.
1. .
1 .
(.2.29): (), ( - ) (0,0) ( ) .
: ; () : [0,0].
. 2.29.
:
,
,
.2.29;
;
, ;
, m .
, , :
( 2.5);
;
(.2.30);
( 2.5 ).
, :
;
.
, .
. 2.30.
,
{NM, NS, ZE, PS, PM} = {negative medium, negative small, Zero, positive small, positive medium}.
:
{NM,, PM} v {NM,, PM}
{NM, NS, ZE, PS, PM}.
.
:
l:
,
,
.
( , - ) :
,
- ()
.
.
, .
.
? (), look-up-table, 2.5 .
( ) 2.5:
1 - NM, 2 - NS, 3 - ZE, 4 - PS, 5 - PM.
|
|
2.5 | 2.5 |
![]() |
![]() |
, 25 - {1432152432124514312211345}, , . : = 100; = 100; = 0.7; = 0.03.
tournament selection: . . , : 1, , 2, refinement stage. , ( ). , [0,0]. 1.
1.
30. () , = 0.5
.
, , .. .
.
1 ():
(if ( < 0.5) and (
< 0.5) then fitness =
%{
}
else (if ( = 175) then
%{ }
(if ( < 1.0) and (
< 1.0) then fitness =
else fitness =
)
if (( > 5.0) and (
> 5.0) then fitness =
).
: % .
. , , . 8 ( 175 ).
175 , ( ( < 1.0) (
< 1.0) , . (
> 5.0) (
> 5.0), .
2.31 2.32 .
. 2.31.
. 2.32.
2.
- 31 - 100-.
2 ( ):
(if ( < 0.5) and (
< 0.5)
%{
}
then fitness =
else (if ( = 175) % { }
then fitness = ))
if (( > 5.0) and (
> 5.0) then fitness =
).
. 2.33 1.
![]() | ![]() |
. 2.33.
|
|
![]() | ![]() |
. 2.33.
. .2.33 (), (b) (d) , (c) .
2. : , , , ( ).
. 2.34.
. .
. 2.34.
: 1) ; 2) , ;
( ) , , , :
,
- .
( )
,
- ( ),
- .
, ,
. , , , .
, .
- .
2: . , , (
). ,
,
,
.
S =
,
- :
,
,
.
, . ,
,
,
,
R .
.
. (GA discretization step).
:
.
( ), , :
:
=
.
. , :
(
).
. 2.34.
. 2.34.
, , , .
2.35 - .
3:
, . .
, SSCQ - ( . simulation system of control quality). SSCQ .2.36.
. 2.35. / -.
. 2.36. / ,
. 2.37.
|
|
. . (control error), (derivative of error) (integral error), () (teaching signal (TS)).
. 2.37.
() (Fuzzy Neural Network (FNN)), .
. .
2.4. :
(Artificial neural networks), .
, , .
:
1) , ( );
2) , .
:
;
;
;
;
;
;
;
;
.
() , , . , . , 1011 1015 . . , . ( ) . , ( ). . () .
McCuloch Pitts (1943), Hebb (1949), Rosenblatt (1958,1961), Rochester (1956), Widrow (1960), Kohonen (1972), Hopfiled (1982), Cohen Grossberg (1983) . , .