Fuzzy logic : principles and applications

Dr. Ing. Dhaker ABBES

Professor-Researcher
Co-responsable of ESEA field :
Energy, Electrical and Automated Systems
Member of Electric Networks Team (L2EP)
Graduate School of Engineering, Lille
(HEI-Lille, France)

30/09/2019 Dhaker ABBES 1

 Fuzzy Logic
What is Fuzzy Logic?
A computational paradigm that is based on how humans think

Fuzzy Logic looks at the world in imprecise terms, in much the same way
that our brain takes in information (e.g. temperature is hot, speed is slow),
then responds with precise actions.

The human brain can reason with uncertainties, vagueness, and

judgments. Computers can only manipulate precise valuations. Fuzzy logic
is an attempt to combine the two techniques.

“Fuzzy” – a misnomer, has resulted in the mistaken suspicion that FL is

somehow less exact than traditional logic

30/09/2019 Dhaker ABBES 2

2
 Fuzzy logic
What is Fuzzy Logic?

FL is in fact, a precise problem-solving methodology.

It is able to simultaneously handle numerical data and linguistic

knowledge.

A technique that facilitates the control of a complicated system without

knowledge of its mathematical description.

Fuzzy logic differs from classical logic in that statements are no longer
black or white, true or false, on or off.
In traditional logic an object takes on a value of either zero or one.
In fuzzy logic, a statement can assume any real value between 0 and 1,
representing the degree to which an element belongs to a given set.

30/09/2019 Dhaker ABBES 3

 Fuzzy Logic
History of Fuzzy Logic

Professor Lotfi A. Zadeh

http://www.cs.berkeley.edu/~zadeh/

In 1965, Lotfi A. Zadeh of the University of California at Berkeley published

"Fuzzy Sets," which laid out the mathematics of fuzzy set theory and, by
extension, fuzzy logic. Zadeh had observed that conventional computer logic
couldn't manipulate data that represented subjective or vague ideas, so he created
fuzzy logic to allow computers to determine the distinctions among data with
shades of gray, similar to the process of human reasoning.

Source: August 30, 2004 http://www.computerworld.com/news/2004/story/0,11280,95282,00.html

(Computerworld)

30/09/2019 Dhaker ABBES 4

4
 Task 1
1) What is the principle of fuzzy logic? Explain the distinction between it and classical logic.
Fuzzy logic is a concept of ‘certain degree’. Boolean logic is a subset of fuzzy logic.
Fuzzy logic is a form of many-valued logic which deals with reasoning that is approximate
rather than fixed and exact. Compared to traditional binary sets (where variables may take on
true or false values), fuzzy logic variables may have a truth value that ranges in degree
between 0 and 1.
2) What is the main difference between the probability and fuzzy logic?
Probability is ADDITIVE, means all its values must add up to one. This is main difference
between fuzzy logic and probability. Although both probability and fuzzy logic contain values
between the range of 1 and 0, fuzzy logic tells the extent of a specific member function,
whereas probability gives the frequency, hence all values of its set must add up to one.

30/09/2019 Dhaker ABBES 5

 Task 1
3) What is the reasons that fuzzy logic has rapidly become one of the
most successful technology for developing sophisticated control
systems?
Fuzzy logic is conceptually easy to understand.
Fuzzy logic is flexible.
Fuzzy logic is tolerant of imprecise data.
Fuzzy logic can model nonlinear functions of arbitrary
complexity.
Fuzzy logic can be built on top of the experience of experts.
Fuzzy logic can be blended with conventional control techniques.
Fuzzy logic is based on natural language.
30/09/2019 Dhaker ABBES 6
 Task 1
4) When not to use fuzzy logic?
When a simpler solution already exists.

5) Give at least 3 famous applications of fuzzy logic.

 Nissan – fuzzy automatic transmission, fuzzy anti-skid braking
system
 CSK, Hitachi – Hand-writing Recognition
 Sony - Hand-printed character recognition
 Ricoh, Hitachi – Voice recognition.

30/09/2019 Dhaker ABBES 7

 Fuzzy Logic Explained
Fuzzy Set Theory

Is a man whose height is 180 cm average or tall?

A fuzzy system might say that he is partly medium and partly tall.

Boolean representation Fuzzy representation

http://blog.peltarion.com/2006/10/25/fuzzy-math-part-1-the-theory/

In fuzzy terms, the height of the man would be

classified within a range of [0, 1] as average to a
degree of 0.6, and tall to a degree of 0.4.
 Fuzzy Inference Process

• What are the steps involved in creating a

Fuzzy Control System?

30/09/2019 Dhaker ABBES 9

 Fuzzy Inference Process

Fuzzy Inference Process

e.g. theta e.g. force

Fuzzification Rule Defuzzification

Evaluation

Fuzzification: Translate input into truth values

Rule Evaluation: Compute output truth values
Defuzzification: Transfer truth values into output

30/09/2019 Dhaker ABBES 10

 Example 1 : Obstacle Avoidance Problem
Robot Navigation

obstacle
(obsx, obsy)

(x,y)

Can you describe how the robot

should turn based on the position
Obstacle Avoidance & Target Pursuit
and angle of the obstacle?
Demonstration

menu
Another example: Fuzzy Sets for Robot Navigation

Angle and Distance

SMALL
MEDIUM

LARGE

NEAR

FAR

VERY FAR

* Sub ranges for angles & distances overlap

 Fuzzy Systems for Obstacle Avoidance
Vision System
Nearest Obstacle (Distance and Angle)
Fuzzy System 3 (Steering)
NEAR FAR VERY FAR
e.g. If the Distance from the Obstacle is NEAR and
SMALL Very Sharp Sharp Turn Med Turn
the Angle from the Obstacle is SMALL
Then turn Very Sharply.
MEDIUM Sharp Turn Med Turn Mild Turn

LARGE Med Turn Mild Turn Zero Turn

Angle

Fuzzy System 4 (Speed Adjustment)

NEAR FAR VERY FAR
e.g. If the Distance from the Obstacle is NEAR and
the Angle from the Obstacle is SMALL
SMALL Very Slow Slow Speed Fast Fast
Then move Very Slowly.
MEDIUM Slow Speed Fast Speed Very Fast

LARGE Fast Speed Very Fast Top Speed

Speed
 Summary of Steps
1. determining a set of fuzzy rules
2. fuzzifying the inputs using the input membership functions,
3. combining the fuzzified inputs according to the fuzzy rules to
establish a rule strength,
4. finding the consequence of the rule by combining the rule
strength and the output membership function (if it’s a
mamdani FIS),
5. combining the consequences to get an output distribution,
and
6. defuzzifying the output distribution (this step applies only if a
crisp output (class) is needed).

30/09/2019 Dhaker ABBES 11

 Fuzzy Inference

Fuzzy inference is the process of formulating the mapping

from a given input to an output using fuzzy logic. The
mapping then provides a basis from which decisions can be
made, or patterns discerned. The process of fuzzy inference
involves : Membership Functions, Logical Operations, and If-
Then Rules.

30/09/2019 Dhaker ABBES 12

 Mamdani Inference System
Two input, two rule Mamdani FIS with crisp inputs

IF Rule THEN
Strength

and

xo yo

and

Input Distribution Output

Distribution

Xo Yo

Fuzzy rules are a collection of linguistic statements that describe how the FIS
should make a decision regarding classifying an input or controlling an output.
30/09/2019 Dhaker ABBES 13
 Mamdani FIS
Mamdani-type inference, expects the output membership functions to be fuzzy
sets. After the aggregation process, there is a fuzzy set for each output variable that
needs defuzzification.

http://www.mathworks.com/access/helpdesk/help/toolbox/fuzzy/index.html?/access/helpdesk/help/toolbox/fuzzy/bp78l6_-
1.html&http://www.mathworks.com/cgi-
bin/texis/webinator/search/?db=MSS&prox=page&rorder=750&rprox=750&rdfreq=500&rwfreq=500&rlead=250&sufs=0&order=r&is_summa
ry_on=1&ResultCount=10&query=fuzzy+logic&submitButtonName=Search

30/09/2019 Dhaker ABBES 14

 Mamdani FIS
Mamdani-type inference, expects the output membership functions to be fuzzy
sets. After the aggregation process, there is a fuzzy set for each output variable that
needs defuzzification.

http://www.mathworks.com/access/helpdesk/help/toolbox/fuzzy/index.html?/access/helpdesk/help/toolbox/fuzzy/bp78l6_-
1.html&http://www.mathworks.com/cgi-
bin/texis/webinator/search/?db=MSS&prox=page&rorder=750&rprox=750&rdfreq=500&rwfreq=500&rlead=250&sufs=0&order=r&is_summa
ry_on=1&ResultCount=10&query=fuzzy+logic&submitButtonName=Search

30/09/2019 Dhaker ABBES 15

 Mamdani FIS

30/09/2019 Dhaker ABBES 16

 Mamdani FIS

30/09/2019 Dhaker ABBES 17

 Flow of Fuzzy Inference

In this figure, the flow proceeds up from the inputs in the lower left, then
across each row, or rule, and then down the rule outputs to finish in the lower
right. This compact flow shows everything at once, from linguistic variable
fuzzification all the way through defuzzification of the aggregate output.
30/09/2019 Dhaker ABBES 18
 Sugeno FIS
Sugeno FIS is similar to the Mamdani method in many respects. The first two parts of
the fuzzy inference process, fuzzifying the inputs and applying the fuzzy operator, are
exactly the same. The main difference between Mamdani and Sugeno is that the
Sugeno output membership functions are either linear or constant.

A typical rule in a Sugeno fuzzy model has the form:

If Input 1 = x and Input 2 = y, then Output is z = ax + by + c

For a zero-order Sugeno model, the output level z is a constant (a=b =0).
30/09/2019 Dhaker ABBES 19
 Sugeno FIS

30/09/2019 Dhaker ABBES 20

 FIS: Sugeno vs. Mamdani
Advantages of the Sugeno Method

It is computationally efficient.
It can be used to model any inference system in
which the output membership functions are either
linear or constant.
It works well with linear techniques (e.g., PID control).
It works well with optimization and adaptive
techniques.
It has guaranteed continuity of the output surface.
It is well suited to mathematical analysis.
Advantages of the Mamdani Method

It is intuitive.
It has widespread acceptance.
It is well suited to human input.

30/09/2019 Dhaker ABBES 21

 Task 2

What is the sequence of steps taken in designing a fuzzy logic

controller?
 Following is the sequence for the designing a fuzzy logic machine:
Fuzzification->Rule Evaluation->Defuzzification
when designing a fuzzy logic, we first have to define the fuzzy sets
and make appropriate member function. The rule evaluation comes
in which matches the sets to its corresponding rules.

30/09/2019 Dhaker ABBES 22

 Task 3 : example of control
The objective of this case study is to perform the speed control of a separately excited
DC motor (figure 1) using fuzzy logic controller (FLC). The controller will be designed
based on the expert knowledge of the system. For the proposed dc motor case, we
recommend 7 fuzzy rules for fuzzy logic controller.

30/09/2019 Dhaker ABBES 23

 Task 3 : example of control
Taking field flux as ∅ and Back EMF Constant as 𝐾𝜑 .
Equation for back emf of motor will be:

𝐸𝑏 = 𝐾𝜑 𝜔

Torque:

𝑇𝑚 = 𝐾𝜑 𝐼𝑎
𝜔 is the angular velocity (speed) and friction in rotor of motor is very small (can be
neglected) so Bm = 0.
Armature Time Constant:

30/09/2019 Dhaker ABBES 24

 Task 3 : example of control
1) Plot the block diagram of separately excited dc motor based on Laplace transformations of
the motor’s armature voltage and balance torque.

Fig.2 Block diagram of separately excited dc motor

30/09/2019 Dhaker ABBES 25
 Task 3 : example of control
2) Define the required fuzzy controller inputs and outputs. Then complete this diagram:

Fig. 3: General structure of the fuzzy controller.

30/09/2019 Dhaker ABBES 26

 Task 3 : example of control
3) Deduce the structure of the fuzzy logic controller with closed loop
(synopsis of all system with fuzzy controller).

Fig.4 Model of system using fuzzy logic controller

30/09/2019

Dhaker ABBES 27
 Task 3 : example of control
4) Represent membership functions for inputs
and output variables.
Input 1 range: [-1 1]
Input 2 range: [- 1 1]
Output range: [-30 30].

30/09/2019 Dhaker ABBES 28

 Task 3 : example of control
4) Represent membership functions for inputs
and output variables.
Input 1 range: [-1 1]
Input 2 range: [- 1 1]
Output range: [-30 30].

30/09/2019 Dhaker ABBES 29

 Task 3 : example of control
5) Enunciate the 7 “if-then” rules necessary for separately excited dc motor speed control.

- If (ERROR is PB) then (CONTROL is AUGMENTER_BEAUCOUP)

- If (ERROR is NB) then (CONTROL is DIMINUER_BEAUCOUP)
- If (ERROR is Z) and (VARIATION is NEGATIVE) then (CONTROL is
DIMUNIER_PEU)
- If (ERROR is Z) and (VARIATION is POSITIVE) then (CONTROL is
AUGMENTER_PEU)
- If (ERROR is VAL_PM) then (CONTROL is AUGMENTER_PEU)
- If (ERROR is VAL_NM) then (CONTROL is DIMINUER_PEU)
- If (ERROR is Z) then (CONTROL is TENIR)

6) What is the inference system type used here? Is there another type?

MANDANI. YES SUGENO METHOD.

30/09/2019 Dhaker ABBES 30

 Task 3 : example of control
7) What is the contribution (benefits) of fuzzy logic in comparison with a conventional PID
controller for these case study?

The fuzzy logic approach has minimum overshoot, minimum transient and steady state
parameters, which shows more effectiveness and efficiency of FLC than conventional
PID controller.

30/09/2019 Dhaker ABBES 31

 Task 4 : Fuzzy logic applied to energy management

You are given 3 slides taken from a power point presentation done by a PhD student from HEI L2EP in
ELECTRIMACS conference (Spain 2014). Well understand the problematic and the objectives then
respond to the question.

30/09/2019 Dhaker ABBES 32

Task 4 : Fuzzy logic applied to energy management

30/09/2019 Dhaker ABBES 33

Task 4 : Fuzzy logic applied to energy management

30/09/2019 Dhaker ABBES 34

Task 4 : Fuzzy logic applied to energy management

Question: Complete this table for work specifications (problem analysis):

30/09/2019 Dhaker ABBES 35

Task 4 : Fuzzy logic applied to energy management

30/09/2019 Dhaker ABBES 36

Task 4 : Fuzzy logic applied to energy management
DESIGN OF THE SUPERVISION STRATEGY
Methodology for HRPS energy management
STEP 1 STEP 4
Work specifications Determination of the membership functions
STEP 2
Design of the supervisor
STEP 5
STEP 3 Chart representation of fuzzy operating modes
Chart representation of operating modes - Operational graphs -
- Functional graphs -

STEP 6
Determination of the fuzzy rules

STEP 7
Determination of indicators to measure
the achievement of objectives

30/09/2019 Dhaker ABBES 37

Task 4OF: THE
DESIGN Fuzzy logic applied
SUPERVISION STRATEGYto energy management
Methodology for HRPS energy management
STEP 1
Work specifications

Objectives Constraints Means of actions

Predictive mode – LONGT TERM
Reducing energy bill Trains consumption predictions Storage power (Psto-ref-lgt)
(regarding short-term trades) RES forecast (Predictive reference
Electricity market fluctuations power)

Fuzzy Logic energy management – SHORT TERM

Limitation of subscribed power exceeding Subscribed power Storage power (Psto-ref-sht)
Favoring local RES consumption Storage limits (Predictive mode
Ensuring storage system availability RES availability adjustment)

30/09/2019 Dhaker ABBES 38

Task 4OF: THE
DESIGN Fuzzy logic applied
SUPERVISION STRATEGYto energy management
Methodology for HRPS energy management
STEP 2
Design of the supervisor

Ptrain_predictive To reduce the electricity bill

Predictive Psto_ref_lgt
PRES_forecast
mode
Electricity cost Long-term
the grid power excess amount
Pgrid ΔPexcess
To limit exceeding subscribed power +- K1 the power difference between train
Psubscribed Psto_ref_sht consumption and RES production
K5
SOC K2
To ensure the storage availability
ΔPlocal
Ptrain +- K3 FL energy
To favor local RES consumption PRES management
K4
HRPS supervision
K1, K2, K3, K4, K5 = normalisation gains

30/09/2019 Dhaker ABBES 39

Task 4OF: THE
DESIGN Fuzzy logic applied
SUPERVISION STRATEGYto energy management
Methodology for HRPS energy management Storage level is Small Storage level is Big
| |

STEP 3 3. To ensure storage

charging when near the
1. To limit exceeding grid
power by load shedding
3. To ensure storage
discharging when near
Chart representation of operating lowest level of
saturation
N2
2. To favor the RES local
consumption
N1
the highest level of
saturation
N3
modes |
Storage level is Medium
|
Storage level is Medium
- Functional graphs - 1
Negative-Big Zero Positive-Big

STEP 4

Plocal
0.5

Determination of the membership -1

1
Small
-0.8 -0.6 -0.4 -0.2 0
Plocal
Medium
0.2 0.4 0.6 0.8

Big
1

functions

SOC
0.5

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
SOC
Negative-Big Zero Positive-Big
Inputs: 1

• ΔPlocal (Negative-Big, Zero, Positive-Big) Pexcess

0.5

• SOC (Small, Medium, Big)

-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
Pexcess
Negative-Big Zero Positive-Big

• ΔPexcess (Negative-Big, Zero, Positive-Big)

1
Psto-ref-lgt

0.5

• Psto-ref-lgt (Negative-Big, Zero, Positive-Big) 0

-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
Psto-ref-lgt
Output: 1
Neg-Big Neg-Med Zero Pos-Med Pos-Big

• Psto-ref-sht (Negative-Big, Zero, Positive-Big)

Psto-ref-sht

0.5

0
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1
Psto-ref-sht

30/09/2019 Dhaker ABBES 40

Task 4OF: THE
DESIGN Fuzzy logic applied
SUPERVISION STRATEGYto energy management
Methodology for HRPS energy management
SOC is M & ΔPexcess is PB
|

ΔPlocal| is PB or Z
STEP 5 Psto_ref_lgt is NB Psto_ref_lgt is PB Psto_ref_lgt is NB Psto_ref_lgt is PB
| |

Chart representation of fuzzy

| |

Psto_ref_sht Psto_ref_sht Psto_ref_sht Psto_ref_sht Psto_ref_sht Psto_ref_sht

is Z is PM is PB is NB is NM is Z
operating modes - Operational |
Psto_ref_lgt is Z
|
Psto_ref_lgt is Z N1.11
|
Psto_ref_lgt is Z
|
Psto_ref_lgt is Z N1.12

graphs - |
ΔPlocal is NB N1.1

from 81 possible cases, only 30 fuzzy necessary rules

STEP 6 Mode Inputs Output

SOC ΔPlocal ΔPexcess Psto-ref-lgt Psto-ref-sht
Determination of the fuzzy rules M not NB PB PB Z
N1.11 M not NB PB Z PB
If SOC is M M not NB PB PB PB
and ΔPexcess_p.u. is NB (Pgrid<Psubscribed) M NB PB NB NB
and ΔPlocal_p.u. is PB (Ptrain>PRES) N1.12 M NB PB Z NM
and Psto-ref-lgt_p.u. is NB (to charge) M NB PB PB Z
then Psto-ref-sht_p.u. is NB (charge reference) …

30/09/2019 Dhaker ABBES 41

DESIGN
Task 4OF: THE SUPERVISION
Fuzzy STRATEGYto
logic applied energy management
Methodology for HRPS energy management
STEP 7
Determination of indicators to measure
the achievement of objectives

Economic indicator (monthly component of the exceeding subscribed power)

•
CMDPS     kt  P 2
( x) five tariff time periods T
tT xX t
excess
• α=0.3584€/kW
• kt (%) is a power coefficient for each time tariff period t
Pexcess  Pgrid  Psubscribed • Xt represents the index set x belonging to each time class t).

Energy indicator (ratio between the locally consumed energy of RES and the
produced one)

ERES  ERES _ noncons

I RES (%)  100
ERES

30/09/2019 Dhaker ABBES 42

Task 4 : Fuzzy logic applied to energy management
Predictive mode storage reference power Depends on electricity price in short term trades
0.1
Psto-ref-lgt
Ptrain_predictive

Power(p.u.)
To reduce the electricity bill
Predictive Psto_ref_lgt
PRES_forecast
mode
Electricity cost Long-term 0

Pgrid ΔPexcess
To limit exceeding subscribed power +- K1
Psubscribed Psto_ref_sht -0.1
K5
To ensure the storage availability
SOC K2
0 1 2 3 4 5 6 7
ΔPlocal 1
Ptrain +- K3 FL energy

price(p.u.)
To favor local RES consumption PRES management
K4
HRPS supervision 0.5
K1, K2, K3, K4, K5 = normalisation gains

electricity price
0
0 1 2 3 4 5 6 7
Weekday

Storage system:
 Csto = 5000 kWh
Production and consumption profiles  Psto-max = 1MW
 ηcharge=90%, ηdischarge=90%
1  SOCmin, SOCmax
Ptrain
0.8 Ppv
 time response constant (τ=0.5s)
Pwind
Power(p.u.)

0.6

0.4

0.2

0
0 1 2 3 4 5 6 7
Weekday
30/09/2019 Dhaker ABBES 43
 Task 4 : Fuzzy logic applied to energy management

Case 3: storage power adjustment

the subscribed power is reduced
thanks to RES local consumption
and storage management
1
Psubsribed
Power(p.u.)

Pgrid
0.5

0
0 2 4 6 8 10 12 14 16 18 20 22 0 2 4 6 8 10 12 14 16 18 20 22
1 storage system charges
SOC when RES exceed train
SOC(%)

0.5 consumption and when

0
economic interesting
0 2 4 6 8 10 12 14 16 18 20 22 0 2 4 6 8 10 12 14 16 18 20 22
0.05
Power(p.u.)

0
Psto-ref-lgt
-0.05 Psto
0 2 4 6 8 10 12 14 16 18 20 22 0 2 4 6 8 10 12 14 16 18 20 22
time(hours)
adjustment of Psto-ref-lgt reference power

30/09/2019 Dhaker ABBES 44

Task 4 : Fuzzy
SIMULATION logic
RESULTS applied
& CASE STUDIES to energy management
Comparison of different supervision cases
Subscribed power is reduced five times in HRPS supervision

Study case CMDPS IRES

Reference case 5338 € 0%

Case 1 (Predictive mode) 1024 € 96,5% well performance of FLEM

strategy compared to
Case 2 (FLEM strategy) 1036 € 95,5% predictive mode results

Case 3 (Adjustment) 942 € 96,9%

RES is locally consumed almost in totality

30/09/2019 Dhaker ABBES 45

 References
This first theoric part is extracted from : Fuzzy logic Massey University
course.
http://www.massey.ac.nz/~nhreyes/MASSEY/159741/Lectures/Lec201
2-3-159741-FuzzyLogic-v.2.pdf

Petronela Pankovits; Dhaker Abbes; Christophe Saudemont; Othman Moumniabdou; Julien

Pouget; Benoit Robyns, « Energy Management Multi-Criteria Design for Hybrid Railway
Power Substations, », 11th International Conference on Modeling and Simulation of
Electric Machines, Converters and Systems (ELECTRIMACS 2014).

Umesh Kumar Bansal and Rakesh Narvey, «Speed Control of DC Motor Using Fuzzy PID
Controller », Advance in Electronic and Electric Engineering. ISSN 2231-1297, Volume 3,
Number 9 (2013), pp. 1209-1220 , © Research India Publications.

30/09/2019 Dhaker ABBES 46

Thank you for
your attention !

Dhaker ABBES