Professional Documents
Culture Documents
net/publication/263466665
CITATIONS READS
2 149
2 authors:
Some of the authors of this publication are also working on these related projects:
Three Phase Three Level Voltage Fed Inverter for a Doubly Fed Induction Motor View project
All content following this page was uploaded by Rami A. Maher on 28 June 2014.
Abstract—Although fuzzy systems demonstrate their ability to As the field of fuzzy computing is an active research field,
solve different kinds of problems in various applications, there is many methodologies are developed for constructing and
an increasing interest on developing solid mathematical computing the FLC. The designer of a fuzzy controller for
implementations suitable for control applications such as that certain control application is faced with the many design
used in fuzzy logic controllers (FLC). It is well known that, wide
choices that the fuzzy set theory provides. Fundamental
range of parameters is needed to be specified before the
construction of a fuzzy system. To simplify in a systematic way comparisons and suggestions are found in the literature, and
the design and construction of a general fuzzy system, and they are well presented by [8-11,16-18]. These computing
without loss for generality a full parameterization process for a approaches are not unique; it is mainly due the lack of having
singleton type FLC is proposed in this paper. The presented a general good mathematical formulation for the fuzzy
methodology is very helpful in developing a universal computing system construction algorithm.
algorithm for a standard fuzzy like PID controllers. An A solution for this problem may be solved if good
illustrative example shows the simplicity of applying the new parameterization process is developed. The parameterization
paradigm. of a fuzzy system is insufficiently addressed in the literature.
RT
Keywords—parameterization;fuzzy logic controller(FLC);
This work is a trial to solve this problem, and mainly devoted
singletone FLC ; PID to present a fuzzy system for control applications, whether
used in construction of the FLC, or in fuzzy system modeling
IJE
fuzzy system, from different aspects according to the systems employ a normalized fuzzy sets to specify the entire
classification of parameters summarized by Table I. partition of a fuzzy variables), a great number of reduction in
the number of parameters defining the input MF can be
introduced. Without loss of generality, this implementation is
B. Parameterization of Input Membership Functions adopted in the work. Hence (3) will be reduced to:
Consider the fuzzy system as shown in Fig. 2. which is a
simplified form of Fig. 1 with input x and output u , b b b
in 11 12 13 (4)
where x nx is the fuzzy system input variables,
b b b 21 22 23
u nu
is the fuzzy system output variables, and { n x ,
C. Parameterization of Membership Degree
nu } are the dimension of input and output variables.
In FLC computing, only the degree of membership is
further processed. Often information is lost during this
TABLE I. PARAMETER CLASSIFICATION OF A FLC
procedure, although it is not required that the MFs are
CLASS PARAMETERS normalized (i.e. their sum is equal to one for all x ), this
LOGICAL REASONING MECHANISM, FUZZY property is called fuzzy partition and often is employed
OPERATORS, MEMBERSHIP FUNCTIONS because it makes the interpretation and computation easier
TYPE, DEFUZZIFICZTION METHOD
[7]. The degree of membership is obtained for the current
STRUCTURAL RELEVANT VARIABLES, NUMBER OF
MEMBERSHIP FUNCTIONS, NUMBER OF input vector by:
mx
x
RULES
i
CONNECTIVE ANTECEDENT PART OF THE RULE, i
(5)
CONSEQUENT PART OF THE RULE, RULE
WEIGHTS
and,
OPERATIONAL MEMBERSHIP FUNCTION VALUES x ( xi ; in(i ) , mx ), i 1,2,3......, n x .
i i
(6)
where (.;.) is a generalized MF producing degrees of
membership for all fuzzy sets related to the input x i and
x [ 1 , 2 ,....., k ,......m ] (7)
RT
i xi
i th input (shape, no. of sets, width of the fuzzy set, spacing x 1 2 3
1
(8)
and overlapping between the sets…).
where the index T is put as an indicator targeting for using
the triangular function in calculating the membership degree.
Example 1:
For a fuzzy system with two inputs, each defined with three 1 T (cx1 ;[a11b11c11 ]), k 1
triangular MFs, then n x 2 ; m x1 3 ; m x2 3 2 T (cx1 ;[a12b12c12 ]), k 2 (9)
and c 3 . 3 T (cx1 ;[a13b13c13 ]), k 3
A triangular MF is normally defined by three points (a, b, c). The membership degree of each fuzzy set (k 1,2,3) for a
Therefore in will be defined by triangular MF is evaluated by:
in 29
a b c a b c a b c
in 11 11 11 12 12 12 13 13 13 (3)
a21b21c21a22b22c22a23b23c23
This approach, which is adopted by the presented Hence, certainty of the 6th rule is evaluated by the product of
methodology, will cover both reduced and full RB the 2nd MF value (defined by mf 2 ) for input 1 (calculated at
representation by letting the certainty of each undefined rule
to be zero, and set one otherwise. the measurement value of input x1 ) and the 3rd MF value
Also, most of FLCs used in control application are (defined by mf 3 ) for input 2 (calculated at the measurement
assumed to have fuzzy propositions connected with fuzzy
and connectives only. Thus the structure of the rules value of input x 2 ).
premise is defined by:
E. Parameterization of the Rule Base Output
rules n R n x
(11 )
After the dof has been calculated for all the rules, and, for
where n R denotes the total number of rules in the RB, and the inference engine process to be completed, it is required
now to consider the outputs of the RB.
n x represents the number of inputs used in constructing the Now, consider the output parameters vector by defining:
FLC. Noting that n R is defined by: out n R nu
(17)
nx
n R m xi (12)
where nu represents the number of controller outputs, n R
i 1 represents the dimension of the rule base.
a row of rules , connects the index of the input fuzzy sets for The i th row within out is defined by:
each input variable defined by each rule, and hence reflects out
(i )
[hi1 hij hin ]; j 1,2,....nu . (18)
the degree of membership xi function to be taken into
u
out
T
for normalization the input and output UODs.
U
UODx1 [ , ] g x1 1 /
nR
(23)
i 1
(i )
UODx2 [ , ] g x2 1 /
where and out are defined by (13) and (18) UODu1 [ , ] g u1 1 / (27)
respectively. Actually they are part of the pre-processing and post-
The generalized form of the defuzzified output can be written processing stages.
in the following form:
U D(, out ) (24) III. FLC COMPUTING ALGORITHM
where D can be any defuzzification formula applied to
Back to Table 1, a complete set of FLC parameters are
evaluate the crisp output U . defined and connected by systematic mathematical
formulation suitable for the computing algorithm, and
G. Parameterization of Input and Output Variables summarized as follows:
In closed loop systems as shown in Fig. 4., there are 1) The inference mechanism uses the singleton fuzzy system
several signals which should be taken into consideration as given by the general RB structure:
when the control signal is calculated. The error signal IF ( rules ) THEN ( out ) (28)
between the set point y and the measurement output
2) Relevant inputs are { e1 , e1 } and outputs are
{ u1 , u1 }.
3) Number of MFs defined for each input and output is given
by m xi and mu j .
Fig. 4. Typical FLC in a closed loop control system. 4) Input MFs characteristics in are defined by (3), and
output MFs are defined by singletons (19).
5) The fuzzy operators are defined to use the product
Hence:
mx1 5; mx2 5; mu1 9
And according to (2) for a triangular MF it gives: c 3
and ni 15 .
RT
Fig. 5. Components of a FLC computing algorithm.
IJE
Consider the problem of balancing an inverted pendulum 4) in is evaluated by (1), and its reduced version (4) gives:
on a cart [15] as shown in Fig. 6. for which the dynamic 1 0.5 0 0.5 1
equation is given by: in
1 0.5 0 0.5 1
with normalization scaling factors evaluated by (27) as:
u 0.05 y 2 sin( y )
9.8 sin( y ) cos( y )
y 1.1 g x1 2 / ; g x2 4 / ; g u1 1 / 20
(30)
4 0.1 rules
0.5 cos 2 ( y ) 5) reflects the indices of the MFs involved in the
3 1 .1 construction of the rule premise, which is identified by
the two inputs x1 and x 2 respectively. n R 25 is
calculated by (13), and RB premise is generated by (12)
where y is the angle of the pendulum with respect to the as:
vertical line and y is the pendulum angular velocity and u T
1111122222333334444455555
rules
1234512345123451234512345
is the applied control force.
To evaluate the presented methodology, a PD like fuzzy
controller is to be constructed for the non-linear control 6) The consequence out of the rule base is constructed
system given by (30), using the developed singleton FLC
according to (18), using the summation formula defined
computing algorithm. We will discuss the regulator problem,
by passino [8], gives the following:
i.e. keeping the inverted pendulum balanced in a vertical
position with reference to different initial positions of y (0) .
out [ 1 .75 .5 .25 0 | .75 .5 .25 0 -.25| the pendulum to its vertical position from different initial
conditions bounded by y(0) [42.97,42.97] .
.5 .25 0 -.25 -.5| .25 0 -.25 -.5 -.75|
T
0 -.25 -.5 -.75 - 1 ] VI. CONCLUSION
1
where the output MFs are assumed to have the same A full parameterization process for a singleton fuzzy
partitions of the input MFs, that is identified by: system is developed. It presents a systematic methodology
{ mf1 , mf 2 , mf 3 , mf 4 , mf 5 } = {NL, NS, Z, PS, PL} for developing a singleton fuzzy logic controller for control
If normalization is used, it will be in the form of: applications. The assumptions made by the parameterization
{-1,-0.5, 0, 0.5, 1} process is highly simplified the FLC computing algorithm. A
well-known inverted pendulum problem is chosen for
The rule base table generated by rules and out could evaluating the capability of the
be summarized as shown by Table II.
TABLE II.
REFERENCES
(1980).
[15] L. X. Wang, ―A Supervisory Controller for 'Fuzzy Control Systems that
Guarantees Stability,‖ IEEE Transactions on Automatic Control, vol.
39, pp. 1845-1847, Sep. 1994.
[16] A.-Lopez , A. Aguilar-martin ,J.A ,"Simplified Fuzzy Controller : The
Natural Logic Control ", vol. 14,no.1,pp.16-30,2006.
[17] Prem Kumar P.I Kar and I Behera, "Variable Gain Controller for
Nonlinear Systems Using T-S Fuzzy Model, IEEE Trans on SMC,
2009.
[18] A.G. Aissaoui, M. Abid, A. Tahour and A. C. Megherbi, "A Fuzzy
Logic and Variable Structure Control for Permanent Magnet
Synchronous Motors", International Journal of Systems Control,
vol.,no.1, pp.13-21, 2010.