

𝐶𝑒𝑛𝑡𝑟𝑜𝑖𝑑=(20+30)2=25Centroid=2(20+30)=25

SMALLEST OF MAXIMA
This method selects the smallest value among the maxima of the aggregated fuzzy membership
func ons.

Here's how the Smallest of Maxima method works with an example:

Let's consider a fuzzy system that controls the speed of a fan based on the temperature input. We
have three fuzzy sets for the temperature variable: Cold, Warm, and Hot. Each fuzzy set has a
membership func on represen ng its degree of membership.

 Cold: μCold(x)

 Warm: μWarm(x)

 Hot: μHot(x)

Suppose the aggregated fuzzy membership func ons a er fuzzifica on are as follows:

 μCold(x) = [0, 0.2, 0.5, 0.7, 0]

 μWarm(x) = [0, 0.3, 0.6, 0.9, 0.3]

 μHot(x) = [0.2, 0.6, 0.9, 1, 0.7]

To defuzzify using the Smallest of Maxima method, we first find the maximum membership value for
each crisp output value. Then, we select the smallest value among these maxima as the final crisp
output.

Let's calculate the maximum membership values:

 For x = 0: Max(μCold) = 0, Max(μWarm) = 0, Max(μHot) = 0.2

 For x = 1: Max(μCold) = 0.2, Max(μWarm) = 0.3, Max(μHot) = 0.6

 For x = 2: Max(μCold) = 0.5, Max(μWarm) = 0.6, Max(μHot) = 0.9

 For x = 3: Max(μCold) = 0.7, Max(μWarm) = 0.9, Max(μHot) = 1

 For x = 4: Max(μCold) = 0, Max(μWarm) = 0.3, Max(μHot) = 0.7

Now, we select the smallest value among these maxima:

 For x = 0: Smallest(Max(μCold), Max(μWarm), Max(μHot)) = Smallest(0, 0, 0.2) = 0

  For x = 1: Smallest(Max(μCold), Max(μWarm), Max(μHot)) = Smallest(0.2, 0.3, 0.6) = 0.2

 For x = 2: Smallest(Max(μCold), Max(μWarm), Max(μHot)) = Smallest(0.5, 0.6, 0.9) = 0.5

 For x = 3: Smallest(Max(μCold), Max(μWarm), Max(μHot)) = Smallest(0.7, 0.9, 1) = 0.7

 For x = 4: Smallest(Max(μCold), Max(μWarm), Max(μHot)) = Smallest(0, 0.3, 0.7) = 0

 Therefore, the crisp output value obtained using the Smallest of Maxima method is 0 when
the input temperature is 0, and it gradually increases to 0.7 when the temperature reaches
3, then decreases back to 0 as the temperature rises further.

LARGEST OF MAXIMA
This method selects the output value with the highest degree of membership from the aggregated
fuzzy output set.

In our example, the degree of membership values are:

 Cool: 0.8

 Warm: 0.9

 Hot: 0.7

Since "Warm" has the highest degree of membership (0.9), the corresponding crisp value for the
control signal is selected. Therefore, the crisp output value for the temperature control system using
the Largest of Maxima method is 25°C, indica ng that the system should be set to a warm
temperature.

7) Different fuzzy membership func ons

8) How are the membership value assigned