Chiziqli algebra_Maruza7(24-25)

Download as pdf or txt
Download as pdf or txt
You are on page 1of 34

JAHON IQTISODIYOTI VA DIPLOMATIYA UNIVERSITETI

Tizimli tahlil va matematik modellashtirish kafedrasi

Fan : CHiZiQLi ALGEBRA

7 - MAVZU

KVADRATIK FORMALAR
Ma'ruza rejasi

▪Simmetrik matritsalar
▪Kvadratik formalar
▪Kvadratik formaning matritsasi
▪Kvadratik formaning aniqlanganligi
▪Kanonik shakl
▪Misollar
Kvadratik formalar

Kvadratik forma bu n ta o'zgaruvchilar yig'indisi bo'lib, uning har


bir hadi o'zgaruvchilardan birining kvadrati yoki ma'lum bir
koeffitsient bilan olingan ikki xil o'zgaruvchining ko’paytmasidir:
n n
L( x1 , x2 , ..., xn ) =  aij xi x j .
i =1 j =1
1- misol
1) Ikki o’zgaruvchili kvadratik forma:
2 2 2
L( x1 , x2 ) =  aij xi x j =  ( ai1 xi x1 + ai 2 xi x2 ) =
i =1 j =1 i =1

= ( a11 x1 x1 + a12 x1 x2 ) + ( a21 x2 x1 + a22 x2 x2 ) =


= a11 x12 + ( a12 + a21 ) x1 x2 + a22 x2 2 .

Masalan:
L( x1 , x2 ) = 9 x −12 x1x2 + 4 x2
1
2 2

yoki
L( x1 , x2 ) = − x + 6 x1 x2 −10 x2 .
1
2 2

Keyinchalik
2- misol
2) Uch o’zgaruvchili kvadratik forma:

3 3
L( x1 , x2 , x3 ) =  aij xi x j = a x + a22 x2 + a33 x3 +
11 1
2 2 2

i =1 j =1

+ ( a12 + a21 ) x1 x2 + ( a13 + a31 ) x1 x3 + ( a23 + a32 ) x2 x3 .

Masalan:
L( x1 , x2 , x3 ) = 3x12 − 2x1 x2 + 2x2 2 + x32
yoki

L( x1 , x2 , x3 ) = −7 x12 − 3x22 − x32 − 2 x1 x3 + 2 x2 x3.


orqaga
Kvadratik formaning matritsasi

Kvadratik formadaning matritsasi uning koeffitsientlaridan


tashkil topgan simmetrik matritsadir.

 a11 a12 ... a1n 


 
 a21 a22 ... a2 n 
A=
 ... ... ... ... 
 
 an1 an 2 ... ann 

Keyinchalik orqaga
3- misol
Kvadratik formaning matritsasini tuzing
 9 −6 
1) L( x1 , x2 ) = 9 x − 12 x1 x2 + 4 x2  A = 
2 2


1
 6 4 
2)  −1 3 
L( x1 , x2 ) = − x1 + 6 x1 x2 − 10 x2  A = 
2 2

 3 −10 
3)
L( x1 , x2 , x3 ) = 3x12 − 2 x1 x2 + 2 x2 2 + x32 
 3 −1 0 
 
 A =  −1 2 0 
 0 0 1
 
Kvadratik formaning rangi

Kvadratik formaning rangi uning A matritsasining r rangidir


( r = rank(A)).
Agar r = n bo'lsa, A matritsa xos bo'lmagan deb ataladi ,
r < n bo'lsa , A matritsa xos deyiladi .
4- misol
Kvadratik forma matritsasining rangini hisoblang

1) L( x1 , x2 ) = 9 x12 −12 x1x2 + 4 x22


 9 −6 
A=   rangA = 1
 −6 4 
2) L( x1 , x2 ) = − x12 + 6x1x2 −10x22

 −1 3 
A=   rangA = 2
 3 −10 
Kvadratik forma matritsasining rangini hisoblang:
3) L( x1 , x2 , x3 ) = 3x12 − 2x1 x2 + 2x2 2 + x32

 3 −1 0 
 
A =  −1 2 0   rangA = 3
 0 0 1
 
4)
L( x1 , x2 , x3 ) = −7 x12 − 3x22 − x32 − 2 x1 x3 + 2 x2 x3.

 −7 0 −1
 
A =  0 −3 1   rangA = 3
 −1 1 −1
 
Kvadratik formani matritsalar orqali tasvirlash

Matritsa yozuvida kvadratik formaning ko’rinishi: L = X T AX


bu erda - o'zgaruvchilarning matritsa ustuni. X = ( x1 , x2 ,..., xn )T

5-misol : kvadratik formalarni matritsalar orqali ifodalash

2 2 9 −6 𝑥1
𝐿(𝑥1 , 𝑥2 ) = 9𝑥1 − 12𝑥1 𝑥2 + 4𝑥2 ⇒ 𝐿(𝑥1 , 𝑥2 ) = 𝑥1 𝑥2
𝑥2 .
−6 4

3 −1 0 𝑥1
𝐿(𝑥1 , 𝑥2 , 𝑥3 ) = 3𝑥1 2 − 2𝑥1 𝑥2 + 2𝑥2 2 + 𝑥3 2 ⇒ 𝑥1 𝑥2 𝑥3 −1 2 0 𝑥2 .
0 0 1 𝑥3
Matritsaning bosh minorlari

Matritsaning 1-tartibli bosh (burchak) minori:


1 = a11
Matritsaning 2-tartibli bosh (burchak) minori:
a11 a12
2 =
a21 a22

Matritsaning 3-tartibli bosh (burchak) minori:


𝑎11 𝑎12 𝑎13
Δ3 = 𝑎21 𝑎22 𝑎23
𝑎31 𝑎32 𝑎33
6- misol
Kvadratik forma matritsasining bosh minorlarini hisoblang:

1) L( x1 , x2 ) = 9 x −12 x1x2 + 4 x2
1
2 2

 9 −6  9 −6
A=   1 = 9;  2 = = 0.
 −6 4  −6 4

2) L( x1 , x2 , x3 ) = 3x − 2x1 x2 + 2x2 + x3
1
2 2 2

3 −1 0
3 −1
𝐴 = −1 2 0 ⇒ Δ1 = 3; Δ2 = = 5; Δ3 = det 𝐴 = 5.
−1 2
0 0 1
Kvadratik formalarning aniqlanganligi

Kvadratik forma

Aniqlangan Yarim aniqlangan Aniqlanmagan

Musbat aniqlangan Nomanfiy aniqlangan

Manfiy aniqlangan Nomusbat aniqlang


Kvadratik formalarning musbat aniqlanganligi

Xos bo'lmagan kvadratik shakl faqat va agar bo'lsa, musbat


aniqlangan deyiladi, agar
a) kvadratik forma matritsasining barcha asosiy (burchak)
minorlari musbat bo’lsa(Silvestr mezoni);
b) kvadratik forma matritsasining barcha xos sonlari musbat
bo’lsa.
7- misol
Kvadratik formaning qanday aniqlanganligini tekshiring:
L( x1 , x2 , x3 ) = 3x − 2 x1 x2 + 2x2 + x3 .
1
2 2 2

Yechish :
 3 −1 0 
1) kvadratik formaning matritsasi  
A =  −1 2 0  .
 0 0 1
2) Silvestr mezoniga ko’ra  

3 −1
1 = 3  0;  2 = = 5  0; 3 = det A = 5  0 
−1 2
kvadratik forma musbat aniqlangan .
Kvadratik formalarning manfiy aniqlanganligi

Xos bo'lmagan kvadratik shakl faqat va agar bo'lsa, manfiy


aniqlangan deyiladi, agar
a) kvadratik forma matritsasining barcha asosiy (burchak)
minorlarining barcha toq tartiblilari manfiy va barcha juft
tartiblilari musbat bo’lsa(Silvestr mezoni);
b) kvadratik forma matritsasining barcha xos sonlari manfiy
bo’lsa.
8- misol
Kvadratik formaning qanday aniqlanganligini tekshiring:

L( x1 , x2 , x3 ) = −7 x12 − 3x22 − x32 − 2 x1 x3 + 2 x2 x3.


Yechish :
 −7 0 −1
1) kvadratik formaning matritsasi  
A =  0 −3 1  .
 −1 1 −1
2) Silvestr mezoniga ko'ra  

−7 0
1 = −7  0; 2 = = 21  0; 3 = det A = −11  0 
0 −3
kvadratik forma manfiy aniqlangan .
9- misol
Kvadratik formani tekshiring:
L( x1 , x2 , x3 ) = −2 x + 3x3 + 4x1x2 − 4x1x3 + 8x2 x3.
1
2 2

Yechish :
1) kvadratik formaning matritsasi  −2 2 −2 
 
A =  2 0 4 .
 −2 4 3 
2) Silvestr mezoniga ko'ra  

−2 2
1 = −2  0;  2 = = −4  0; 3 = det A = −12  0 
2 0
kvadratik shakl aniqlanmagan.
10- misol
Kvadratik formani tekshiring:

L( x1 , x2 ) = 9 x −12 x1x2 + 4 x2
1
2 2

Yechish :
 9 −6 
1) kvadratik formadaning matritsasi A= .
 −6 4 
det A = 0 
L ( x1 , x2 ) = ( 3x1 − 2 x2 ) = y12  0 
2
2)

Kvadratik forma nomanfiy aniqlangan.


11- misol
Kvadratik formani tekshiring:

L( x1 , x2 , x3 ) = − x12 − x22 − x32 − 2x1x2


Yechish:
1) kvadratik formadaning matritsasi  −1 −1 0 
 
A =  −1 −1 0  .
det A = 0   0 0 −1
 

L ( x1 , x2 ) = − ( x1 + x2 ) − x32 = − y12 − y2 2  0 
2
2)

Kvadratik forma nomusbat aniqlangan.


Kvadratik formaning kanonik shakli

Agar L( x1 , x 2 , ..., x n ) kvadratik formaning barcha aij i  j


koeffitsientlari nolga teng bo'lsa, u kanonik ko’rinishda deyiladi:

n
L( x1 , x2 ,..., xn ) =  aij xi = a x + a22 x2 + ... + ann xn
2
11 1
2 2 2

i =1
Kvadratik formani kanonik shaklga keltirish

n
L( x1 , x2 ,..., xn ) =  aij xi 2 = a11 x12 + a22 x2 2 + ... + ann xn 2
i =1

1) Lagranj usuli yordamida;


2) Ortogonal almashtirishlar yordamida;
Lagranj usuli bilan kanonik shaklga keltirish

12- misol f (x1 , x2 ) = 2 x12 + 8x1x2 + 10 x22 .


Yechim .

f (x1 , x2 ) = 2
2 x1 + 8x1 x2 + 10x2
2
=2 ( 2
x1 )
+ 4 x1 x2 + 10x2
2
=
= 2 𝑥12 + 2 ⋅ 𝑥1 ⋅ 2𝑥2 + 4𝑥22 − 4𝑥22 + 10𝑥22 = 2 𝑥1 + 2𝑥2 2 − 4𝑥22 + 10𝑥22 = 2 𝑥1 + 2𝑥2 2 + 2𝑥22 ,

𝑥1′ = 𝑥1 + 2𝑥2 , x'2 = x2 ; f (


x1' , x'2 ) = 2( )
' 2
x1 +2 ( )
' 2
x2

𝑥1 = 𝑥1′ − 2𝑥2′ , 1 −2
X = T  X ,
ቊ 𝑇=
𝑥2 = 𝑥2′ . 0 1
13- misol f (x1 , x2 , x3 ) = x12 + 2 x1 x2 + 3x22 + 4 x1 x3 + 6 x2 x3 + 3x32
Yechim. f ( x1 , x2 , x3 ) = x12 + 2 x1(x2 + 2 x3 ) + 3x22 + 6 x2 x3 + 3x32 =

= x12 + 2 x1 (x2 + 2 x3 ) + (x2 + 2 x3 )2 − (x2 + 2 x3 )2 + 3x22 + 6 x2 x3 + 3x32 =


= ( x1 + x2 + 2 x3 )2 − x22 − 4 x2 x3 − 4 x32 + 3x22 + 6 x2 x3 + 3x32 =

(
= ( x1 + x2 + 2 x3 )2 + 2 x22 + 2 x2 x3 − x32 = ( x1 + x2 + 2 x3 )2 + 2 x22 + x2 x3 − x32 =)
 2 x x 2
x 2
= ( x1 + x2 + 2 x3 ) + 2 x2 + 2 x2  +
2 3 3
−  − x32 =
3

 2 4 4 
2
 x 
= ( x1 + x2 + 2 x3 )2 + 2 x2 + 3  − x32
3
 2 2

x1' = x1 + x2 + 2 x3 ; x'2
1
= x2 + x3 ; x'3 = x3
2
f (
x1' , x'2 , x'3 )= ( )
' 2
x1 +2 ( )
' 2
x2 ( )
3 '
− x3
2
2
14- misol f (x1 , x2 , x3 ) = x1x2 + x1x3 + x2 x3

xi = yi − y j , x j = yi + y j , xk = yk , k  i ,k  j

( )( )
xi  x j = yi − y j yi + y j = yi2 − y 2j

Yechim.
. x1 = y1 − y2 , x2 = y1 + y2 , x3 = y3 ,

f (x1, x2 , x3 ) = ( y1 − y2 )( y1 + y2 ) + ( y1 − y2 )y3 + ( y1 + y2 )y3 =

= y12 − y22 + y1 y3 − y2 y3 + y1 y3 + y2 y3 = y12 − y22 + 2 y1 y3 =

( )
= y12 + 2 y1 y3 + y32 − y32 − y22 = ( y1 + y3 )2 − y22 − y32

x1' = y1 + y3 , x'2 = y 2 , x'3 = y3 f ( x1 , x2 , x3 ) = ( ) −( ) −( )


' 2
x1 ' 2
x2 ' 2
x3
1
y1 = (x1 + x2 ), y2 = 1 (x2 − x1 ), y3 = x3 x1'
1 1 1 1
= x1 + x2 + x3 ; x2 = − x1 + x2 ; x'3 = x3
'
2 2 2 2 2 2
Ortogonal almashtirishlar usuli bilan kanonik shaklga keltirish

Teorema. A simmetrik matritsa bo'lsin . U holda A :


1) Musbat aniqlangan bo’ladi, agar uning barcha xos qiymatlari musbat bo’lsa;
2) Manfiy aniqlangan bo’ladi, agar uning barcha xos qiymatlari manfiy bo’lsa;
3) Nomanfiy aniqlangan (yarim musbat aniqlangan) bo’ladi, agar
uning barcha xos qiymatlari nomanfiy bo’lsa;
4) Nomusbat aniqlangan (yarim manfiy aniqlangan) bo’ladi, bo’ladi, agar
uning barcha xos qiymatlari nomusbat bo’lsa;
15- misol

f( )
x1 , x2 = 2 x1
2
+ 8 x1 x2 + 8x2
2

Yechim.

 2 4 2− 4 (2 −  )(8 −  ) − 16 = 0,
А =  , = 0, 16 − 8 − 2 + 2 − 16 = 0,
 4 8 4 8−
2 − 10 = 0,

 = 0 ,
 = 10.

f ( x1 , x2 ) = 10 ( )
' 2
x2
2 x1 + 4 x2 = 0,  − 2a 
 =0  x1 = −2x2 ;  , a  0. 5a 2 ,
4 x1 + 8 x2 = 0;  a 
− 8 x1 + 4 x2 = 0, b
 = 10  x2 = 2 x1;  , b  0.
4 x1 − 2 x2 = 0. 5b 2 .
 2b 
−2 1 1 2
e1' = e1 + e2 ; e'2 = e1 + e2 .
5 5 5 5
 2 ' 1 '
 2
−
1 

 x1 = − 5 x1 + 5 x2 ,
T = 5 5 ,
X = T  X , 
 1 2   x = 1 x' + 2 x' .
   2 1 2
 5 5 5 5
16- misol

𝑓 𝑥1 , 𝑥2 , 𝑥3 = 4𝑥12 + 4𝑥1 𝑥2 + 4𝑥1 𝑥3 + 4𝑥2 𝑥3 + 4𝑥22 + 4𝑥32

Yechim.

A xarakteristik tenglamani yechamiz :


(𝜆 − 8)(𝜆 − 2)2 = 0
𝜆 = 8 yoki 2
A ning xos sonlari:  1 = 2  2= 8

𝑓 𝑥1 , 𝑥2 , 𝑥3 = 2 𝑥 ′ 1 2 + 2 𝑥2′ 2 +8 𝑥3′ 2
1 0 1Τ 2 0
 1= 2 0 , 1 0 , 1Τ 2
−1 −1 −1Τ 2 −1Τ 2
 2= 8
1 1Τ 3
1 1Τ 3
1
1Τ 3

1Τ 2 0 1Τ 3 1 1
𝑥1 = 𝑥1′ − 𝑥3′ ,
𝑃= 0 1Τ 2 1Τ 3 2 2
𝑋 = 𝑃 ⋅ 𝑋′, 1 1
−1Τ 2−1Τ 21Τ 3 𝑥2 = 𝑥2′ − 𝑥3′ ,
2 2
1 1 1
𝑥3 = 𝑥1′ + 𝑥2′ + 𝑥3′ .
3 3 3
17- misol

𝑓 𝑥1 , 𝑥2 , 𝑥3 = 5𝑥12 + 8𝑥1 𝑥2 + 4𝑥1 𝑥3 + 4𝑥2 𝑥3 + 5𝑥22 + 2𝑥32

Yechim.  5 4 2
A = 4 5 2
2 2 2
 

A uchun xarakteristik tenglamani yechamiz :


− ( − 10)( − 1) 2 = 0
 = 10 or 1
A ning xos sonlari 10 va 1 ga teng.

𝑓 𝑥1 , 𝑥2 , 𝑥3 = 10 𝑥 ′ 1 2 + 𝑥2′ 2 + 𝑥3′ 2
2 2Τ3
 1 = 10 xos songa mos xos vektor: 2 2Τ3
1 1Τ3
0 −1Τ 5
0 −1
 2 = 1 xos songa mos xos vektorlar: 1Τ 5 , 0
1 , 0
−2 2 −2Τ 5 2Τ 5

2Τ3 0 −1Τ 5 2 ′ 1 ′
𝑥1 = 𝑥1 − 𝑥3 ,
𝑃 = 2Τ3 1Τ 5 0 3 5
1Τ3−2Τ 5 2Τ 5
𝑋 = 𝑃 ⋅ 𝑋′, 2 ′ 1 ′
𝑥2 = 𝑥1 + 𝑥2 ,
3 5
1 ′ 2 ′ 2 ′
𝑥3 = 𝑥1 − 𝑥2 + 𝑥3 .
3 5 5
E'tiboringiz uchun rahmat !

You might also like