Assignment 2
Assignment 2
DISCUSS AND SUMMARISE A JOURNAL ABOUT THE APPLICATION OF S-NORM AND T-NORM
IN REAL LIFE.
JOURNAL TITLE:
TABLE OF CONTENTS:
TOPIC PAGE
1. INTRODUCTION 1
2. PROBLEM STATEMENT 2
3. OBJECTIVE 2
4. METHADOLOGY 3
6. CONCLUSION 7
7. REFERENCES 7
8. APPENDICES 7
The real world is complex. Control applications are kinds of problem for which fuzzy
logic has had the greatest success and acclaim. Many of the consumer products that we use
today involve fuzzy control. Control systems abound in our everyday life [1]. Since Zadeh
proposed the basic principle of fuzzy logic controllers in 1968, the basic principle of fuzzy
logic controllers has not only found tremendous applications in variety of industrial use and
gone through substantial theoretical developments. The three most important concept
associated with fuzzy logic controller were intersection and union, implication function and
composition rules of inference.
Fuzzy logic control algorithms MIN-MAX operators have been extensively used in the
membership functions. Theoretical and experimental studies have indicated that some
operators may work better than other in some situation. This implies that some operators
may be more suitable in the context of a given decision making process. The choice of
operator is always a matter of context and it mostly depends on the real world problem
which is to be modelled.
Therefore, to use the general notion of T-operators in the design of fuzzy logic
controllers so that more options and flexibility are available for the selection of T-operators
that may be better suited for a given control problem.
3. OBJECTIVE
In the design of almost all existing fuzzy controllers, only one of many sets of T-operators
has been used and these are MIN(x,y), MAX(x,y) and (1 – x). Table 1 below shows the
various choices of T-operators.
Table 1: T-operators
Example:
If the error (e) is large and the change in error (Δe) is small, then the change in the
process input (Δu) is medium.
If the error (e) is small and the change in error (Δe) is zero, then the change in the
process input (Δu) is small.
If …
Where;
e, Δe and Δu linguistic variable
large, small, zero, etc fuzzy set value
Extension of Zadeh’s
implication functions
Now if the actual process error, the actual change in error and the actual change in the
process input take on fuzzy values E’, ΔE’ and ΔU’ at a particular moment. The actual change
in the process input ΔU’ can then be obtained as follow:
A fuzzy controller with the simplest structure is used in which the proposed fuzzy control
algorithm is utilized. It is used to control a given class of processes for given reference inputs
and the system response is a function of the T-operators.
The purpose of the following simulation studies is to see what the system responses
look like when other parameters are constant. Figure 1 below shows the block diagram of
the fuzzy controller. For simplicity, only a single input and single output structure is
considered.
Also, the following set of control rules are proposed for the controller based on the
knowledge of an expert operator. Figure 2 shows the membership function of fuzzy sets.
Computer simulation we used to examine the impact of each of the following five typical T-
operators on the proposed fuzzy controller’s. it is also done to reveal the differences among
these T-operators in term of control problems. In this study, it was decided to use only five
T-norms (T1 to T5) and five T-conorms (T*1 to T*5) to investigate the effect. The summarise
simulation result for all 30 couples of T-norms and T-conorms are shown in Table 2 below.
In order to obtain more general conclusion, several different plants are used which
are given as follows:
1) First-order : y’ + ay = bu
2) Second-order : y’’ + ay’ + by = cu
From the process output, a general conclusion that can be drawn is that there are
differences between using one couple of T-norms and T-conorms or another, under the
condition proposed here for most of the 25 couples of the T-norms and T-conorms.
Fuzzy control algorithm proposed in this paper is an extension of the conventional one. This
is done by replacing the MIN-MAX operators in the conventional control algorithm by the
generalized notion of T-operators. This new fuzzy control algorithm provides a general and
flexible methodology for the design of fuzzy logic controllers.
Generally speaking, in the proposed fuzzy controller using one couple of T-norms
and T-conorms or another does make a difference in terms of the system responses. Further
studies are underway using the notion of adaptivity which optimizes the location and width
of the membership function during the operation.
7. REFERENCES
1. M.M. Gupta and J. Qi, Design of fuzzy logic controllers based on generalized T-
operators, Fuzzy Sets and Systems 40 (1991) 473-489
2. Timothy J. Ross, Fuzzy logic with engineering applications, Wiley third edition (2010)
8. APPENDICES