Unit -1

Introduction- Matrices
Aset ofmm numbers (real or magmary) arrangedin the form ofrectangular array of mrows,
each owconsisting of an ordered set of nnumbersis calledanm xn matnx. (read asm by
n).

Anim 71 matrix is usually represented

as
Oy, Oy. Oyy
A=] ow were ue or A = [ Gi) men Where 1 sismilsjan
Om Omg Gey,

Basic Definition

Matrices

Matrix- Definition

A set of mn numbers (real or imaginary) arranged in the form of rectangular array of m

rows, each row consisting of an ordered set of n numbers is called anim x n matrix. (read
as m by n).

Anm xn matrix is usually represented as

M4, Gy2ww = =Ain

aan wena wae orA = [aij |mxn wherel <ismilsjen
fn Gna Ama
Types of matrices

1. Rectangular matrix: Ina matrix if the number of rows are not equal to the number of
columns, then the matrix is called a rectangular matrix.

a=| |
—-9 5 -2
4 3 1
2. Square matrix: In a matrix if the number of rows are equal to the number of
columns, then the matrix is called a square matrix.
2 6 5
B=/1 4 j
—7 7 2
3. Row matrix: A matrix having only one row is called row matrix.
la b el
4, Column matrix: A matrix having only one column is called a column matrix.

5.
H
Null matrix or zero matrix: A matrix in which each element is zero is called a null
matrix.
Definitions:

. Transpose of a Matrix: The matrix obtained by changing its rows to columns

and columns to rows is called transpose of a matrix.

3-4
3 1 7
raf 7
0 ar =|
_9 -4 0 -9
|
. Orthogonal matrix: A square matrix is called orthogonal if AA’ = I =

A’Aor AT =A}

. Determinant of a 2 by 2 square matrix A = [* | is a unique number ad-cb.

c d
_fa b . . .
Or, If A= [? a then its determinant is ,

|A| = |? | =ad—cb
c d
. Minor of an element in a square matrix: Let A = {a;;} be a square matrix. The

minor of an element a;; in Ais the determinant of the square matrix that

remains after deleting the ith row and jth column of A. It is denoted by M;,;.

. Cofactor of an element of a square matrix: The cofactor of an element in the

ith row and jth column of a matrix is defined as its minor multiplied by

(—1)'*/. If Ay is the cofactor of ayy, then

i+]
Ay = (-1) My;
8. Inverse ofa matrix: Let A be a non-singular matrix of order n, if there exists a

square matrix B of the same order such that AB=BA=l, then B is called the

inverse ofA and we write as A-?.

. a apa
The inverse of Ais A7? = at

Problems

ae ob
1. IA= f a | then find the cofactor of a3, in A + the cofactor of a5, in A’.
e ba
Solution: a3, = b

Cofactor of ag, =_ (—1)?*? lfa bl)__7,._

_ =—(ac—b*)22

a be
A=lc a b
boca
Cofactor of a3, = (—1)?*7 I; nl = —(ab—c?)

Ans= —[(ac — b’) + (ab —c*)]

Try Yourself

4 2 1 1 —-2 1
2. fA=|]-2 3 5|andB=|-3 3 2
3 2 3 3 1 5
find (AB)' and B'A' .Is (AB)’ = BA’.

Aset of mm numbers (real or mmagmary) arrangedin the form ofrectangular array ofmrows,
each mw consisting of an ordered set of numbers is calledanm xn matin. (read asm by
n).

Anim7 matrix 1s usually represented

as
Oy, Oy. Oyy
A=] ee ee ford = [Gila Where] Simi s<jen
Oy Omg Gan,

Problems
3. Show that A and B are orthogonal matrices.
1 2 2 leo.
2 1 aa-5f
au2
2

Solution: Show that AA? = | = A’A and BB" = B'B =]

1 2 -3
4. Find the determinant of matrix |: —1 7
2 4 —6

1 2 -3
Solution: Let A=|4. -—1 7
2 4 -6

detA-|4l-1|
_jaiq4|-l
‘7] _5|4
2\; {+6
7
3);
_a\|4
41
—1

=1(6 — 28) — 2(—24 — 14) — 3(16 + 2)

— —22-—2x —-38-—3x«18
— —-22+76—54-0

Hence A is a singular matrix. Hence A~? does not exist.

1 4 2
5. Find the determinant
of matrix] 2 —1 4
—3 7 6

deta-[4l-1]—"
_yatal[ol
f4)_,]2
[—4[ “7 42|
4
4,
2
|
-1

=1(—6
— 28) — 4(12— 12) + 2(14 — 3)
= —-34—-4x042x11=—34+22=—12
cos@ siné 0O
6. Ifthe nats ( snd cos@ ) is singular find @.
0 0 1
cos@ sin@ 0O
Solution: Given sin@ cos@ 0 =0
0 0 1

= 1(cos*4@ — sin*@) = 0
. . TT
=> cos*@ = sin?@
> cos@ =sind >@ =F

7. Find the inverse of A-[4 2]

detA= 124+12=24

hence A is invertible.

. [6 —4
Cofactor matrix of A-[) 3

-1_4djA 176 3
Aw = 4] 241-4 ]

102
8. Find the inverse of A=]2 1 O
3 2 1
Solution: |A| = 3

1-0 -(2-0) 4-3

Cofactor matrix of A= |—(0 — 4) 1-6 —(2 —0)
0-2 —(0—4) 1-0
1 -—-2 1
=|4 -5 -2

relat, 4a
1 -2 1

|| 3 1 -2 1

Rank of a Matrix- Normal Form

We reduce the given matrix to Normal Form and find the rank.
Normal Form

Nonmal Form of a Matrix

Every non zero matrix can be reduced

to any one of the form|[* alle): 0], [,] by applyinga

finit number of elementary

row and column operations. Hered, isthe unit matrix
of orderrand Ois
the zero element.

Example
11 1 2
Find he nk he mts by rein toma fom. A= 1 -3 -s|
3 -3 1 2

Solution:

io 0 0
A~|]2 -1 -1 -10]€; + C€,—C,Cs 7 C,—C,.C, = C,— 20,
3 6 -2 -4
10 0 90
A~ 10 -1 -1 -10 R, + R,—-2R,,R; +R, —3R,
0-6 -2 -4
0 oO
0 Oo Jos =e. coe, = 64-206
-4 —56

R; > R,—6R;

Hence rank (A) =3

Rank of a matrix-Cond...

Examples
For each of the following matrices, find a row-equivalent matrix which is in reduced row echelon form. Then determine
the rank of each matrix.

@a=| i,
1 3aI:

2 6 -2
mele 2 3}
2-2 4
(VC=]4 1 —-2).
6 -1 2

-2
()D=| 3 |.
1
()E=[-2 3 1).

Solutions

The matrix A has rank 2, which can be seen by computing

[2 3 Re 42R, F ‘ a he 2 ‘ R,-3Rp [: |
De oa oe .
2 2 0 8 01 01

The matrix B has rank 2, which can be seen by computing

2 6 -2]) #71 3 -1]) B3R [1 38 —1

— —>
3-2 8 3-2 8 Oo —-l11 11

aw [1 3 1] B 3h [1
— ——+
01-1 0

The matrix C has rank 2, which can be seen by computing

2-2 4) 12 fl -1 2) pu fl -1 2
4 1 -2/—3]4 1 -2}— Glo 5 —10
6 -1 2 6 -1 2] B® |g 5 —10

nn ft) 1 2 ]2mf1 1 2] aap ft 9 0

— slo 5 -10}] 3]o0 1 -2}—S]lo1 —2
0 0 O 0 0 0 0 0 0

Inverse of a matrix- By row transformations

Inverse of a Matnx- By Elementary Transformation

Elementary
row operations canbe made use to find the mverse ofa nonsmgular matix.

We canwnte A=I A.

Now we can apply row operations on Ato reduce it tl I, and each step apply the same
transformations
on nght side also. Finally we get J = BA, hence B is the mverse of A.

Inverse of a Matrix

Example

Find the nverse ofthe matnx by row transformations

1 1 3
1. IFA=} 1 3 —3) Find the inverse of A,
—2 -4 —4

Solution:

Consider A=IA

1 1 3 10 0
Consider
[A: I=] 1 3 -3:0 1 0
—2 -4 -40 0

1 1 0 °
0 2 “6 “1 1 0) Rs 3 Ry- RR; 3 R3+ 2Ry

1 1 0
1 1
o 1 o-ip 1/2 0 Ry 7 5RaRs aR

0 -1 1 1 0 1/2

io 6 3/2 -1/2 0
: 1 —3:-1/2 1/2 0| 3R,—Rz,R3 Ag+ PR
Oo 0 —2 1/2 1/2 1/

10 6 3/2 —1f2 0 1
0 1 -3:-1/2 1/2 0 |R3 3 —5Rs
Oo 0 1 -1/f4 -1/4 —1/4

bt 0-8 1 8603/2
o 1 0-5/4 -1/4 M315], = Rotts > Ro 3,
0 0 1-1/4 -1/4 -1/4
3 1 3/2 2 7
ices [5 -1/4 “i= i|-s -1 =
—1/4 -1/4 -1/4 -i -1
Homogeneous System of Linear Equations
Consider the system of linear equations

a,x+byy+e,z=0,

a,x+
by y+e,7= 0,

dgx+ hy y+egz=0

The system is homogeneous

and can be written as AX=0.

Solution of Homogeneous system of equations

Consistent and Inconsistent systems

Consider the system of linear equations

ayxt+byyt+e.z=0,

ax + by y+c,7=0,

gx
+ by yt eqz=0

The system is homogeneous

and can be writtenas AX=0_

Working Rule

1. Reduce the coefficient

matrix A to row reduced echelon form by elementary
row
transformations.

2. lfif rank(A) <3, will have mon trivial solution

3. Choose (n-r) parameters i@ 3-2 or 3-1 to get complete solution. Where n is number of

equations and ris numberof unknowns.

Here Ais called

the coefficient matrix, Xis the matrix of unknowns and is the column

matrix
of the right side.

Homogeneous System of Linear Equations

Consider the system of linear equations

ayxt+tbhbyyte,z=0,

a;x+b,y+e,2=0,

agx+ hb, y+egz=0

The system is homogeneous

and can be written as AX=O.

Example
Examine whether the following system has non trivial solution.

1. xt+dy—2z=0,2x-yt4z=00.-11y+14z=0
Solution: The given equation can be written as

AxX=0

1 3 —2 Xe
WhereeA=]2 -1 4 =|
1 -l1l 14 z
The augmented matrix is
1 3 -2
~ 10 -¥ 8 R, ~R,-2R,R, -R,-R,
oO —-14 16
13 -
~ 10 —7 §|R, + Ry—2R,
o oO 0

Rank(Aj=2<3(Number of unknowns)

Hence the system has non trivial solutions.

Let us write the system of equations again

x+a3y—2z=0,

Jy —82=0

>y=nrr
7

Substitute this value of z in first equation,

_ 24 22 _ 10
xa Et (Pea Tz

Substitute z=k{a constant or parameter) then x = — = ky = =k

Criteria for Consistency for a system of Equations

Consider
the system of linear equations

qyxt+bhy+ez=d,,

a,x + by te.z=dy,
y
agx + by y +e32 = dg

Ifthe system has at least one solution, itis said to be consistent.

If it has no solution it is said to be inconsistent.

Solution of System of Equations

Consistent and Inconsistent systems

Consider
the system of linear equations

aqxth y+e.z=d,,

a,x + by y teqz=dy,

a3x + by y+cgz = dz

ifthe system has at least one solution, it is said to be consistent.

If it has no solution itis said to be inconsistent.

Result: The system of three equations in three unknowns AX=6 has

(i) A unique solution if rank(A)= rank(AB)=3

(ii) Infinitely many solutions if rank(A) = rank(AB)<3

(iii) No solution if rank(.A)+ rank(AB)

We can extend this to m equations also.

Where A is called the coefficient matrix, X is the matrix of unknowns and 6 is the column

matrix of the right side. Here AB is called the augmented matrix.

Criteria for Consistency for a system of Equations

Consider
the system of linear equations

axtbhytaz=dy,

a,x+
bh, y+e,z= dz.

agx + by y +e32 = dg

Ifthe system has at least one solution, itis said to be consistent.

If it has no solution it is said to be inconsistent.

Problems

Examine whether the following system is consistent or inconsistent. Solve if possible.

1. xt+ytz=—3,2x+2y—z=-3,.x+y—-z=1
Solution: The given equation can be written as
AX=B
11 i
WhereA=|]2 2 —-1 x=[y].2=[3
11 -1 z
The augmented matrix is
11 i
AB=|2 2 -1 3
1 i -1
111 3
~]0 0 -3 -3| R, +R,—2R,.R, -R,—R,
o -2 -2
111 3
~|0 0 -3 —3|R, +3R,—2R,
00 0
Rank(A)=Rank(AB)<3(Number of equations)

Hence the system is consistent and has infinite number of solutions.

Let us write the system of equations again

xtyt+z=3.

—3z=-3

=3z= = =1
—3

Substitute this value of z in first equation,

ktyt1l=3,5>x+y=2

Substitute y=k{ a constant or parameter) then x=2-k

2-k
Hence the solution is X= k |jorx=2-k,y=k,z=1
1

Criteria for Consistency for a system of Equations

Consider
the system of linear equations

qyxt+bhy+ez=d,,

a,x
+ by te.z=dy,
y
agx + by y +e32 = dg

Ifthe system has at least one solution, itis said to be consistent.

If it has no solution it is said to be inconsistent.

Example 2 and 3

2, xtytz=4,2x4+5y—-—2z2=3,x4+7y
—7z=5.
Solution: The augmented matrix is
111 4
AB=/2 5 -2 3
17 -7 5
1 i141 4
~j0 3 -4 -5 R, ~R,—2R,,R; +R; —R,
6 -8 1
111 4
~|0 3 —4 —S/R, +f, —2R,
0 0 11
Rank({AB)=3 But Rank(A}=2
Hence rank(A)+ rank(AB)
Hence the system has no solution.
3, 2x—yt+3z=868,-x+2y+2z2=4,35x+y¥—-4z=0.
Solution: The augmented matrix is
2 -1 3 8
AB: -1 2 1 4
1
-1 3 8
~10 3 5 16 R, > 2R,—(—1)R,,.R, ~ 2R; — 3R,
0 5 -17 —24
2-1 3 8
~ 0 3 5 16 R, + 3R,—5R>
0 0 -76 —152

Rank(A)=Rank(AB)=3(Number of equations)

Hence the system is consistent and has unique solution.

Let us write the system of equations again

2x-—yt+3z=8------—-- cD
Ox+3y+5z=16—-—-—-—-—— (2)
Ox + Oy — 76z = —152 ---——-— @)
152
77g 2
Substitute the value of z in (2)we get
3y4+5x2=16 =>3y=100ry=2
Substitute
in (1)we get 2x —2+3x2=8 > 2x=40rx=2
2
Hence solution is X= |
2

Eigen Value and Eigen Vectors

Definition :

Let A be a square matrix of order n. if there exists a non-zero column matrix X of ordern x 1 anda

scalar A such that AX = AX, then X is called an eigen vector and A is called the corresponding eigen

value.

Eigen Values and Vectors

Eigen Values and Eigen Vectors
Definition :

Let A be a square matrix of order n. if there exists a non-zero column matrix X of order n x 1 anda

scalar A such that AX = AX, then X is called an eigen vector and A is called the corresponding eigen

value.

Working Rule to find Eigen Values and Eigen Vectors

Step 1. Write the characteristic equation of the matrix as |A — AZ| = 0

Step 2: Solve the characteristic equation. The roots of this gives the the eigen values say A,,A3,..A,

Step 3: To find the eigen vector corresponding to the eigen value(say) A,, put A = A, in the

equation|A — Aal|X = 0 and solve.

Eigen Value and Eigen Vectors

Definition :

Let A be a square matrix of order n. if there exists a non-zero column matrix X of ordern * 1 anda

scalar A such that AY = AX, then X is called an eigen vector and A is called the corresponding eigen

value.

Problems
Find the eigen values and the corresponding eigen vectors of the given matrix.

1. a=(5 A)
Solution: The characteristic equation of the matrix |A — AZ| = Ois
_fl 2 1 0 1 2 A 0
Now, A—ar= (3 )-4 2G 3-6 )

=" 22))
1-A 2

Hence |A —Al| =0
_ =|
1-A
3
27y_
5 4l=°

(1-A)(2-—A)-6=0

4-—34-4=053(A-4)441)
=0
=>A=4, A=-1

Hence the eigen values are

A = 4, A=-1

To find the corresponding eigen vectors

ForA=4

(A—alx =0 =(*5* a a)(5)=0

=(3 2)0)=°
=> -3x+
2y =0

3x —2y=0 => —3x+2y=0 = 3x

= 2y

Put y=k then 3x = 2k =x=2 k

2
Hence eigen vector is ¥ = F |
k
ForA=—1

(A-anx =0 =(07 2p) G)=2

=(3 3)(y)=°
= 2x+2y=0

and 3x+3y=0

=x+y=0
S>x=—-y
Put y =k thenx
= —k

Hence eigen vector is ¥ = al

6 -—2 2
2 |-2 3 —-1
2 -1 3
Solution:

The characteristic equation of the matrix Ais |A—Al| = 0

>} f= -2 3-a 2 “|- 0

2 -1 3-

(6 — A}{(3- A)(3 — A) — 1} + 2{-2(3 — A) + 2} + 2{2 — 2(3 —A)} =0

(6 —A)(9 —6A + A* —1) + 2(-6 +22 + 2) + 2(2 —6 424) =0

(6 —A)(8 — 6A + A?) + 2(2A — 4) + 2(-4 4 24) =0

(6 —A)(8 —6A4 a7) 42% 2(A—2)42x2(A—

2) =0

(6 —A)(8 — 62 + a7) + 84-2)

=0

(6-A-4O-2)+8-2)
=0

(A —2){(6—a)(A—
4) + 8} =0
(A—2){64
—24-— a7 +444 8} =0

(A —2){10 A—16 —a?7} =0

(A —2){a?
— 104416} =0

(A—2)(A—8)(A—2)
=0
(A—2)
= 0or(A—8) =0or(A—2) =0

A=2,8,2

Hence the eigen values are A = 2, A=8

To find the corresponding eigen vectors

ForA= 2

6-2 -2 2)
(-anx=0 =|" 3-2 =: |p]=c
2 -1 3-al'z

4 -2 2)]px
=|/-2 1 -1 [»}=0
2 -1 1dtlz

4x — 2y + 2z =0-—--—(1)

(2)
—2x +y —z=0———
(3)
2x —y+z=0-—-—-

(1) gives, 2x —-y+z=0—-—-—(4)

we observe that all three equations are the same. Hence we have one equation and three variables.

Hence we should have two parameters, let y= k, and z=k2

Then x= > (ki —kz) ky

1
3 (ki — kz)
Hence eigen vector is X = ky
ke
ForA=8

6-8 -2 2]
U-apr=o =['2 3-8 “1 |p]=e
2 —-1 3-—8Il'tz

—2 -2 2)]/%
=/-2 -5 -1 [»]=0
2 —-1 —5litz

—2x —2y+2z=0-——-—(1)

—2x —5y—z=0-——-—-—
(2)

2x —y —5z=0-—-—-—(3)
—2 —2 2
The coefficient matrix E —5 - can be reduced to
2 -1 —-

—2 -—2 2
fo —3 -3|, > (—2)R,— 2R,,R3 — (—2)R3+ 2R,
0 -3 -3

—2 —2 2
fo -3 -3 > R34+R,
0 0 0

Hence the equivalent system becomes

—2x —2y + 2z=0-—--—(1)

Ox — 3y —3z =0 —-—— (2)

Sytz=0 >y=-2

Put z=ktheny
= —k and x = 2k

2k
Hence eigen vector is ¥ = —k
k

Eigen Value and Eigen Vectors

Definition:

Let A be a square matrix of order n. if there exists a non-zero column matrix X of ordern * 1 anda

scalar A such that AX = AX, then X is called an eigen vector and A is called the corresponding eigen

value.

Example 3
1 0 0
3. A=|0 1 O
-1 1

Solution: The characteristic equation of the matrix Ais |A— AZ| =0

(1-4? =0 s=a=111
To find the corresponding eigen vectors

ForA=1

1-1 0 0 17
0 1-1 #0 b=
3 -1 1-ul
We get the system as 3x —y = Oor3x=y

Put x=k, z=k’ then y=3k. [ Note when we have 3 variables and only one equation, we have to

assign 3-1 =? constants or parameters]

k
Hence eigen vector is¥ = 3k
kK’

Eigen Values and Eigen Vectors(Cond...)

Example 3
1 0 0
3. A=|0 1 O
-1 1

Solution: The characteristic equation of the matrix Ais |A— AZ| =0

(1-4? =0 s=a=111
To find the corresponding eigen vectors

ForA=1

1-1 0 0 17
0 1-1 #0 b=
3 -1 1-ul
We get the system as 3x —y = Oor3x=y

Put x=k, z=k’ then y=3k. [ Note when we have 3 variables and only one equation, we have to

assign 3-1 =? constants or parameters]

k
Hence eigen vector is¥ = 3k
kK’

Cayley- Hamilton Theorem

Theorem: Every square matrix satisfies its characteristic equation.

Cayley Hamilton Theorem

Cayley-Hamilton theorem

Statement: Every square matrix satisfies its characteristic equation.

1. Verify Cayley Hamilton theorem for the matrix A = [3 A

Soln: Consider|A— AJ|=0

3 4 {=2

=> (@-AG@-y-6=0
> 4—5a4+¥-6=0

> #W-5A-2=0

Let A = A, then the above equation becomes

> AP - 5A -— 22 =O. 1)

#-aa=[3 S103 alelas 22

equation(1) becomes

> [is x2l-Li5 20]-[o 2

= [ ol=9

2. Verify Cayley Hamilton theorem for the matrix A = i > |

Soln: Consider|A— AJ|=0

Py" oa =0
=> (4-4) @-aA+4+1=0

=> 8-6A4¥4+1=0

=> 7-6A4+9=0
Let A = A, then the above equation becomes

> A -6449 =O... (1)

#-aa=(t oft Sle ol

equation(1) becomes

= [@ F)-lé wl+lo s
> [o ol=°

Cayley- Hamilton Theorem

Theorem: Every square matrix satisfies its characteristic equation.

Problems- Using Cayley Hamilton Theorem

100
1. IfA=]1 0 iow by using Cayley Hamilton theorem that A? = 2A? + A — 21.
010

102
2. IfA=!/0 2 1], usimg Cayley Hamilton theorem find A~*
2 0 3

2 -1 1
3. Using Cayley Hamilton theorem find AT*if A=|-1 2? —-1
1 -1 2

Cayley Hamilton Theorem(To find A inverse, A square )

Problems
3. Verify Cayley Hamilton theorem for the matrix
1 0
A=|0 2 1 ana hence find A.A and a
20 3
Solution:
The characteristic equation of the matrix Ais |A —AI| = 0
1-aA 0 2
=> i 2-A 1 =0
2 0 3-A
On simplifying
we get 4° —6474+7A42=0
We have to verify the theorem

Cayley Hamilton Theorem states A? — 6 A"

+ 74+ 22 =0

—6 0 4
Hence 21 = (A? —6A24+7A) = 2LA4=—a4?+6A-—71=|-2 1 1
4 0-
30
2
At=/-17 =
11 =
= 137
20-1

Problem 4

Using Cayley Hamilton theorem find A?,A* and A~* for the matrix
102
A=|0 2 | .
20 3

The characteristic
equation of the matrix Ais |A—AlI| =0
1-A 0 2
> 0 2—A 1 =0
2 0 3-—A
On simplifying we get A? -—6a°+7A+2=0
Cayley Hamilton Theorem states A? — 6 A? + 74+ 21 =0
A*=6A*-—7A+2I1=0
5
We have Ae =| 2
8

ate
A =|12 8 23
340 55
Again from (1) we have,
A = 64° - 7A - 21
= A‘ = 6A° - 7A? - 2A
A’ = 66 A? — 7A - 21] — 7A” - 2A
u

> A‘ = 29A? - 44A - 121

Substituting for A’, A and I, we get

89 0 144
A’ =| 58 16 101
144 0 233

Again we have from example 2,,

-3 0 2
A'=|-1% &
2 0-1
Now from (1) we have,
2I = —A? + 6A* -7A
2A a — Are 6A ~ Ol
uy

oaA a A or = 7a

4 (71 0-2 600 -21 0 14

2A77=| 0 -2 =1) +1060]@ 772”
-2 0 -3 006 14 0 -7

130 -8
A2=| AU -%
4 6 3&8

Problem 5
Using Cayley Hamilton Theorem, express A? — 27 A? + 65 A? as a quadratic
0 ot
polynomial ina, where =| 3 1 0
—Z2 1 4

Solution : The charactristic equation of A is

[A—
AL | = 0
“vw a 1
=> 3 1-a O =0
22.4 4-4
= - A) =0
—A(1-A) (4 - A] +3421
> Ni? — 50 44145
= 21 =0
=> 3 - 507+ 64-5 =0
By Cayley- Hamilton theorem, we have
A’ - 5A? + 6A -51=0
Consider
A? — 27A° + 65A? = A?A® - 542 + 6A - 51]
+ 5A* - 334° + 70A?
= 5A* — 334° + 70A? [By (1}]
= 5A[A® - 5A? + 6A - 51]
- 8A? + 4042 + 25A
-~ 8A + 40A? + 25A [By (1)]
- 8[A° - 5A? + 6A - 517]
+ 73A - 401
73A
- 401

Revision

Revision Questions
1. Define the rank of a matrix.
6
a —-1
2 Find ifthe mst omti A~ [ 3 1 fea
3 4 2 12?
3. Mind the eigen values and eigen vector ofthe matnx A= [ ab

4. Find inverse
of A= [5 5] using Cayley Hamilton theorem .
12-14
5. Reduce thematnx/? 4 3 | to echelon fon.
12 3 4

6. Are the equations x + 2y+22=L2ax+y+2=2,3x+

2y +22 =3,¥+2
= 0 consistent?

7. Verify C-H theorem

for the matrix A= G i

Groups
Properties of group is the basic concept of vector space.

Basic- Definition
UNIT
I Linear

Definition cae

Let G be anon-empty set and let + be a bmary operation on G:

#G*G46,

Or (a, bh—aszb.

Then (G, +) is a group if the followmg axioms are satisited:

(G1) Associative property: a+(bsc)=(a*eb)*cvabceG

(G2)Existence
of identity element: 3 e € G such thatas e=e+a=afor
alae G.

(G3) Existence of mverse: Wa € G there exists a@~+€ G such that

asat=a@tsane

If m addition to the above axtoms, the followimg axiom also holds

(G4) Commutative property: a*b=beava, beG

then (G.+) is calledam abelian group, or simply a commutative group.

lf the set G is finite, we define the order of G to be the number of elements im G, and denote
it (G|, Otherwise we say that G has infinite order.
Example (1) Consider the set R with operation ‘+’. ie. (R.+)

Closure property: Clearly ‘+*is a bmary operation.

For,a+beR, forall ab ER

(G1) Associative property: a+ (b+ c)=(at+b)t+ewa

bceR

(G2) Existence
of identity element: 3 0 € R such thata+ 0=0+a=aforallacR.

(G3) Existence of inverse: va € R there exists a7 + © FR such that

at a@t=at4+ a = Oora+(—a) =0=(-a)+a.

If im addition to the above axioms, the following axiom also holds

(G4) Commutative
property: a+b=bt+awabeG

then (R.+)is called 2m abelian group, or simply 2 commutative

group.

Sub- Space
Anon empty subset W of a vector space V over field F is called a subspace of V, if W
itself is a vector space over field F, under the same operations of addition and scalar
multiplication.

Subspace
Subspace
Anon-emptyset W ofa vectorspace V overa fieldF is calleda subspace of VjifW
is itselfa vectorspace over F, underthe same operations
of addition and scalar
multipheation
as defined im V.

Result:
Anon-empty
subset W ofa vectorspace V overthe field Fis asubspace
of V,ifand
onky if
a WiuyeWixtyeW

Gi) cinFandxeW>ax eW

Sub- Space
Anon empty subset W of a vector space V over field F is called a subspace of V, if W
itself is a vector space over field F, under the same operations of addition and scalar
multiplication.

Examples

Examples

1. Let ¥ = 2? be the vector space of all ordered n tuples( triplets) of real numbers ,
over the field of real numbers. Show that the subset W= {(x, 0,0)/x € R} isa
subspace of V.
Solution:
The element 0=(0,0,0)
€ W. Hence W is non-empty.
Leta =(«,,0,0)and G=( x,,0,0)be
my two elements of W.
Thana +8 =(x,0.0)+ ( x,00)=(4,+2,006W
Hence W is closed under addition.
Agam for any scalar
c€R
ea =e.(x,0,0)= (ex, 00)eW
Wis closed under multplication.
Hence W is asubspace
of ¥ = R3.
2. Show that the subset W=[(x,.4,.%,): ¥, +4, +", =0 jor
W=f(x y.z)ix + y+ 2 = 01 of vector space VW (2) is a subspace of V,(R).
Solution:

The element (=(0.0.0)

€ W. Hence W is non-empty.

Leta
=( x, V.2,Jand P=( x, y..2,) be my two dements of W.

Then x, +y,+2,=Oand x,+7,+2,=0

Thena + $= (xy y2)+ ( tp Yt)

= Cy tay Mate tee W
Smee, (x, +0) +07, +44 02,42)

=x,t+y,+2,+ x +y,+27,=0

Hence W is closed under addition.

Again for my scalarc

€R

aa=cCx,
yp 2 )=(cxcy,.czy ew

Since cx, + ey, +ez,=c(x2,+y,+2,)=0

W is closed under multiplication.

Hence W is asubspace
of ¥ = 2°.

Sub- Space
Anon empty subset W of a vector space V over field F is called a subspace of V, if W
itself is a vector space over field F, under the same operations of addition and scalar
multiplication.

Example 3
2
3. Let ¥ = be the vector space of all ordered n tuples( triplets) of real numbers ,
over the field of real numbers. Show that the subset W= {(x, y.z): 2x + Jy +2 = 0}
is a subspace of V.
Solution:

The element 0=(0,0,0)

€ W. Hence W is non-empty.

Leta=( x, V)Z,and @=( x,,y,,2,)

be my two dements of W.

Then 2x, + 3y,+2,=O0and

2x,+ 3y,+72,=0

Thena + f= (xp ¥pZ)+( Mp ¥p Fp) = Cay tay Wt Vy 2y+2,)

6
Smee, 2( x, +2",)+ 30 y,+¥,)4+(02,4+2,)

= 2x, +3y, +2, + 22,+3y,+2,=0

Hence W is closed under addition.

Again for any sealarc

€R

ja@=e.(0 4, Vy 24) = (ex, cvy.cz) &

Since 2c x, + 3ey, + cz, = c( 2x, + Jy, +2,)=0

W is closed under multiplication.

Hence W is a subspace
of V = 83.

Linear Combination

A vector expressed in the form civ; +C: V2+..+Cn,V., is called a linear combination of
the given vectors.
Linear Combination
Linear Combination and Linear Span of a set
Defirtion
Let V bea vector space over the field F and a,.a,,a@, ...a@, be any nvectors of V.
The vector of the formc,a@,+e,@,+ ¢20 +°" + ¢,a, Where C,,03,...,0, © Fis
called a linear combination ofthe vectorsa,,a4,...@,-
For example
Consider the vectors a, = (—1,3, —1).@, = (—1,2,3),a3 = (1.0.1) of ¥ = BR?
Then 2a, —3a, — ag = 2(-1,3, —1) — 3(—1,2,3) — (10,1) = (0,0,-12)
Isa lmear combimation of the vectors @,,03,03.
By choosing different scalars we get different lnear combinations
of a,, cy, a.
Definition:
Let Sbe a non-empty subset of a vector space V[F].The set ofall lmear
combinations
of finite number of elements of Sis called the lmear span of$ andis
denoted
by L[S].
That is
LISF fe, 0, +e,.0@,4+ cgagt+e,a,: 6,6 Fanda; €5,i=1,2,..0
Tfwe saya € L[S], then a = c,a,te,0,+ coagt + ty ty,
Problems
1. Express the vectors (3,5,2) as a linear combmmationof the vectors (1,1,0),(2.3,0)
(0.0,1) of ¥,(A).
Solution: Let
(3,5,2) = c, (1,1,0) + ¢,(2,3,0) + 02(0,0,1)
= (¢,.¢,,0) + (2e,, 3¢,,0) + (0,0, 65) = Ce, + 202,06, + 32,3)

6,+ 2c, =39,e,+3c, =3,0e, =2

Solving the above system and finding the values of (c,,cz,c,

Jas
e, =—-Le,=2,0,=2

We get (3,5,2) = —1 (1,1,0) + 2(2,3,0) + 2(0,0,1)

Linear Independent and Dependent Sets

1. A set of vectors is said to be linearly independent if civ; +¢2 V2t+...+C n V n=O implies
all c, =0

2. A set of vectors is said to be linearly dependent if there exist at least one non
zero c such that G.Vi +C2 V2t..+CnV n=O.

Definition
Lineay debendence and Lndependence
Definitve : A set Jd, player yon ot vectorw of a V's
VIF] ig Said to be Lineanty dependent if JF $Scalaxn
C1405, ---, ln EF » not all Zews Such that
Cr ky te Cady tees bh Cndn =O —OH

Definitve: A Bert 9A) ns yond of vectors ofa ws

VLE] is Said +o he haba Independent if

Ch hy eC edgy Ant Chen = 0
=> Cy em Cy eo Cp .0..

QB. Show that she Vectors

Ba cl te EO ac Ng can Rie
of he Vector Space Vy CR) ve Linearly todeflenalent.
Solution:
. Gf possible foe 2) Ba ia CHER, bet
Cy by Aly gee +tCn ly =D.

Ci 1,0, ...,0) + CalO,l,0,+..,0) +--+ Cnlo,a-..,1)=C0,0,..,0)

> (6) 1 O,-20) 00) Cay, +50) +o > F CO)0) -- -2Cn) =(0,0, -++,0

=> (Cry La, C3, +--+» Ca) =D

=> Creo, Cavo, ---, ned.

G1 Shous—thas the Set Sef Coy), Cho), lo, 4
__| 2 nae Mfenclenk \n yn 5 CA). ere a
| Coludens s =
eg T+£_tHoss' ble “bet,
2 err ren ipa bP Meals oth es en
—— —> Ce, +tC¢y SE ess Cay ¢ p- 2 2) => lo,o,n)

=> C4+¢,-c2 =o
aioe Oy =n
= OE ep => Cr =o
S

—i thus 3 Sealave 0, £05 mnt vem Sur that

—| 4 tn 2 =10,-1) =
Hence pe se+ S ue ppt Hnearly Laclepenclent.

Pesutts the cet $ Covariates)» Coe fary 1 (2,2 zal af

t a7 eT
Vechys of yes. Vg hoislinearly dependent iff
Mi MN a teen

r | Y, Yes Ys =O
Z| 23. 23

Ercnaple fs {tu ee ee oe oe

3 -3 ele 1E1a-S)
~1 (8416) — (2-6) +d
fe ! Al :
“es & MS dinearty
__|h d tadepencent . 7 e
Example: $2938, ¢4-F-e rey y—h
, 3 2 - —_
»——y_-g | = 134+ 9)-
3 C1414) +c; 414) —
a aot -l) = ye BxIS
4+ Ax IS =d .
_ fa’ Steg dineady dependent

A Basis of a vector space V[F], is the linearly independent subspace which spans the
whole space.

Definition
| Dagic Qud "Winaeutious
man .'
: a Neckoy Space VIF] 1S called
la basic nf Vif
tif je linpawlas independent Set.
iqy
IT
z.fB]-v 7

| Of bir Gon 2 byoie wf vie wIF] je wo Jinpavts

\| 3s
—pradependent subset whieh Shanc othe wrhole share

Ay

N Qolution che cet Cp Ze, C3 2a is Linearly Ande pendent

Sek -
| Fax .
1 Hoey oS

bo Cy Coy ,O) +43 CO,1, ,Dy+-- +Cn(o,0,--.,)=—0

ss (4, 7nd
23 ron Cn) =D
=> C=
bf
CaeCy=Ss
---—o

Also Quy Nechr {H,,mi,--- »%5) eVacQ)

| OU ag a _dinear Combinadicn
. of
Can he ex pressecl
Clenae nde nf CS.
112 CH, Na, 95,22 Und = 1 Oy te Hy Cat === — nen
| : L[sl=y-

| Boas a basis of VaR).

|

).
| Deeutte: _ : Si
bs In ann Abcaandioant wists anaes vce ——————
1 n+1 Clemente of Vo are Hin early dependent - ee
at
Li no Set of n- element Can—Span— Me

ee ee ee eel any sebot — ==

ee_VCF
dadependent vector te a hasis. _
—

z- Fer n_ vechrs of a-dimencional Vector space Vito he a

bagtc , it 16 Sufficient hat hey Span px dbat th
Jd d
Axe linea ly dndépenclent -

A ous veduced @c a
A: he. Noo wZWexD DiS o£

of a metrx aye Linearly todepencleat :

any mxn matrix TOW | equivalent to OA Fun

gO» Let Lhe
Chen the non zem ours af
on matrix:
foynn oo bacirs of the Subspace i an
LE

Bern yous of E js the dimensiss of

‘(| Ge nne of pon
matrix A-
pa cubepace Spanned— by the ynws pf the

Basis and Dimension

A Basis of a vector space V[F], is the linearly independent subspace which spans the
whole space.

Problems
if the Given set forms & hasis of Vac)
G.| fect
| 6 we Sti) Amant, C31
dT Botan: =
| Simee Vac) is_a_Nector Shace of dimension _Hree,it

is sufficiens if nee ieeisleg site given Get vf Jinear

-
—_____.Ladepemdleace—____—
| Conctdev log \ i ee

\ 32 4 -7] = I(do+F)—2 Cle +1) + (3-12) —_

3 l & +b — eh

'_ Given set are nearly tndehenclend

Hence : @o+ is 2 b Sic oft VoCR)

Cet Rips: a. basis of 3 In case S

Q No she folloratng

_is not a hasan deternsine the olimensian and oa hasis _

of Ihe Subspace gNanned ha,
S=1C) 9,3), ve 0), (= 1,2)4
Golutien: Coneider he matyix-
1 oa Z

A-\|3 1 » Here |Al=o ae

l-a 1 J
&_is linearly dependent Subset of 2+
Bur S ve not-a basje. -
To find he ddinaeusion & basis of dhe Subs pace Spanned

by the Vector S$ —
eS. ri a 3&3| 2s > R»-3R,

3 > Rca
g)
, 2a 3 |
-5& -4q |g,
Concider
geah 11 ele
3] | |o
0 | 3

| er el
~w
a “Se +4
| 0 °o Oo
—
T

das +u.0 Anon Zorn — WorasS

: he MG+ANK
1. Almension of
Also the non zero rows *01m a
fo 1 Cte3), Co,-s,=U4 o eee
basts of =the subgnare Shaaned be, S- a
is the

Examples - Basis and Dimension

Examples
8. Find she Dacis and She olimensian. ot he. SubShace

Spann ed. she vectors a eer

(2, A,),-Cl, =p) , Cl @,1) 2 ( o-2, 1) in Vs3fQ)_
ipnt
hes 8 2 469,4,9),01,-10) ,01,2,15 ee SSS

_f2a2 4+ a] fi 2& 1
A_s tT a] TT +1 i 2.52.
i jg a rs a &
lo 2 1) lion = td
- 2 17]

2
° —— a a
= J elt 8 -
° “s =}
| A_ basis of the subgpace is a ==
__| {he 5 bagels ne a

+f bah teh th ep
8. Shon has the: SEE

| Sowa a_ basis of Vector Shace Vo of au axe ruatrices

Over RR. ——
ee >
_ Ransider
Cy !oo |
(iJ tal t oO 1 jee br AN +C rs i

=> pe fe | e5
CO3 Ca)
=> Cc =Cyg=l3 =Cy =o

CO fe Sineacvly indehenden}-
Jo 2:7 Ss geaerSet dhe tshole Space
tat ane py ey be arbitnany
She Matix Can be Uniquely Lotiken as

fa b\|=—a/t ay gk ht jae eo o}pd{o 4

d) lo 1) lo o) UI oo} oT
Le

aac oe ap ee
bom binatihe of the vectors of &

ee ee a a
ae Seof all Axa matrices -
§. Fiad +he basis and dimengion of the Subspace Spanned
dhe subeet
+ ote (5 2) 4 ett
— of -the Vector Space of all dxa matrices over Q.
_ Qolirtreny ! —
Consider the ma+atx __
[| -S -4 a 1 -5S - &/2,3aR 5-8,
A =I 5 =A = i ——=3 Roar>P,-alk,
a- -S o
=T S57 or i Ra PRy- 2,

] =F -—h _ .
fas Oo &6 3< 3] Rg> a R%3-FRe
oO oO 0 o Rgz>rRt+tR
5 4 2

== .. dimensisn of the Subspace =.

iy Basis of the Subspace Spanned bss S

= oc re eer
Linear Transformation

Let U and V be two vector spaces over field F. A mapping T: U- Vis a linear
transformation if

TX + yJ=T(X) + T

T(c x)= T(x)

For allx, yin U and cinF

Linear Transformation

Linear transformations are mappings from one vector space to another, which
possess a certain property called linearity.
LetU and V be two vector spaces over the same field F.
The mapping T : U > V is said to be linear
transformation, if
T(a+ 8) =T(a)+T(S) Va,f EU
Tle.a) =c. Tle) We E€F,a EV

= Zero map: Isthe mappmgT : U + V defined by

T(a) =0 We EU which maps every vector of the
vector space U to zero vector 0 of the space V.

= Identity map: Is the mapping J, : V - V defined

by J,(@) =a Wa €V which maps every vector
of the vector space V to the vector of space V

Linear Transformation

Let U and V be two vector spaces over field F. A mapping T: U- Vis a linear
transformation if

TX + yJ=T(X) + T

T(c x)= T(x)

For allx,
yin U and cinF

Problems
I Secesh desis =. a _
I. Define FT: VaR)
> Vg CQ) ba
TlH1,Ns,
Ws) = lo, Hy,H9) - Show that Tis a Linear
transfomahen-
Boluticn : fo s4 9 ds a linear transformation toe She
prpve shat _
Gla+p) =Tla+ TCP) » Va, Pp € Vee)
GLC +2) =C-dtey ȴ ce 2 wevsCR)
Let ay pe V9) he Ar bitsrany

a= CH, Y ys 21) » Pera, Yar a)

Consider <+/9 = CH tHe, Yea, 2+ Zw
ae T lx +fs) = (o, YstYa, Zit 2a) [by Givens detnitioe]-
—— = 00, Y,,2:2 +00, Ys, Zry
_t = T(«) +9(p)
Consider, Tle) = 9 COM, CY 4,024) =(0,04,,¢2,) =
= Clo, Y,,2,)=¢:Tes)
Hence 7 is Linear

Properties of Linear Transformation

We have two properties , stated as theorems. We prove them in this lecture hour.

Theorems

* Properties of linear transformation:

Theorem 1: 1 T:U - Vis alinear transformation, then

i) T(O) = 0" where0 and O’are zero vectors of U andV
respectively.
ii) T(—a) = —T(a) Va eu
i) Teyay + C202 + cee ee Hoyng ) =
Tay) +e:T(a2)+ ..........+¢,T (a)

Theorem 2: If f,,2, ....... By, be any basis of vector

spaceV and ay, d@%,... ...., 0, be any m vectors of the
vector space W’, then 3 only one linear transformation
T:V 3 WwithT(9,) =a; fori =1,2,....,m

Properties of Linear Transformation

We have two properties , stated as theorems. We prove them in this lecture hour.
Problems

lL Fupd the f: R222 Ruch that —

Aineay +eansformadion
Pt) =C11) aad +o,1) =(4,2)- —
_. Botutten: =
Che vectors (1,0) » (011) 4foenr the basis elemen
/ ;
of R&.
: het tu ,4 5) € 22 be arbhite. .
— Gheo C4} = Kh) + YCo,t)
Define - £ : IR» QX bay 7
_
—_||_ =a.sp Pe sa ee
i =H fig Eah —————————
a Ca
Sate a = (1-4 2 ay) — Sed

: { aber: ¥3-= Conny otaty ) Meyer?

-o):!| Find he Binecy sransfommahben f: 2% > 8" Bich oe
aeee ee ee ee ee
Solute + p
| Are element C1) ye af 2 fore —— : au
fasts
| pf 2%. Ben

het atm Coty 2B be arbitrans _

—_Let+ LA sy) 0, C11) FO ac—1,1) ee a
! ples eps SS

—f— a Cp ne Oe @ ay
| Cy + a — Y —(a)

| Qo turn toe Get

| a0, = H+ of [ Ada eel =

=\ (= X+Y
ae

@O-D => ac,= Y-e > cae yx

a

~~, y J me HAY (1,1) + Yee (1,0) =

aL a - a
—=>Ployi= ty Fl + Yon fem) =

Matrix of a linear transformation

Let B = fay, @ 2) wu... @,} be an ordered basis of the

vector space. Then a € V can be uniquely written as
= yO + 20gto vee tly Oni Cf EF
The scalars
cy, €2,........C,, are called coordinates
of the
vector@ w.r.t the ordered basis B. The n-
tupple (c,, C2, ........€,) 1s called the coordinate vector
of¥ relative to the basis B.

Let T: V —@ W bea lmear transformation whereV and

W are vector spaces of dimensionm and n, defined by
T(a;) EW ,i =1,2,.....m. ThusT(a,) is a umique
linear combmation of ff,, 85,........,
8 Which are the
ordered bases of W’.
Thus T(a,) = ayy By + Gay Bat ocr eee Ont Br
Tay) = ayy By + Gaz Bet oe eevee Ona Boy

T (Gm) = Gim By + Gam Bot we vee ee Gam Bn

The transpose of the coefficient matrixis called the

matrix of the linear transformation and denoted by A;

yy Agee im
G21 Az7eevweee 2m
Ar= cone cone

Qni Onz+ an nxm

Matrix of a linear transformation

Problems

- |. Fine Hoe nnatesx o dhe Linear tyang-foamatisa_-

—}—T: Vac Rios Ve cp)
——dlepined. They ap = Cry», Bu- y)-
lo. t4t gkandayd bases
Soludirin hes “Ape st 254... Ps fst be the std awe
oF Vgc) 8 Vg CR) peep. pS as a
CC, =(1,0) , @y
= lo,1) =

ee ae ree - fy =(041,0) , be elve 48; . ——————

Nou ,_ a
TleyeTUuM=Cyiygethar bards
Tes=Tloyi) =(1,0,- -D=i-fyro-farenf,
= fe ! ")
4 1 6
Ss ~
ol. Fane
, the raat xii t_op lhe Li lnear_tyansternatian in the
—_—.@bove_eranap Selative
le,fo the basis
ITEM gee PARAS ee eo Ene ge Lee (0,014
————_f}_ Matar _& VgC2) resp: a
—_—Soluten: Given The yy = CK +y 1 Sn =y ) eee
EE 22) eS
TOZ,1) = (4,2 ,2) ———

(29,1 2) £(4,2,2) Can he Apres sed ag linear ell tN gi fy

ole bases
le Cea =f mee + Caltol Im + CoCijo,p)
eaAPO eg ah ar ppl le |)
| CQ+Catls =

Ci. + @o- =)
| QC, =e 2 a

| eee GE ede Me een I

__| => Cs) a = = £ C= ee

hi « Ear, \ yor) — AL tod, 1) — Ke Ly i; 0.) = | (Lob)

ee eS 2 eS
——J So ee ee — =

|. Aq = “! -o | a. - =——— eee

Linear Transformation of a matrix

Linear Transformation from a matrix of the transformation

i “J 7 = =a :
L + Gist the mie A=

Seruaiboensdilits 11 Nol Vs5td

> 9) -
“te
—— Reladuse to the pases BO, & Ba Given bay : - =
0) By 2 By , the Standard bases op V3CP) 2 V2) mpl
Ease eet C123),gE Ae aes eft OLbald

Define —> Vis CR)

T? V3 (Q) >
by Ci n,0) = Ld),5) + 3¢0,1) Ay
att
T(o,},n) = ip tem baton) = Flor) 7
Tln,o,1) = a(n) + alo st) = Cd,o) =
"Yo coeTot 2) 2 VotQ)
3
=
= do) ,2 _
4 oT Tipo) +ZTlo,o0,1)
= U1,3) e+Yyt=-li+Zzl9,o)
-(y-y
¥. +22, 3. du+y y) |
——~»s=._' : ; ___
1 the 2 quized trans fom adie

Wy A, Lu ta,0 50,04, i= ts peti, -$

GT: V9(Q) Sv, (AR) Le Adfinedt bay

VT Ct,i,1)= $011)
+ 25-1) = (4, —o2)
ThU,aABeri(pN+il1,-)) = oe “a)
eel as LNrocr-1) = (a, 2)
y Quy, Z) 2 N52)
(ory, ZB) = 2, Cyd + Cal 10d 3) + 6301,0,0)

Y= C4+Aac,) 4
0 = k =e
z = Ci + 3ey a =e Hy

Cre Yodl, = =
You ¢ 2-4) S44 = 22
=a “ 69 = H— 4 0) = ee
pf ar Spiele ct _ re
7 1H y = Sy matz Gyan $24) L192, 3) nay 4230409)
“Te ifiaad TL ipl)
att eA) sp Coay) Tl pad, 2) __
+(H=Ay +7) TL 1,0,0)
ee ai) Ch p=) + C2> se¢ inter cm
= a a

= (int ey—67, aX = By +hz)

__ Ais the required trans fomnahen — __

Problem 2

a. Givin A= b ~ tind the Jinear +transtownaten

T! Vg tR) >ValQ) relate te she basic -

Belt, iy) , BaHAtiyiu, Cy, lo,o,10)
—_—Jolutiorn:—
— Dene 1 Myf ag th) ty
——1 FC pt iee (Ck bt = 1)—ldo,0o, = 1,0)
TOS Se APN A TY che BT ac) Wt aya 63,16) ==
het 0904) We LQ
—— ON Y= CN) pp Call) -
aa ee Ly = C4 = DL

eee 1 — ly ) ay fearn +t ya=00 See

ee ee Ly) =< C4 7+ ie 0
“-TLUY y TOI)
| = H+ + Yon Thr)
ol

— AX+Y C1,1,.0) + Y— C3, b)

ol ol

wen Je = a + SCYuo), X+y4 Yeu 4 Gh

i
ae ol a ot a
es Tee
i =n, Yy , Sy = 3a)

oe Bee lb dhe required trans fomatcn -

Range and Kernel of a Linear Transformation

Range and Kernel

Range
and Kernel of a linear transformation

= Range: LetT: V - W bea linear

transformation. The range ofT is the set
R(T) ={T(@)| @ EV}.
i.e R(T) is the set of all T- images of the elements
of V. Clearly R(T) CW.
= Kernel: LetT : ¥ — W be
a linear
transformation. The kemel (or null space) of T
is the set N(T) ={T(a@) =O|a@ EV}
where0 is the zero vector
of W’.

Le N(T) is the set of all those elements inV whose

image under T is the zero vector of W. Clearly
N(T) CY.

= For the identity linear map J: V — V

the range is the enitre space V and the kemel is
the zero subspace of V.
= For the zero lmear map T : V — W defined by
T(a) =0Va@ EV, the range R(T) = {0} =
zero space of Vand the null space N(T) = V.

Range and Kernel of a Linear Transformation

Rank and Nullity

= Rank of linear transformation:

LetT : V - W bea lmear transformation. The

dimension of range space R(T) the is called the rank of
the linear transformationT and denoted by r(T).

ie | r(T)=dim[R(T)]

= Nullity of linear transformation:

LetT : V — W bea linear transformation. The
dimension of null space N(T) the is called
the nullity of
the linear transformationT and denoted by n(T).
te | n(T) = dim [N(T)]

Rank Nullity Theorem

Rank Nullity Theorem

Let T : V — W bea linear transformation and V be a finite dimensional vector

space. Then
r(T)+ n(T)=d[V]
ie = «d[ R(T)]+ d[ N(T)] =a[V]
Proof:

Let V be of dimension m

That is d{V}=m

Let d{N(T)}=n That is nullity=n(T)=n

Since N(T) is a subspace of V, we have n < m

Let Bi= {@1, @2, @3 .... @n } be a basis ofN (T).

As B1 is linearly independent in N(T) so is in V. [Since N(T) is a subspace of V]

We shall extend this set B: to a basis of the space V.

Let this basis of V be

Bo= {01y, @2, M3 --.- An Br, Bo, Bs, }

Clearly

nt+ts=m _ (Since d[V] =m)

Now

T(@,), T(@2), .... T(@y),T(B,), -...T(B,) € R(T)

But T(@,) = T(@z) = --- = T(@y) = 0 since a; € N(T)

Consider the set S = {T(f1,),T(Bz,) -.-.-T(Bs,)}

We shall show that the setS of s vectors of R (T) is a basis of R (T).
That is we will shaw that

(i) 5 spans R(T)

For
Since Bz is a basis of V, it spans V.
Hence the set {T(a,), T(a2), .... Tay), T(Bi,), ---- T(Bs,) } spans R(T).
But T(a,) = T(a,) = --- = T(a,) = 0 [sincefay, az, a3 ..-.@,, } be a basis of N (T)].
Hence S spans R(T).
(u) ~— ‘Sis linearly independent
Consider c,T(B1) + coT(B2) + ---. +¢,T(B.) = 0
> T(c,B, + C2B2 + “= C5Bs,) =0

> (c1B1 + coB2 + --- C5Bs,) EN(T)

Hence there exists scalars d,, dz, ....d,, such that
CyB, + c2Bz +---¢sBs, = dyayt+dz a2 + ---+dy Gy [Since Bi= {0y, O2, G3 -... py }

is a basis of N (T).]

=> 8, + CoB. + = C6By, — dya,—d,z a, +--—d, a, =0

Since B> is the basis of V, it is linearly independent.

Hence § is linearly independent.

Thus S is a basis of R(T).

d[R(T)] =s = rank of the linear transformation T

n+s=m >rank + nullity = dimension of V

Hence the theorem is proved.

Problems - Rank Nullity Theorem

Problems
bg ct
l.ihet 7: V—>ll be a J! detined

Vie ene Vedy a

eee la an
4
So luboo'

Ale,)= Tlhio,n) ge C5 Ny 50) we ob

= D = = =
=
é =T (9,0,1) =at3
——ne_Knots Petite ks} generates P07).
Consider A Ty {>
[1 -\ +8
Lo oT

IAI= 1-1) -101) = -a =o.

—y-
ea ae, g
ey Jot, sera ee st is dinearly dodepende ui Qurd shenee ——s

a ee ee ee ee ee re
= (Mt
As , UH= Ms, 29,
+ 2) re
< Liq ,.= 3, (x, + AS - Mi = Ms AK, +3) JH Me, gen} —

To_tsind NID
tet oto, 2) 2alo, 0,n) ee
=> sry, ay, da+2) =(0,5,0)
| Say so, osyen, aaazeg —
=> — Hay, HEY, AM Se Boo

== ai ee ul Y =O ,2z=05 . a

tot. NCT) @antoains— Zeer Vectoxy

#—~*_Nlit) = {Co,n,n) %
—Alse dD nay] =0 x eo
—}* —Lank = dim Tv
+ Nattity.- = Z+o 2
=

Problems - Rank Nullity Theorem

Problem 2
—_Aa- | Gaiven Tiv, (2) SV nck) deined bay s_= _.
—_ Nt (m
2) =y,
(y=, Y= 2). Verify A.N + Thenvene.
Rolutico!
— T11,0,0) =(-I,n) =o, ———
tT Ey i Sef
ty iat: — a ee
= —T(0,0,1) = (0, -1) =e
ut. RIT) == Span i-, pea yetad =2 pte P) 21___-_—
—____
—_ Pr
Lonsider
Batis=\
—_—_————_-e-
07
=\—1—_,;
=i bo
'
ore eee
74
a
fos
—og
=

_ . en > _
_ ~ fs]. Ra rR eR 1 o]
oO =i |
SSS ae ee RES RR
. 2

——_ = fst ft Ratko. 7% —

I a a) es eon aE Se
! ALR] =2 aes eg
£17) = Lubspace Geto erated by tw 1), do 1) "9
____. fo__find Nia eee —
I} —_ Considler T(oa,y,%) = Co,0) as
—___ => Yy-m, y=2) =(0,0) a
= => = She Yo 2 2o- ad

a f => “a = ——a moe: — = . =

oo NVUT)et eae a ——
——|_& Basis. gq Nia =tdrioy ae
——— Nlullizty =niq) = |

Eigen Values and Eigen Vectors of a Linear

Transformations
Eigen Values and Eigen vectors

Let T from V,(R) into V,(R) be a linear transformation on an n dimensional vector space over the field
F. Aneigen value ofT is ascalarA € F such that T(a@) = Aq, for some non zero vector we V. Every
vector satisfying T(a) = Aq, ford € F, is called an eigen vector of T.

Let T: V,(R) = d
V,(R) be defineby
T(1,0)= (1,2) ; T(0,1) = (4,3)

(i) Find all eigen values of T and the corresponding eigen vectors.
(ji) Find the matrix P such that P~ AP in diagnal, where A is the
matrix of T.
Solution : The matrix A of lineasr transformation is

A (3 | ; T(x, y) = (x + 4y, 2x + 3y)

x € R isaneigen value of T if there exists a = (x,y) € V, (R) such
that
T(x,y) = A(x, y)
sing (x + 4y, 2x + 3y) = (Ax, Ay)
=> (1-Ax+ 4y,2e+B-Ay=(, 0) .. (1)
=> (1-Ax+4y=0, 2+@G6-Ay=0
This system will have no trivial solution if

2 3-y|
1-~A 4 =79 2 2_ P-M-5=0
a

=> (A-5)(A +1) =0

> 25, ==1
. eigen values are 5, — 1.

Let 4 = -1, then we have again from (1)

2x + 4y =0 | > x+%=0
2x+ 4y = 0
This equation have the independent solution as (2, - 1)
.. (2,-1) is the eigen vector corresponding to eigen value - 1. Now

r(t
consider the matrix

Introduction- Calculus of Variations

Introduction
Functional:

Let °S* be a set of functional ofa variable ‘x’ defined overa

interval (a,b).Then a function which assigns to every element in ‘S’ a
unique real number is called a functional.

Variation of a function:

Consider,f = f(xy). y= 2
, here “x° is arbitrary function yx) and their derivatives
yp’, we get a unique real number as value off = f(x,y, ¥).

Now, let y be changed from ytoy +handy’toy

+k
From Taylor's theorem,
’ a a 1 a

Sy2eu f+ ,
ko

: « a a . * ©
fluyt hy +k) -fluyy)§ ag + ke +higher derivatives

Af =df = ne + ke (Here Af is an increment

in f dueto
increments/#t and kin y andy’)

Applying
for y,
dy =A
— 4, 3 +k
ay
Similarly applying for 4,

a a
df = aay + ay!

Also we can use, k =en(x) and k = en'(x) , where¢ arbitrary small and
n(x) is arbitrary fimction.
Thus from equation (1),

f(x.yteny +en)— flxy.y) = en(x) Sh + en'(x oa

Variation
of f Sf is df = a by + by Looe. 2)
Equation (2) is variation of df of f

Euler's Equation

Euler's equation:-

A necessary condition for ](y) = io f(x,y,

y )dx to be an extremum
: af
isthat £— 4od sh)
fary _
=0 oo. a)
Equation (1) is the Euler’s equation for solving the Variation problem.

Euler's Equation

Euler's equation:-

A necessary condition for ](y) = Es f(x,y,

y )dx to be an extremum

is that 24
ay
— sa) O.....
Equation (1) is the Euler’s equation for solving the Variation problem.
Alternative forms of Euler’s equation:

1) We have ‘f as a function of x, y, y then,

af af, oF oF
dx ax) ay) Taye uae iC)

£ (yf) = y (2) 44 ofLyy

é(v Ve of) a 2)

2) We have ‘fas a function of x,y, y, thus “is alsoa function

of x,y,y-
ie. 5 = (x, ¥,9) -.G)
. a *

of _
a'r __y PrFf “oF
of SP SP Zhao. 4
_,2fug

Solution of Euler's equation under particular cases:-

a) When /f is independent of ‘x’,

iect = 0, then the Euler’s equation is given by,

OF _
f-y ay" = constant

b) When ‘f is independent of “y’,

In this case 2 = 0, then Euler's equation is given by,

£(2f\-9
dx \ay

a
*f = constant
ay

c) When /f is independent of »’,

a = 0, then Euler's equation is given by,

af
ay

d) When ‘7 is independent of both ‘x’ and ‘y’,

af
3 _ 9 And af
a_ 0

z
*f ~oAnd e
Qf =0
axay" ayay’
Then Euler's equation becomes,

y iF$=0.....0)
a

Euler's Equation
Euler's equation:-

A necessary condition for /(y) = J f(x,y,

y )dx to be an extremum

is. that af 2aa (%(ar\_=O... ()

Equation (1)is the Euler's equation for solving the Variation problem.

Problems

E xample 1 ‘$ Find the extremal of the functional I=

as foex2y? dx

|
Solution : Gj 6 ' , ;
n= Given fayyyy )=y +x7y'2 Here f does not contain y explicitly, therefore use the
particular form of Euler’s equation of wees 1a 2x2y' at
oy
1_K-1 dy k-] 7
— I 2x2 oF a x2 or dy -()5 on integration

fay aK=1
a 2 fl
Ja 5 => y=- k-1( [-+} 1 +e

or y-(5*
1-kV1
[Fhe or
¢ -
ae | ( where c=")

*2
Example 2 : Find the extremal of the functional I= \o? +y'?4+2ye* dx
*
Solution : Here f = y* + y'* +2ye* contains x,y,y' explicitly

use the general Euler’s equation =) a =0

‘dy dx\ dy’

i.e Qy +2e")--—(2y')=0 or 2y+2e*-2y"=0

d
or y'"-y =e* or (D*-1)y =e* where p=

AEis m7-1=0>m=#1 ye* *

«. C.F=y,=c+c7¢
I e* xe” x ~~ ee
PI.= y)= e* = x, —— = OVERVo pt Yp =Cye FOe +
Yp D2 -1 2D 2 a 2
"4
Example 3 : Find the extremal of the functional I= ov? - y'*—2ysinx )dx under the
0

Conditions y(0) =0 ; 1(5] =0

2
Solution ; Here f = y? ~ y?-2ysinx contains x,y,y' explicitly

“.. use the general | Euler’

Eulerss equation eh
dy dk e3,
eq tion ——-—-——| dy |~0

+ 2y"=0
i.e (2 ~2sinx)-(-2y')=0 or 2y-2sinx
: 2 d
or y'+y =sinx > (D* +l)y =sinx, where De

AFis m?+1=0O>m=Hi + c2sinx

«. CF=y, =ccosx
x
; x sinax =-——cosax
Now P.I. = y.= sinx = ——cosx [ 5
D? +1 D* +a 2 2a
x ee ,
F which is the general solution
~ Y=YotYp= Cy Cosx + Cp sinx — so
To determine the constants c, andc, apply the given end condition

=> 0=c,cos0 +7 5in0-$-c0s0 = c; =0 and

y(0)=0; 1-0
t-te Tt 3 x
Tt =>c,=0 «.. theextremalforlis y= “panes
O=c, ae tee sine

*2

Example 4 : Find the function y which makes the integral I= [a +xy'txy” )dx an
*1
extremum.
the
Solution : Here f(x,y,y') =1+ xy'txy” which does not contain y explicitly. Therefore use
k-x . dy _k 1
particular form of Euler’s equation a =k => x+2xy'=K or y'=
¥
Kd. 1 k ¢l 1
dy =—.——— onintegration |dy=— |—dx-— |dx
eis. 2 Joh J
“ _k 1 _ x _k
aS Eee T or yr eigea Ste crs

Geodesic on a plane
We find the shortest path connecting two points on a plane.
Geodesic on the plane

Example 1 : Geodesic on a plane :

Prove that the shortest distance between two points in a plane is along a straight ling
joining them OR Prove that the geodesics on a plane are straight lines.

Consider the XOYplane. Let y=y(x) be a curve joining of two points A(+),¥;) and Bix>.¥>)

in the XOY plane. An element of arc length in a plane is ds= (dx)? +(dy)” and the total
length of any curve between A and B is given by
"2 ds *2 dy 2 *2
S=_ fds 4
fee dy
fy+() dx =_ [ity 2 dx
x xy xy
Xo

i.e S=l= fyi +y"? dx. Now, we are required to find that particular curve y=y(x} for which I
ad|

is minimum. We have f(x,y,y')= l+y" . f does not contain x or y explicitly therefore

corresponding Euler’s equation is y''=0, on integration twice w.r.t.x, we get y = ar+b. But this
is the equation of a straight line. Hence geodesics on a plane are straight lines.

Geodesic on a sphere
We derive the shortest path connection two points on sphere.

Geodesic on a sphere
Find the geodesics on a sphere of radius a or show that the shortest arc joining two points
on a sphere is the minor arc of the great circle through them.
Solution : The equation of a sphere of radius ‘a’ with centre at the origin is given by
2
x+ y- +z°
2: 2
=a". In the case of spherical polar coordinates (r, 6) the elementary arc length

is given by ds=y(dr)* +(rd0)* +(rsinOdd)* ,since for a sphere r=a,

we have dr=0 and
2
oe ds=1/(ad6)? + (asin dé)? = a2(do
)? +a” sin? Q(do)? or ds= afiesn
e(#) 6
Suppose A(a,6;,,) and B(a,8,0)) are two points on the sphere r=a then the are length
B 6, 2
between A and B is given by S= fas =a | 1+sin? (=) AB csivkk (1)
A 8,
Now the geodesic is the curve on the sphere r= a for which s is the minimum.
' Here f =/1+sin? 0g? which is independent of and hence the corresponding Euler's
i 2 t HOF 1 2 st ea
equation is ——~ = constant 1. —~———————sin" 0.29'=c squaring and cross multiplying,
do 2y1 +sin? Oy?
we have (9)? sin* @ =c?(1+ sin? 80" ) or (sin* @-c? sin? 9)07 =¢?
(y= ¢? _ c* cosec’6 deen ccosec8
sin? @(sin?@-c”) sin? @—c? sin? Q-c2’
2
=> o . ccosec"O ccosec20 — SS rr
ccosec*@
aoe ee
Vi-c cosec?® 1-c?(1 +cot 0) Vi-c2) ce cot? 0
putting ccot@=t wehave ccosec20d0= —dt

{ao=- f dt oe fie cokt| pela tae Gmi ccotO I (2)

a-e)-2 vi-c2 Vi-¢?
The constants c¢ and b can be determined by using the conditions that 6 =6; and 0=6,
A and B respectively. The curve (2) is an extermal of the functional (1). The curve (2) is the
of radius r=a. Now to show
curve of minimum length between the points A and B on the sphere
_1| ecot® 4b
that the geodesics on a sphere are arcs of great circles, consider 9=COS 2
‘ l-c

c
=> come = cos(—b) => kcot@=cosdcosb+singsinb c 7 5
vi-c?
l-c

si ‘sinb\. ...
> cot = Peas + sing > cos={ SS") cosdsin 6+| sae sin sind
\

or acos0= A asindcos) + BlasinOsind) 8) { oe k

We have on the sphere x + y? +27 =a’, x=asin@coso, y=asinOsing, z=acos6

(3) becomes z=Ax+By which represents a plane through the origin, which is the centre
of the sphere. Thus the equation z=Ax+ By together with x + y? +z? =a” represents the
A
curve of intersection of the plane with the sphere. This curve is a great circle passing through
great
and B. However though the points A and B on the sphere there are two arcs of the same
arc of the great circle gives
circle one minor arc and the other major arc. Obviously the minor
the minimum length between A and B. This is the required geodesic on the sphere.

Geodesic on a Cone

Find the geodesics on a right circular cone.

along
Consider a right circular cone of semi vertical angle @ , with vertex at the origin and axis
the z-axis, Its equation in spherical coordinate system is 9=a. Elementary arc length ds

spherical coordinates is given by ds=y(dr)? + (rd0)? + (rsin@d)?

; 8=a=> dd=0 «. ds= (ary? +(rsinado)? =,|1+sin? a( doy
Since dr
r

2
Now S will be minimum if S= [,/1+sin” ar?” dr is minimum
~ f= inbevtGe. wa < os peginmienn 1
Here f=yl+sin° ar’ g* js independent of , hence the corresponding Euler's equation ™
=constant i.e 1
wae OF 2h'=c or sin? or?rg'=cy i +sin’ oF ‘O°
2y1+sin aur2g?
Snag ~~ snptyng. we get
2
sin’ a r4g? = 7(1+sin ar 292) > sin’ o r4o'?-c? sin?
o 70! = 7
; 2
= sin?a rq? (sin? or? ~¢?) =c* = 7 adapta eames
sin? or? (sin? or? -c*)
_ c . 4d c
“af
= fea at ee
singe sin* ar“ —c } dr. sinar{ sin? as? —e |
4

1 ccosec?o
do=ccoseca . dr = dr
sin orvr2 =e cosec™a. ryr? —c? cosec"a

ccosec zo
On integration [a0 ——-
ryr? cttemsont

=> 0=ccosec’a 2 .———-sec

1 -| ————— SS =~see
x

ccoseca sooty ier ~9?

: -}({ rsina : _{ rsince

=> 0=cosecasec { ve => (—b)sina= sec ( |
et c
> rsina=csec|(— b)sin a] or r= ccosecatsec|psin a —bsina]
or r=Asec|(sina)o + B] (where A=ccoseca, B= —bsina)
which is the geodesic on a right circular cone. A and B are constants and can be cetermines
using the condition that curve passes through A(1, &, ,) and B(t.c.0>)

Geodesic on a Cylinder
Find the geodesic on a right circular cylinder.

The arc length element ds in the cylindrical polar coordinates (R, ©,z) is given by

ds=(dR)? +R2(do)? + (dz)?

If the are lies on a right cirentar cylinder with the z-axis as its axis and radius of cross section
‘a. the "2
a. then we have R =a so that dR=0 and hence ds=\a2(do)? + (day? = a’ , + (= dep
\ dip

The length of a curve connecting two points A and B on the cylinder R= ais given by

_ fas dz
S= dig | a (2) dQ cece (1) where , and ¢y are the values of @ at A and B

reapectivey. Now the geodesic is the curve on the cylinder R = a is the curve for which S is
minimum.

Here f= a’ +z" which does not contain and.z explicitly. (independent and z dependent)
Hence the corresponding Euler's equation is z''=O on integration twice we get
Z=CO+CQ 0.0... (2) where cj and cy are constants. The curve (2) is an extremal of the
functional (1). The constants c; and cy in (2) can be determined by using the end conditions
$=, and@= 0) at A and B respectively. Thus z=c)$+c represents the geodesic on the
cylinder R =a. :

Surface of Revolution

Surface of Revolution
Mini mal sSurface of revolution : enerates a surface of
when rota ted abou t a line 8°
which
Prove that catenary is the curve
minimum area. passing between two
surf ace of revo luti on by taki ng som e curve
Solution : Suppose we form a then is to
y2) and rev olv ing it abou t x-axis. The problem
fixed end points A(),y))an d B(x), p of the surface is
the surface area is minimum, The area of a str
find that curve for which xX

[yv! +y" de
t { 2

2nyds = ny Sax
j = 2nyyl +y'?. dv and the total area is 27
*|
Euler's equation is
the correspondin g
Here f = yyl + y” does not contain x explicitly, hence y” )-yy" _ (it
' y ey? c
wher c is constant ic yyl[ + y” -y' y : 2y'=c >
or =c
f-y'—
QV1+y' l+y’
dy :
2
> 2
y?=c*(I+y") ory
2_yr-e
35
or y —=c
Il+y"
2_ 2 (a
=e or dy way
il or dy vy
dx c y?-c? c
c

On integration f oy =+far+K i.e cos’ ht L=7+K

; fy2—¢2 c c C

or 2 xcosh{ x4 k] me y=coosi( ***) where a= Ke

c c c
a minimum. And this is Catenary.
Thic is the curve which makes the surface area of revolution

Hanging Cable
Hanging Cable

Show that
A heavy cable or chain hangs freely under gravity between two fixed points.
the shape of the cable (or chain) is a catenary.
Solution : Let A(x;,y;) and B(x2,y2) be the two fixed points of the hanging cable. lf ds is an
elementary arc length of the cable and p the density (mass/unit length) of the cable so that pds
is the mass of the element. Let g be the acceleration due to gravity, then the potential energy o!
the element m.g.h is given by (Ods}.g.y where x-axis is taken as the line of reference.
i
“. Total Potential Energy of the cable
is given by T Jods) wey dx
A
‘ ‘
= ds . " /
=> Tsp fy = Px fv ty? da Aix yy) Bir. ¥2)

" y \

Here f =a yy + y'" 2 which

ich isis ind
j ependent é
of x, hence the
corresponding Euler's equ
ation is
of
i-¥ooee icyy +y? ~ Lp ae
2yi+y'

y(l+y" )-yy’ cae or y=

eatin or y 2 scX(+y?) = y? =

or y=
dy? =<? ww j ;
a. a on integration
bape c
d ]
f y “cae i.e. cosh"! = 24K or & =cosh] —~+K
ia aet c Cc

+
. y= coost( =**) where a= Kc. This is catenory

Brachistochrone Problem

Shortest Time problem

Brachisto chrone Problem: (Shortest time problem)
This well known problem is to find the curve joining two points along which a particle falling
from rest under the influence of gravity travels from the higher to the lower point in the Least
time. (the name Brachisto chrone is being derived from the Greek words brachistor which
means shortest and chronos meaning time).
The problem is : Find the path in which a particle in the absence of friction will slide from
one point to another in the shortest time under the action of gravity.
Let the particle start from the point O, initially at rest and
slide along the curve OQ. Let P(x,y) be the position of the
particle at any time t and Let OP=s be the arc length. It is
given that the particle is moving only under gravity (with out
friction). Therefore from the principle of work and energy
that the work done in moving the particle from O to P is equal
to the difference between the Kinetic energy at P and at O.

ie mgy= sm? -( (mgy = Potential energy at P)

2
But y a x ngyae m( or —=
~= ay being the velocity of the particle at the point
dt 2 t
Px,y),

He nce the time taken by the particle to move from

O to Q is given by
1 \ it 4

T= fa- [Sars ae oo Vi eytas

0 0 4
i ° ds ¢ ‘ vey

“dy. Now we need to find y(v) such that Tis minimum.

oh i
isis independent ‘ i
of ; x, hence the corresponding duler’s
Euler's equation ;js
equation
which

a adie
fe eal =()
gaye =c¢ ie

1 Fah kt
CI ty Fe
vyvl + y~ Vy 1+ y

ie Jy vi ty"= le va (say) on squaring we have

Cc
a-

yi+y")=a or a alae or y'= Ao

i.e a lil Y= |
are - dy on integration fax= he wer Le x= [ —_ay+k
dx yo any
6 61
. dy=a. 2sin—cos—.—.
asin ote ° dO
weet
Put y=asin e2 x wie. dain
Cy =a Gal reteCSS . dd

asin— ris :
*; x= f 2 -asin eos 40 a fsin? 40 =a [- wer BK ie x=—(6-sin6)+K
bone 5 2 2 2 2

Also we have when 6=0, y=0 and we must have x = 0 also since (0, 0) is a point on the curve.

This give K=O. Thus x=5(0~sin®) ; y=asin? == ~ 088)

Taking <= b, x=b(@-sin®) ; y=b(1-cos6)
These equations represents the curve cycloid.

Quick Revision

CALCULUS OF VARIATION

Functional:

Let ‘S’ be a set of functional of a variable ‘x’ defined over a

interval (a, b).Then a function which assigns to every element in ‘S’ a
unique real number is called a functional.

Variation of a function:

Consider, f = f(xy.) ==
, here ‘x’ is arbitrary function v(x) and their derivatives
y’, we get a unique real number as value off = f(x, y,y)-

Euler’s equation:-

A necessary condition for J(y) = fe f(x, y,y) dx to be an extremum

1
. of ad (af) _
is that 37S su) =0 —— (1)
Equation (1) is the Euler’s equation for solving the Variation problem.
Solution of Euler’s equation under particular cases:-

a) When ‘f is independent of ‘x’,

ect = 0 , then the Euler’s equation is given by,

r of
f-y ay constant

b) When ‘f is independent of ‘y’,

In this case a 0 , then Euler’s equation

is given by,

£@)-0
of
ay = constant

c) When ‘f is independent ofy’,

or = 0 , then Euler’s equation is given by,

ar
ay 2

d) When ‘f is independent ofboth ‘x’ and ‘y’,

ar ar
5, 7 9 And ay 9

arr arr
axdy-— 0 And ayay 9

Then Euler’s equation becomes,

y"4=0.....@
w Gf
ee = 0, then y’’*=0 whose solution is of the form,

Derivations

Geodesics:

A Geodesic ona surface is the curve of shortest length joining two

points on surface.

a) Geodesic on a plane:

“To show that the shortest distance between two points in a plane
is a straight line”

ie = m(a constant)..... qd)

On integration, y = mx+c..... (2)

The constantm andc can be determined by using the end conditions that
the line passes through the points A(x,,y,) and B(x3,y2)

b) Geodesic on Right Circular Cone:

The element of arc length in spherical coordinate system

(R, @,@) Is given by,

— (ds)* = (dR)* + R7(d0)* + (Rsin 6 d@)*

sina.(@ — b) = sect , is a Geodesic on curve @ = @

d) Geodesic on Right Circular Cylinder:

“To show that geodesic on a right circular cylinder is a circular

helix”

The element of arc length in cylindrical coordinate system

(R, 9, z) Is given by,

+ (ds)* = (dR)* + R*(dO)? + (dz)*
The equation of geodesic on cylinder is given by

2=C10+ C2
Hanging Cable

Total Potential Energy= fr pgy ds

da
Here we use f —y’ a = constant

We get catenary y = c cosh( “)

Minimum Surface of Revolution

Total Area=[- 2ny ds

Here we use f—y' 5=of = constant

We get catenary y = c cosh( =)

Shortest time Problem

Time, T= f, dt = fc dt
— — dx
0 ds dx
ds

a
Here we use f — Ws = constant

We get the curve as cycloid.