ROHINI COLLEGE OF ENGINEERING & TECHNOLOGY

REDUCTION OF QUADRATIC FORM TO CANON ICAL FORM BY

ORTHOGONAL TRANSFORMATION
Quadratic Form
A homogeneous polynomial of second degree in any number of variables is called a
quadratic form.
The general Quadratic form in three variables {𝑥1 , 𝑥2 , 𝑥3 } is given by
𝑓(𝑥1 , 𝑥2 , 𝑥3 ) = 𝑎11 𝑥12 + 𝑎12 𝑥1 𝑥2 + 𝑎13 𝑥1 𝑥3 +
𝑎21 𝑥1 𝑥2 + 𝑎22 𝑥22 + 𝑎23 𝑥2 𝑥3 + 𝑎31 𝑥3 𝑥1 + 𝑎32 𝑥2 𝑥2 + 𝑎33 𝑥32
This Quadratic form can written as 𝑓(𝑥1 , 𝑥2 , 𝑥3 ) = ∑3𝑖=1 ∑3𝑗=1 𝑎𝑖𝑗 𝑥𝑖 𝑦𝑗
𝑎11 𝑎12 𝑎13 𝑥1
𝑓(𝑥1 , 𝑥2 , 𝑥3 ) = (𝑥1 𝑥2 𝑥3 ) (𝑎21 𝑎22 𝑎23 ) (𝑥2 )
𝑎31 𝑎32 𝑎33 𝑥3
= 𝑋 ′ 𝐴𝑋
𝑥1
Where 𝑋 = (𝑥2 ) and A is called the matrix of the Quadratic form.
𝑥3
Note: To write the matrix of a quadratic form as
coeff. ofx 2 1/2coeff. ofxy 1/2coeff. ofxz
A = (1/2coeff. ofxy coeff. ofy 2 1/2coeff. ofyz )
1/2coeff. of xz 1/2coeff. ofyz coeff. of z 2

Example: Write down the Quadratic form in to matrix form

(i)𝟐𝐱 𝟐 + 𝟑𝐲 𝟐 + 𝟔𝐱𝐲
Solution:
coeff. of x 2 1/2coeff. of xy
A=( )
1/2coeff. of xy coeff. ofy 2
2 3
=( )
3 3
(ii) 𝟐𝐱 𝟐 + 𝟓𝐲 𝟐 − 𝟔𝐳 𝟐 − 𝟐𝐱𝐲 − 𝐲𝐳 + 𝟖𝐳𝐱
Solution:
coeff. ofx 2 1/2coeff. ofxy 1/2coeff. ofxz
A = (1/2coeff. ofxy coeff. ofy 2 1/2coeff. ofyz )
1/2coeff. of xz 1/2coeff. ofyz coeff. of z 2

MA8251 ENGINEERING MATHEMATICS II

ROHINI COLLEGE OF ENGINEERING & TECHNOLOGY

2 −1 4
= (−1 5 −1/2)
4 −1/2 −6
Example: Write down the matrix form in to Quadratic form
𝟐 𝟏 −𝟑
(i)( 𝟏 −𝟐 𝟑)
−𝟑 −𝟐 𝟓
Solution:
Quadratic form is 2 x12 − 2x22 + 6x32 + 2x1 x2 − 6x1 x3 + 6x2 x3
𝟏 𝟏 𝟐
(ii)(𝟏 𝟑 𝟏)
𝟐 𝟏 𝟔
Solution:
Quadratic form is x12 + 3x22 + 6x32 + 2x1 x2 + 4x1 x3 + 2x2 x3 .

Example: Reduce the Quadratic form𝐱 𝟐𝟏 + 𝟐𝐱 𝟐𝟐 + 𝐱 𝟐𝟑 − 𝟐𝐱 𝟏 𝐱 𝟐 + 𝟐𝐱 𝟐 𝐱 𝟑 + 𝟔𝐱 𝟐 𝐱 𝟑 to

canonical form through an orthogonal transformation .Find the nature rank, index,
signature and also find the non zero set of values which makes this Quadratic form as zero.
Solution:
1 −1 0
Given A = (−1 2 1)
0 1 1
The characteristic equation is 3 − s1 2 + s2  − s3 =0
s1 = sum of the main diagonal element
= 1+2+1= 4
s2 = sum of the minors of the main diagonalelement
2 1 1 0 1 −1
=| |+| |+| | =1+1+1=3
1 1 0 1 −1 2
1 −1 0
s3 = |A| = |−1 2 1| = 0
0 1 1
Characteristic equation is 3 − 42 + 3 = 0
⇒  = 0 ; (2 − 4 + 3) = 0
⇒  = 0,1,3
To find the Eigen vectors:

MA8251 ENGINEERING MATHEMATICS II

ROHINI COLLEGE OF ENGINEERING & TECHNOLOGY

𝑥1
Case (i) When  = 0 the eigen vector is given by(A − I)X = 0 where X = (𝑥2 )
𝑥3
1 −1 0 𝑥1 0
𝑥
⇒ (−1 2 1) ( 2 ) = (0)
0 1 1 𝑥3 0
x1 − x2 + 0x3 = 0 … (1)
−x1 + 2x2 + x3 = 0 … (2)
0x1 + x2 + x3 = 0 … (3)
From (1) and (2)
x1 x x
= −12 = 2−1
3
−1
x1 x2 x3
= =
−1 −1 1

−1
X1 = (−1)
1
𝑥1
Case (ii) When  = 3 the eigen vector is given by(A − I)X = 0 where X = (𝑥2 )
𝑥3
1−3 −1 0 𝑥1 0
⇒ ( −1 2−3 1 ) (𝑥 2 ) = (0)
0 1 1 − 3 𝑥3 0
−2x1 − x2 + 0x3 = 0 … (4)
−x1 − x2 + x3 = 0 … (5)
0x1 + x2 − 2x3 = 0 … (6)
From (4) and (5)
x1 x2 3 x
= = 2−1
−1 2
x1 x2 x3
= =
−1 2 1

−1
X2 = ( 2 )
1
𝑥1
Case (iii) When  = 1 the eigen vector is given by(A − I)X = 0 where X = (𝑥2 )
𝑥3
1−1 −1 0 𝑥1 0
⇒ ( −1 2−1 1 ) (𝑥 2 ) = (0)
0 1 1−1 𝑥 3 0
0x1 − x2 + 0x3 = 0 … (7)

MA8251 ENGINEERING MATHEMATICS II

ROHINI COLLEGE OF ENGINEERING & TECHNOLOGY

−x1 + x2 + x3 = 0 … (8)
0x1 + x2 + 0x3 = 0 … (9)
From (7) and (8)
x1 x
2 3 x
= 0−0 = 0−1
−1−0
x1 x2 x
= = −13
−1 0
x1 x2 x3
= =
1 0 1

1
X 3 = ( 0)
1
1 −1 1
Hence the corresponding Eigen vectors are X1 = ( 1 ) ; X2 = ( 2 ) ; X3 = (0)
−1 1 1
To check X1 , X2 & X 3 are orthogonal
−1
T
X 1 X 2 = (1 1 −1 ) ( 2 ) = −1 + 2 − 1 = 0
1
1
T
X2 X3 = (−1 2 1) (0) = −1 + 0 + 1 = 0
1
1
X3 T X1 = (1 0 1) ( 1 ) = 1 + 0 − 1 = 0
−1
Normalized Eigen vectors are
1 −1
1
√3 √6
1 2 √2
(0)
√3 √6 1
−1 1
√2
( √3 ) ( √6 )
Normalized modal matrix
1 −1 1
√3 √6 √2
1 2
N= 0
√3 √6
−1 1 1
( √3 √6 √2 )

MA8251 ENGINEERING MATHEMATICS II

ROHINI COLLEGE OF ENGINEERING & TECHNOLOGY

1 1 −1
√3 √3 √3
T −1 2 1
N = √6 √6 √6
1 1
0
( √2 √2 )

Thus the diagonal matrix D = N T AN

1 −1 1 1 1 −1
√3 √6 √2 1 −1 0 √3 √3 √3
1 2 −1 2 1
= 0 (−1 2 1)
√3 √6 √6 √6 √6
−1 1 1 0 1 1 1 1
0
( √3 √6 √2 ) ( √2 √2 )

0 0 0
D = (0 1 0)
0 0 3
y1
𝑇 y
Canonical form = 𝑌 DY where Y = ( 2 )
y3
0 0 0 y1
𝑌 DY = (y1 , y2 , y3 ) (0 1 0) (y2 )
𝑇

0 0 3 y3
= 0𝑦1 2 + 𝑦2 2 + 3𝑦3 2
Rank = 2
Index = 2
Signature = 2 – 0 = 2
Nature is positive semi definite.
To find non zero set of values:
Consider the transformation X = NY
1 −1 1
x1 √3 √6 √2 y1
1 2
x
( 2) = 0 y
( 2)
√3 √6
x3 −1 1 1 y3
( √3 √6 √2 )
y1 y2 y3
x1 = − +
√3 √6 √2
y1 2y2
x2 = + + 0y3
√3 √6
−y1 y2 y3
x3 = + +
√3 √6 √2
Put y2 = 0 & y3 = 0

MA8251 ENGINEERING MATHEMATICS II

ROHINI COLLEGE OF ENGINEERING & TECHNOLOGY

y1 y1 −y1
x1 = ; x2 = ; x3 =
√3 √3 √3

Put y1 = √3
x1 = 1; x2 = 1; x3 = −1 which makes the Quadratic equation zero.
Example: Reduce the Quadratic form 𝐱 𝟐𝟏 + 𝐱 𝟐𝟐 + 𝐱 𝟐𝟑 − 𝟐𝐱 𝟏 𝐱 𝟐 to canonical form through an
orthogonal transformation .Find the nature rank,index,signature and also find the non
zero set of values which makes this Quadratic form as zero
Solution:
1 −1 0
A = (−1 1 0)
0 0 1
The characteristic equation is 3 − s1 2 + s2  − s3 =0
s1 = sum of the main diagonal element
= 1+1+1= 3
s2 = sum of the minors of the main diagonalelement
1 0 1 0 1 −1
=| |+| |+| | =1+1+0=2
0 1 0 1 −1 1
1 −1 0
| |
s3 = A = |−1 1 0| = 0
0 0 1
Characteristic equation is 3 − 32 + 2 = 0
⇒  = 0 ; (2 − 3 + 2) = 0
⇒  = 0,1,2
To find the Eigen vectors:
𝑥1
Case (i) When  = 0 the eigen vector is given by(A − I)X = 0 where X = ( 2 )
𝑥
𝑥3
1 −1 0 𝑥1 0
⇒ (−1 1 0) (𝑥2 ) = (0)
0 0 1 𝑥3 0
x1 − x2 + 0x3 = 0 … (1)
−x1 + x2 + 0x3 = 0 … (2)
0x1 + 0x2 + x3 = 0 … (3)
From (1) and (2)
x1 x2 x3
= =
1−0 0+1 0

MA8251 ENGINEERING MATHEMATICS II

ROHINI COLLEGE OF ENGINEERING & TECHNOLOGY

x1 x2 x3
= =
1 1 0

1
X 1 = ( 1)
0
𝑥1
Case (ii) When  = 1 the eigen vector is given by(A − I)X = 0 where X = (𝑥2 )
𝑥3
1−1 −1 0 𝑥1 0
⇒ ( −1 1−1 0 ) (𝑥 2 ) = (0)
0 0 1−0 𝑥 3 0
0x1 − x2 + 0x3 = 0 … (4)
−x1 + 0x2 + 0x3 = 0 … (5)
0x1 + 0x2 + 0x3 = 0 … (6)
From (4) and (5)
x1 x2 x
= = −13
0 0
0
X2 = ( 0 )
−1
𝑥1
Case (iii) When  = 2 the eigen vector is given by A − I X = 0 where X = (𝑥2 )
( )
𝑥3
1−2 −1 0 𝑥1 0
⇒ ( −1 1−2 0 ) ( 𝑥 2 ) = ( 0)
0 0 1 − 2 𝑥3 0
−x1 − x2 + 0x3 = 0 … (7)
−x1 − x2 + 0x3 = 0 … (8)
0x1 + 0x2 − x3 = 0 … (9)
From (7) and (8)
x1 2 x3 x
= 0−1 = 0−0
1−0
x1 x x3
= −12 =
1 0

1
X3 = (−1)
0
1 0 1
Hence the corresponding Eigen vectors are X1 = (1) ; X 2 = ( 0 ) ; X3 = (−1)
0 −1 0

MA8251 ENGINEERING MATHEMATICS II

ROHINI COLLEGE OF ENGINEERING & TECHNOLOGY

To check X1 , X2 & X 3 are orthogonal

0
T
X1 X2 = (1 )
1 0 ( 0 ) = 0+0+0 = 0
−1
1
X2 T X3 = (0 0 −1) (−1) = 0 + 0 + 0 = 0
0
1
X3 T X1 = (1 −1 0) (1) = 1 − 1 + 0 = 0
0
Normalized Eigen vectors are
1 1
√2 0 √2
( 1 )( 0 ) (−1)
√2 −1 √2
0 0
Normalized modal matrix
1 1
0
√2 √2
N= (1 0
−1)
√2 √2
0 −1 0
1 1
0
√2 √2
NT = ( 0 0 −1)
1 1
0
√2 √2

Thus the diagonal matrix D = N T AN

1 1 1 1
0 1 −1 0 0
√2 √2 √2 √2
= (0 0 −1) (−1 1 0) ( 1 0
−1)
1 1 √2 √2
0 0 0 1
√2 √2 0 −1 0
0 0 0
D = (0 1 0)
0 0 2
y1
𝑇 y
Canonical form = 𝑌 DY where Y = ( 2 )
y3
0 0 0 y1
𝑌 DY = (y1 , y2 , y3 ) (0 1 0) (y2 )
𝑇

0 0 2 y3
= 0𝑦1 2 + 𝑦2 2 + 2𝑦3 2
Rank = 2

MA8251 ENGINEERING MATHEMATICS II

ROHINI COLLEGE OF ENGINEERING & TECHNOLOGY

Index = 2
Signature = 2 - 0 = 2
Nature is positive semi definite.
To find non zero set of values:
Consider the transformation X = NY
1 1
x1 0 y1
√2 √2
(x2 ) = ( 1 0
−1) (y2 )
x3 √ 2 √2 y3
0 −1 0
y1 y3
x1 = +0+
√2 √2
y1 y3
x2 = +0−
√2 √2
x3 = 0 − y2 − 0
Put y2 = 0 & y3 = 0
y1 y1
x1 = ; x2 = ; x3 = 0
√2 √2

Put y1 = √2
x1 = 1; x2 = 1; x3 = 0 which makes the Quadratic equation zero.
Example: Reduce the Quadratic form 𝟐𝐱 𝟐𝟏 + 𝐱 𝟐𝟐 + 𝐱 𝟐𝟑 + 𝟐𝐱 𝟏 𝐱 𝟐 − 𝟐𝐱 𝟏 𝐱 𝟑 − 𝟒𝐱 𝟐 𝐱 𝟑 to
canonical form through an orthogonal transformation .Find the nature rank, index,
signature
Solution:
2 1 −1
A=( 1 1 −2)
−1 −2 1
The characteristic equation is 3 − s1 2 + s2  − s3 =0
s1 = sum of the main diagonal element
= 2+1+1= 4
s2 = sum of the minors of the main diagonalelement
1 −2 2 −1 2 1
=| |+| |+| | = −3 + 1 + 1 = −1
−2 1 −1 1 1 1
2 1 −1
s3 = |A| = | 1 1 −2| = −4
−1 −2 1
Characteristic equation is 3 − 42 −  + 4 = 0

MA8251 ENGINEERING MATHEMATICS II

ROHINI COLLEGE OF ENGINEERING & TECHNOLOGY

 = −1,1,4
To find the Eigen vectors:
𝑥1
Case (i) When  = −1 the Eigen vector is given by(A − I)X = 0 where X = (𝑥2 )
𝑥3
2+1 1 −1 𝑥1 0
⇒( 1 1+1 −2 ) (𝑥2 ) = (0)
−1 −2 1 + 1 𝑥3 0
3x1 + x2 − x3 = 0 … (1)

x1 + 2x2 − 2x3 = 0 … (2)

−x1 − 2x2 + 2x3 = 0 … (3)
From (1) and (2)
x1 2 3 x x
= −1+6 = 6−1
−2+2
x1 x2 x3
= =
0 5 5

0
X 1 = ( 1)
1
𝑥1
Case (ii) When  = 1 the Eigen vector is given by(A − I)X = 0 where X = ( 2 )
𝑥
𝑥3
2−1 1 −1 𝑥1 0
⇒( 1 1 − 1 −2 ) (𝑥2 ) = (0)
−1 −2 1 − 1 𝑥3 0
x1 + x2 − x3 = 0 … (4)
x1 + 0x2 − 2x3 = 0 … (5)
−x1 − 2x2 + 0x3 = 0 … (6)
From (4) and (5)
x1 2 3 x x
= −1+2 = 0−1
−2+0
x1 x2 x3
= =
−2 1 −1

2
X2 = (−1)
1
𝑥1
Case (iii) When  = 4 the eigen vector is given by(A − I)X = 0 where X = (𝑥2 )
𝑥3

MA8251 ENGINEERING MATHEMATICS II

ROHINI COLLEGE OF ENGINEERING & TECHNOLOGY

2−4 1 −1 𝑥1 0
⇒( 1 1−4 −2 ) ( 𝑥 2 ) = ( 0)
−1 −2 1−4 𝑥 3 0
−2x1 + x2 − x3 = 0 … (7)
x1 − 3x2 − 2x3 = 0 … (8)
−x1 − 2x2 − 3x3 = 0 … (9)
From (7) and (8)
x1 2 x
3 x
= −1−4 = 6−1
−2−3
x1 x x3
= −52 =
−5 5
x1 x2 x3
= = −1
1 1

1
X3 = ( 1 )
−1
0 2 1
Hence the corresponding Eigen vectors are X1 = 1 ; X2 = −1) ;
( ) ( X3 = 1 )
(
1 1 −1
To check X1 , X2 & X 3 are orthogonal
2
X 1 T X 2 = (0
1 1) (−1) = 0 − 1 + 1 = 0
1
1
X2 T X3 = (2 −1 1) ( 1 ) = 2 − 1 − 1 = 0
−1
0
T
( )
X3 X1 = 1 1 −1 (1) = 0 + 1 − 1 = 0
1
Normalized Eigen vectors are
2 1
0 √6 √3
1
−1 1
( √2 )
1 √6 √3
1 −1
√2
( √6 ) ( √3 )
Normalized modal matrix
2 1
0
√6 √3
1 −1 1
N= √2 √6 √3
1 1 −1
( √2 √6 √3 )

MA8251 ENGINEERING MATHEMATICS II

ROHINI COLLEGE OF ENGINEERING & TECHNOLOGY

1 1
0
√2 √2
T 2 −1 1
N = √6 √6 √6
1 1 −1
( √3 √3 √3 )

Thus the diagonal matrix D = N T AN

2 1 1 1
0 0
√6 √3 2 1 −1 √2 √2
1 −1 1 2 −1 1
= (1 1 −2)
√2 √6 √3 √6 √6 √6
1 1 −1 −1 −2 1 1 1 −1
( √2 √6 √3 ) ( √3 √3 √3 )

−1 0 0
D = ( 0 1 0)
0 0 4
y1
𝑇 y
Canonical form = 𝑌 DY where Y = ( 2 )
y3
0 0 0 y1
𝑌 DY = (y1 , y2 , y3 ) (0 1 0) (y2 )
𝑇

0 0 3 y3
= −𝑦1 2 + 𝑦2 2 + 4𝑦3 2
Rank = 3
Index = 2
Signature = 2 – 1 = 1
Nature is indefinite.

MA8251 ENGINEERING MATHEMATICS II