J.R.

INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 1

1. Simple variational problems

Variational Calculus is the branch of mathematics concerned with the problem of finding a function
for which the value of a certain integral is either the largest or the smallest possible. That integral is
technically known as a functional.
Variational Calculus deals with functionals. Briefly, a functional is a function of a function. The
difference between a functional and an ordinary function can be appreciated in the table.

Input : argument x Output : function

(independent Function value y (dependent
variable) operator variable)

1.(a) x 
 f 
 y  y( x)  f ( x) Functions

Input 1 : argument x Input 2 : function

(independent y  y ( x ) (primary Functional Output : functional
variable) dependent variable) operator value J (a scalar)

1.(b) x 
 f 
 y  f ( x) 
 J  J  y   J  x, y 


Input 1 : argument x Input 2 : function

(independent y  y ( x ) (primary Functional Output : functional
variable) dependent variable) operator value J (a scalar)

1.(c) x 
 f 
 y  f ( x) 
 J  J  y   J  x, y , y '

 
Input 3 : derivative of primary dependent value
y '  dy / dx Functionals

We shall consider the functional

b
I  y  x     F  x, y, y ' dx ; y  a   y1 , y  b   y2
a

where x is independent variable, y is dependent variable and y ' is first derivative of the dependent
variable.

Three forms of Euler’s equation :

F d  F 
(i)   0
y dx  y ' 
d  F  F
(ii)  F  y'  0
dx  y '  x
F 2F 2F  2F
(iii)   y'  y '' 2  0
y x y ' y y ' y '
 2
Note : For a functional extremal can be unique, infinite or does not exist depending on the boundary
conditions.
Def. C n  a, b  is the set of all functions having continuous nth order derivative in  a, b 

Some results from calculus :

1. The distance between two points ( x1 , y1 ) and ( x2 , y2 ) along a curve y  y ( x ) is given by
x2

I [ y ( x)]   1  y2 dx .
x1

2. When a curve joining P1 ( x1 , y1 ) and P2 ( x2 , y2 ) is revolved about x-axis, the area of the surface of
x2

revolution is given by S [ y ( x)]  2  y 1  y '2 dx .

x1

3. Time taken by a particle in sliding from ( x1 , y1 ) to ( x2 , y2 ) along a curve y  y ( x ) is given by

x2
1 1  y '2
T [ y ( x)]   dx where g is gravitational constant.
2g x1 y
4. Cycloid is formed during rolling of any circle. Let a be the radius of that rolling circle then
parametric equation of cycloid is x  a (  sin  ) , y  a (1  cos )

Type : Integrand contains y only

1
 y3 
Example : Find the extremal of the functional I  y ( x)     y   dx subject to the conditions
0 
3 
y  0   1and y 1  1.
y3
Solution : F  x, y , y '   y  , y  0   1, y 1  1
3
F d  F 
Euler’s equation :   0
y dx  y ' 
d
 1 y2   0  0  y2  1  y  x  1
dx
y (0)  1, y (1)  1 , both the boundary conditions are satisfied.
Hence the extremal is y ( x)  1 .

Exercise 1.1
1
 y3 
1. Find the extremal of the functional I  y ( x)     y   dx subject to the conditions y  0   0 and
0 
3 
y 1  1.
b
 y3 
2. Find the extremal of the functional I  y ( x)     y   dx subject to the conditions y  a   y1
a 
3 
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 3

and y  b   y2 . What are the conditions on y1 and y2 so that given functional can have an extremum ?
b
 y3 
3. Find the extremal of the functional I  y ( x)     y   dx .
a
3 
2
4. Find the extremal of the functional I  y ( x)    y 2dx subject to the boundary conditions y 1  0
1

and y  2   0. Also tell that extremal curve is a minimum curve or maximum curve.
1
5. Find the extremal of the functional I  y ( x)    y 3dx subject to the boundary conditions y  0   0
0

and y 1  1.
1
6. Find the extremal of the functional I [ y ( x)]    2e y  y 2  dx subject to the conditions y  0   1 and
0

y 1  e.

Answers

1. No extremal 2. y1  y2  1 or y1  y2  1 3. No extremal
4. y  x   0, minimum curve 5. No extremal 6. No extremal
---------------------------------------------------------------------------------------------------------------------------
Type : Integrand contains x and y only
3
Example : Find the extremal of the functional I [ y ( x)]   (3x  y ) y dx subject to the conditions
1

3 9
y (1)  , y (3)  .
2 2
3 9
Solution : F  x, y , y '    3 x  y  y , y 1  , y  3 
2 2
F d  F  3
  0  3x  2 y  0  0  y ( x)  x
y dx  y '  2
3 9
y (1)  and y (3)  , both the boundary condtions are satisfied.
2 2
3x
Hence the extremal is y ( x)  .
2
 4

Exercise 1.2
3
1. Find the extremal of the functional I [ y ( x)]   (3x  y ) y dx subject to the conditions y (1)  1 ,
1

9
y (3)  .
2

2
  
2. Find the extremal of the functional I [ y ( x)]   y (2 x  y )dx ; y (0)  0 , y   
0 2 2
1
3. Find the extremal of the functional I [ y ( x)]   (5 x  2 y ) y dx subject to the conditions y (0)  0 ,
0

5
y (1)  .
4
4. What changes in the values of y  0  and y 1 should be made in above problem so that it may have
an extremal ?

Answers
5
1. No extremal 2. y  x 3. No extremal 4. y 1  
and y  0  is unchanged.
4
---------------------------------------------------------------------------------------------------------------------------
Type : Integrand is of the form M ( x, y )  N ( x , y ) y '
1
y
Example : Find the extremals of the functional  (e  xy ')dx that satisfy the boundary conditions :
0

y (0)  0 , y (1)  0
Solution : F  x, y , y '   e y  xy ' , y (0)  0, y (1)  0
F d  F  d
  0  ey   x  0
y dx  y '  dx
 e y 1  0  ey  1
Taking log both side, we get y  x  0
y (0)  0  y (1) , both the boundary conditions are satisfied.
Hence the extremal is y ( x)  0

Exercise 1.3
1

  xy  y  2 y 2 y 'dx , y (0)  1 , y (1)  2

2
1. Find the extremals of the functional
0

1
2. Find the extremals of the functional I [ y ( x)]   ( y 2  x 2 y ')dx ; y (0)  0 , y (1)  a
0
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 5
x2

3. Test on extremum the functional I [ y ( x)]   ( y  xy ')dx subject to boundary conditions y ( x1 )  y1

x1

and y  x2   y2
b
4. Test on extremum the functional I [ y ( x)]   ( y  y ')dx .
a

1
5. Test on extremum the functional I [ y ( x)]   x y y ' dx subject to boundary conditions y (0)  0 and
0

y 1  1 .

Answers
1. No extremal 2. If a 1 then the extremal is y  x , and if a  1 then no extremal.
3. Problem is meaningless and so no extremal. 4. No extremal 5. No extremal
---------------------------------------------------------------------------------------------------------------------------

Type : Integrand contains y ' only

x2

Example : Find the extremals of  1  y '2 dx subject to the conditions y  x1   y1 and y  x2   y2 .

x1

Solution : F  x, y , y '   1  y '2 , y ( x1 )  y1 , y ( x2 )  y2

F d  F  d  y' 
  0  0  0
y dx  y '  dx  1  y '2 

y'
  c2  y '2  c 1  y '2 
2
1 y '
c
 y '2 1  c   c  y '2  c  y' c  y '  c1 (say)
1 c
 y ( x )  c1 x  c2
y  x1   y1  y1  c1 x1  c2 …….(1)
y  x2   y2  y2  c1 x2  c2 …….(2)

y1  y2 y y 
By equation (1) and (2), we get c1  , c2  y1  x1  1 2 
x1  x2  x1  x2 
y2  y1
 y  y1   x  x1 
x2  x1
 6

Exercise 1.4
x2
2
1. Find the extremals of  1  y '( x ) dx subject to the conditions y  x1   y1 and y  x2   y2 .
x1

2. By using Euler’s equation, find the curves passing through (0,3) and (4,11) such that its length
between the given points is shortest.
1
3. Find the extremal of  y ' dx subject to the conditions y  0   0 , y 1  1 .
0

1
2
4. Find the extremal of  y' dx subject to the conditions y  0   0 , y 1  1 .
0

Answers
y2  y1
1. y  y1   x  x1  2. y  2x  3
x2  x1

3. Problem is meaningless and so no extremal 4. y  x

---------------------------------------------------------------------------------------------------------------------------
Type : Integrand contains x and y ' only
1
1
Example : Find the extremal of the functional I [ y ( x)]  x
2
y '2 dx ; y    1, y 1  2
12 2

1
Solution : F  x, y , y '   x 2 y '2 , y    1, y (1)  2
2
d
0
dx
 2 x 2 y '  0

dy c
 x2 y '  c  x2 c  dy  dx
dx x2
x 1 c1
 y  c1  c2  y  c2
1 x
1
y  1  1  2c1  c2 …..(1)
2
y 1  2  2  c1  c2 …..(2)
By equation (1) and (2), we get c1  1 and c2  3
1 1
 y ( x)  3  y '( x ) 
x x2
1 1
1 1 1
and I  y ( x)    x  4 dx   2 dx   x 1 1   1  2  1
2

1 x 1 x 2
2 2

Value  1
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 7

Exercise 1.5
x2

1. Find the extremal of the functional I [ y ( x)]    xy ' y '2  dx .

x1

2
2
2. Find the extremals and extremum value of the functional   x  y '
0
dx ; y  0   0 and y  2   4

4
3. Find the extremal of the functional I [ y ( x)]    xy ' y '2  dx ; y (0)  0 , y (4)  3
0

x2

 y '1  x y ' dx
2
4. Find the extremal of the functional I [ y ( x)] 
x1

5. From among the curves connecting the points P1 (1,3) and P2 (2,5) , find the curve on which an
2

 y ' 1  x y ' dx
2
extremum of the functional can be obtained.
1
1
  x  y '  dx
2
6. Find the extremal of the functional that satisfy the boundary conditions y  0   1,
0

y 1  2.

x2 2
 dy  n
7. Find the extremal of the functional I [ y ( x)]   x   dx where n  1 , passing through the fixed
x1  dx 
points ( x1 , y1 ) and ( x2 , y2 ) .
2
y '2
8. Show that the extremal of the functional I [ y ( x)]   dx , y (0)  0 and y (2)  1 is a parabola.
0
x
x2
y '2
9. Find the extremal of the functional  dx
x1
x3
2
x3
10. Find the extremal of the functional  dx with y (1)  0 and y (2)  3
1 y '2
4 2

11. Find the extremal of the functional

 y ' dx ; y (0)  0 , y (4)  4

0
x 3

2 2
x
12. Find the extremal of the functional t 3
dt with x(1)  3 and x(2)  18
1

x2
1  y '2
13. Find the extremal of the functional I [ y ( x)]   dx
x1
x
 8
2
1  y '2
14. Show that the curve through (1,0) and (2,1) which minimizes  dx is a circle.
1 x
b
15. Find the extremal of the functional I [ y ( x)]   x 1  y '2 dx
a

b
16. Find the extremal of the functional I [ y ( x)]   x 1  y '2 dx
a

Answers
x2 x2 x2  x
1. y  c1 x  c2  2. y  x  ; value = 2 3. y 
4 2 4
c1 4
4. y   c2 5. y  7  6. y  x  1 .
x x
y1  y2 1 n y2 x11n  y1 x21n
7. y  x  8. x 2  4 y 9. y  c1 x 4  c2
x11n  x21n x11n  x21n
x2
10. y  x 2  1 11. y  12. x  t 4  2
4
2  y  c2 
13. x 2   y  c1   c22 14. x 2  ( y  2)2  5 15. x  c1 cosh  
 c1 
16. y  2 c1 x  c1  c2
---------------------------------------------------------------------------------------------------------------------------
Type : Integrand contains y and y ' only
1
Example : Find the extremal of the functional I  y  x      y 2  y2  dx, where y  0   0, y 1  1
0

Solution : F  x, y , y '   y 2  y '2 , y (0)  0, y (1)  1

d
2y   2 y '  0  y  y"  0
dx
D2 1  0  D  1
 y ( x)  c1e  c2 e  x
x

y 0  0  c1  c2  0  c1  c2 ……(1)

y 1  1  c1e  c2 e 1  1 ……(2)

By equation (1) and (2), we get c1  e  e 1   1 .

Now divide both side by 2, we get
 e  e 1  1 1 1 1
c1    c1 sinh1   c1  , c2  
 2  2 2 2sinh1 2sinh1
1 sinh x
 y ( x) 
2sinh1
 e x  e x  
sinh1
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 9

Exercise 1.6
x2

1. Obtain the Euler’s equation for the extremal of the functional I ( y ( x))    ay 2  by '2  cyy ' dx
x1

x2

2. Find the extremal of the functional I ( y ( x))    y 2  yy ' y '2  dx

x1


2
   
3. Find the extremal of the functional I ( y ( x))    x 2  x 2  dt , x(0)  0 , x    1
0  2
x2

  y '  12 yy ' 16 y  dx
2 2
4. Find the extremal of the functional I [ y ( x)] 
x1

1
5. Find the extremal of the functional I [ y ( x)]    y '2  4 y 2  dx , y (0)  e 2 , y (1)  1
0


6. Find the extremal of the functional J  y  x    
0
4
 y '  y dx that satisfy the boundary conditions
2 2

  1
y  0   1, y    .
4 2

7. Find the extremals and the stationary function of the functional J  y  x      y '2  y 2  dx that
0

satisfy the boundary conditions y  0   1, y     1.

2
8. Find the extremals and the stationary function of the functional J  y  x    
0
 y '  y  dx that
2 2

satisfy the boundary conditions y  0   1, y  2   1.

1 1
9. Find the extremal of the functional I [ y ( x)]   yy '2 dx ; y (0)  1 , y (1)  4 3
0

1
1  y2
10. Find the extremal of the functional I [ y ( x)]   dx passing through the origin and the point
0
y'

(1,1).

x2
1  y2
11. Find the extremal of the functional I [ y ( x)]   dx
x1
y '2
x2

12. Find the extremal of the functional y 1  y '2 dx

x1
 10
x2
1  y '2
13. Find the extremal of the functional I [ y ( x)]   dx
x1
y
x2
1  y '2
14. Find the extremal of the functional I [ y ( x)]   dx
x1 y

Answers
a
1. y "  y0 2. y  c1 e x  c2 e  x 3. x  sin t
b
4. y  c1 sin  4 x  c2  or y  c1 cos  4 x  c2  or y  c1 cos 4 x  c2 sin 4 x 5. y  e 2(1 x )
6. y  cos x 7. y  cos x  c sin x. 8. y  cos x  c sin x.
2
x 
9. y   x  1 3 10. y  tan   11. y  sin h  c1 x  c2 
 4 
 x  c2  2
12. y  c1 cosh   13.  x  c1   y 2  c22 14. x  a   sin   , y  a 1  cos  
c
 1 
---------------------------------------------------------------------------------------------------------------------------
Type : Integrand is of the form y 2  y '2  f ( x ) y or y 2  y '2  f ( x ) y or y '2  y 2  f ( x ) y

Example : Find the extremals of the functional I  y  x      y '2  y 2  4 y cos x  dx with
0

y  0   0, y    0 .
Solution : F  x, y , y '  y '2  y 2  4 y cos x, y (0)  y ( )  0
d
2 y  4cos x   2 y '  0
dx
  y  2cos x  y "  0  y " y  2cos x
A.E. D2 1  0  D  i
 C.F.  c1 cos x  c2 sin x
1
P.I.  2  2cos x  , (case of failure)
D 1
x
  2cos x   x sin x
2D
y ( x)  c1 cos x  c2 sin x  x sin x , y  0   0  c1  0
 y ( x)  c2 sin x  x sin x , y    0  c2  0   0
 c2 is arbitrary.
 y ( x)  (c  x) sin x
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 11

Exercise 1.7

1. Find the extremals of the functional I  y  x     y  y '2  8 y cos h x  dx with y  0   2,
2

  
y    2cosh .
2 2

 
2. Find the extremals of the functional I  y  x     y'  y  2 xy  dx with y  0   0, y    0 .
2 2

0 2

 
3. Find the extremals of the functional I  y  x     y  y '2  2 y sin x  dx with y  0   0, y    1 .
2

0 2

  1
4. Find the extremals of the functional I  y  x     y'  y  4 y sin 2 x  dx with y  0   y    .
2 2

0 2 3

5. Find the extremals of the functional I  y  x     y  y '2  2 y e x  dx .
2


6. Find the extremals of the functional I  y  x     y  y '2  2 y sec h x  dx .
2

Answers
 x
1. y  x   2 cos h x 2. y  x   x  sin x 3. y  x   sin x  cos x
2 2
1 1 x
4. y  x    cos x  sin x  cos 2 x  1 5. y  x   c1e x  c2e  x  e x
3 3 2
6. y  x   c1 cosh x  c2 sinh x  x sinh x  cosh x log 2 cosh x

---------------------------------------------------------------------------------------------------------------------------

Integrand contains all x, y, y ' and not of any of the above forms
e
Example : Find the extremal of the functional I [ y ( x)]    xy '2  yy ' dx subject to the conditions
0

y (1)  0 , y (e)  1 .
Solution : F  x, y , y '   xy '2  yy ', y (1)  0, y (e)  1
d
y '  2 xy ' y   0
dx
 y ' 2 xy " 2 y ' y '  0  xy " y '  0
 x 2 y " xy '  0
 12

Put x  e z  z  log x
D2  D  D  0  D2  0  D  0,0
 y ( z )  c1  c2 z
y  x   c1  c2 log x , y 0  0  c1  0
 y ( x)  c2 log x , y e  1  1  c2  y ( x )  log x

Exercise 1.8
1
1. Find the extremal of the functional I [ y ( x)]    y '2  12 xy  dx with y (0)  0 , y (1)  1 .
0
2
2. On what curves can be functional J  y  x      y '2  2 xy  dx; y 1  0, y  2   1 attain an
1

extremum ?
0

 12 xy  y '  dx
2
3. Find the stationary function of the functional I  y  x    which is determined by
1

the boundary conditions y  1  1, y  0   0 .

1

  y '  2 xy  dx that satisfy the conditions

2
4. Find the extremal of the functional J  y  x   
1

y  1  1, y 1  1.

x2

  a( x) y '  2b( x) yy ' c( x) y  dx is a second

2 2
5. Show that the Euler’s equation for the functional
x1

order linear differential equation.

Answers
x
1. y  x3 2. y 
6
1  x2  . 3. y   x 3

7 x3
4. y  x  5. a( x ) y '' a '( x) y '  b '( x )  c( x)  y  0
6 6
---------------------------------------------------------------------------------------------------------------------------

Type : Invariance of Euler’s equation

x2

Example : Find the extremals of the functional I  y ( x)    1  y '2 dx .

x1

or
2
Find the extremals of the functional  r 2  r '2 d by using the transformations x  r cos and
1

y  r sin  .
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 13

Solution : Let x  r cos  , y  r sin 

dx  cos  dr  r sin  d ……(1)
dy  sin  dr  r cos d ……(2)
dx dr d
 cos   r sin 
d d d
dy dr d
 sin   r cos 
d d d
dx dr
 cos   r sin  ……(3)
d d
dy dr
 sin   r cos  …….(4)
d d
Squaring and adding equation (3) and (4)
2 2 2
 dx   dy   dr  2 2 2
      r  r r'
 d    d   d  
x2 2 2 x2 x2 2
 dx   dy  2 2  dy 

     d    dx    dy  
 1    dx
x1
 d   d  x1 x1
 dx 
x2

1  y '2 dx


x1

y'
Let F  x , y , y '   1  y '2  Fy  0, Fy ' 
1  y '2

d  y'  d  y' 
0  0   0
dx  1  y '2  dx  1  y '2 
   
y' 1  y '2 1
Integrating both sides c   c
1  y '2 y' c

 1  y '2  y ' c  1  y ' 2  y '2 c 2   

1  y '2 c 2  1
1 1 dy
 y '2   y'   c1  say    c1
c 2
1  c2 1 dx


 dy   c dx 1  y  c1x  c2

 r sin   c1r cos   c2

 14

Exercise 1.9
2
1. Find the extremals of the functional I  r      r sin  r 2  r '2 d by using the transformations
1

x  r cos and y  r sin  .

log 2

 e y '2  e x y 2  dx, by using the transformations

x
2. Find the extremals of the functional I  y  x   
0

x  log u and y  v , where log 2  log e 2 .

x2

3. Find the extremals through  0,0  and 1,1 of I  y  x     y 2  y '2  x 2  dx by using the
x1

2 2
transformations x  u and y  v.

Answers
r cos 7 x  x4
 2
1. log r sin   r sin   c1 2 2
  c2 
c1
x
2. y  c1 cos e  c2 sin e x
3. y 
6
---------------------------------------------------------------------------------------------------------------------------
Type : Applications of Euler’s equation
Example : Show that the shortest distance between two points in a plane is straight line.
Solution : As we know that the shortest distance between two points in plane is
x2

I  y  x    1  y '2 dx

x1

y'
Let F  x , y , y '   1  y '2  Fy  0, Fy ' 
1  y '2

d  y' 
By Euler equation 0  0
dx  1  y '2 
 
y' 1  y '2 1
Integrating both sides c   c  1  y '2  y ' c
1 y ' 2 y' c
1
 1  y '2  y '2 c 2  
y '2 c 2  1  1  y '2  2
c 1
1 dy
 y'   c1  y '  c1   c1
c2 1 dx


 dy   c dx 1  y  c1x  c2

which is a required equation of a straight line.

J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 15

Exercise 1.10
1. Find the curve passing through the point  x1 , y1  and  x2 , y2  which rotated about the x-axis gives
a minimum surface area.
2. Find the path on which a particle in the absence of friction will slide from one point to another in
the shortest time under the action of gravity.

Answers
 x  c2 
1. y  c1 cosh   2. x  a   sin   , y  a 1  cos  
 c1 
--------------------------------------------------------------------------------------------------------------------------------------
2. Isoperimetric Problems
Isoparametric problem : The problem of finding closed curve of the given length which in closes
maximum area the called this problem as isoparametric problem.

Working Rules :
x2

(i) Suppose we wish to find a curve y=y(x) that gives extreme value of the functional  f ( x, y, y ') dx
x1

x2

keeping another integral  g ( x, y, y ') dx = constant.

x1

(ii) Let y  y ( x ) satisfy the boundary conditions y  x1   y1 and y  x2   y2 .

(iii) Let F  f ( x, y, y ')   g ( x, y, y ') where  is called Lagrange’s multiplier.
F d  F  d  d 
(iv) Euler’s equation     0 i.e fy  dx fy '   g y  dx g y   0
y dx  y '   
(v) We have to determine  and two constants of integration.
2 2
2
Example 1. Find the extremal of the functional  y' dx under the constraint  y dx  1 given y (0)  0
0 0

and y (2)  1 .
Solution : F  y '2   y, y (0)  0, y (2)  1
d 
  2 y '  0    2 y"  0  y" 
dx 2
  2
y ( x)  x 2  c1  c2 x , y 0  0  c1  0  y  x  x  c2 x
4 4
 16

1 
y  2  1  1    2c2  c2 
2
 2  1  
 y ( x)  x  x
4  2 
2 2
 2  1   
 y dx  1
0
   4 x
0

 2  
 x  dx  1

2
  x3  1    x 2 
  12   2  2   1
   0
8
  1     1  8  12  12  12   0
12
1 x
 y ( x)  x 
2 2
Example 2. Find the plane curve of fixed length having maximum area.
Solution : Let I be the fixed parimeter of a plane curve, between points x  x1 and x  x2 then
x2

1  y '2 dx
I

x1
……(1)

x2

Let A
 y dx
x1
……(2)

f  x, y , y '   y, g  x, y, y '  1  y '2

F  f  g  F  y   1  y '2
F F y'
  1, 
y y ' 1  y '2

F d  F  d  y' 
  0  1  0
y dx  y '  dx  1  y '2 
 

d  y' 
  1
dx  1  y '2 
 
y'
Now, integrating on both sides we get,  x  c1
1  y '2

Squaring both sides we get,  2

  x  c1 
2
 y' 2
  x  c1 
2

y'  
 x  c1 
2
 2   x  c1 
2
Integrating above equation we get, y    2   x  c1   c2
2
 y  c2    2   x  c1 
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 17

 x  c1 2   y  c2 2   2 which is a circle.
Hence the curve is a circle.

Exercise 2.1

1
1. Find the external for which the functional I   y '2 dx , y (0)  0, y (1)  1 is extremum, subject to
0

1
the condition I   y dx  2.
0

1 1
2
2. Find the extremal of the functional  y ' dx ; y (0)  1 , y(1)  6 subject to the condition
0
 y dx  3
0

4
3. Prove that the extemal of the functional I [ y ( x)]   y '2 dx ; y (1)  3 , y (4)  24 subject to the
1
4
condition  y dx  36 is a parabola.
1
x2 x2
2
4. Find the extremal of the functional  y'
x1
dx subject to the condition  y dx  c , a constant.
x1

 
5. Find the extremal of the functional  ( y '2  y 2 ) dx under the constraint  y dx  1 ; y(0)  0 , y( )  1 .
0 0

6. Find all the functions y, that satisfy the conditions y (0)  y ( )  0 and which under the integral
 
2 2
I   y ' dx is stationary subject to the constraint y dx  1 .
0 0

1 1

  x  y '  dx under the constraint

2 2 2
7. Find the extremal of the functional y dx  2 ; y (0)  0 , y (1)  0 .
0 0

Answers
1. y   9 x 2  10 x 2. y  3 x 2  2 x  1

3 2c  c1  x2  x1   2c2  x2  x1  2
2 2
2
3. y  x  2 x 4. y  . x  c1x  c2
2 x23  x13
1 2  2
5. y 1  cos x   sin x 6. y   sin nx , n  7. y   2sin n x, n  
2 4 
 18

3. Variational Problems for functionals involving several dependent variables

x2

If I  y1  x  , y2  x  ,...., yn  x     F  x, y1 ( x ), y2 x,...., yn x, y1' ( x ), y2' ( x),...., yn' ( x)  dx then

x1

d
Euler’s equation is: Fy  0 where i  1, 2,...., n
Fyi 
dx i
d d d
i.e. Fy1  Fy   0, Fy2  Fy   0 ,...., Fyn  Fy   0
dx 1 dx 2 dx n
1
Example : Find the extremals of the functional I [ y ( x), z ( x )]    y '2  z '2  dx ; y (0)  0, z (0)  0,
0

y (1)  1 , z (1)  2
Solution : F  x, y, y ', z , z '   y '2  z '2 ; y (0)  0  z (0), y (1)  1, z (1)  2

F d  F  F d  F 
  0 and   0
y dx  y '  z dx  z ' 
d d
   2 y '  0 and   2 z '  0
dx dx
 y"  0 and z"  0
 y ( x)  c1  c2 x and z ( x )  c3  c4 x
y (0)  0  c1  0 and z 0  0  c3  0
y 1  1  c2  1 and z 1  2  c4  2
 y ( x)  x and z ( x)  2 x
 y ( x)  x and z ( x)  2 x

Exercise 3.1
1
1. Find the extremals of the functional I [ y ( x), z ( x )]    y '2  z '2  4 z  dx ; y (0)  0, z (0)  0,
0

y (1)  1 , z (1)  0
1 2 2
  dx   dy  
2. Show that the functional   2 x        dt s.t x(0)  1, y (0)  1, x (1)  1.5, y (1)  1 is
  dt   dt  
0 

2  t2
stationary for x  t   , y  t   1.
2

2
3. Find the extremals of the functional I [ y ( x), z ( x )]   ( y '2  z '2  2 yz )dx satisfying
0

   
y (0)  0, y    1, z (0)  0, z    1
2 2
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 19

2
4. Find the extremals of the functional I    x 2  y 2  2 xy  dt , such that at t  0, x  y  0 and at
0

 dx dy
t , x  y  1 where x  and y 
2 dt dt

5. Find the extremals of the functional I [ y ( x), z ( x )]    2 yz  2 y 2  y '2  z '2  dx ;
0

y (0)  0, y ( )  1, z (0)  0, z ( )  1
1
6. Find the extremals of the functional I [ y ( x), z ( x )]   1  y '2  z '2 dx ; y (0)  0, y (1)  2,
0

z (0)  0, z (1)  4
1
7. Find the extremals of the functional I    2 xy
  y 2  x 2  dt s.t. at t  0, x  y  1 and at
0

t  1, x  y  0

Answers

1. y  x   x and z  x   x 2  x 3. y  x   sin x and z  x    sin x .

 
4. x  t   y  t   cosec h   sin ht .
2
x 1
5. y  x   c3 sin x  cos x and z  x   c3 sin x  (2sin x  x cos x ) .
 
sinh(1  t )
6. z  x   4 x and y  x   2 x 7. x  t   y  t  
sinh1
---------------------------------------------------------------------------------------------------------------------------

4. Functional dependent on higher order derivatives

x2

If I   F  x, y, y ', y '',...., y ( n )  dx , then

x1

d d2 dn
Euler’s equation is: Fy 
dx dx dx
 
Fy '  2  Fy ''   .....  (1) n n Fy( n )  0 also known as Euler-Poisson

equation.
x2
d d2
Particular cases : If I   F  x, y, y ', y '' dx then Fy  Fy '  2 Fy ''  0 .
x1
dx dx
 20
x2
d d2 d3
If I   F  x, y , y ', y '', y '''  dx then Fy  Fy '  2 Fy ''  3 Fy '''  0 .
x1
dx dx dx
Example 1. Find the extremal of the functional
1
1 1
I  y ''2 dx ; y (0)  0, y (1)  , y '(0)  0, y '(1)  1 .
20 2
1 2 1
Solution : F  x, y , y '   y " , y (0)  0  y '(0), y (1)  , y '(1)  1
2 2
F d  F  d 2  F 
    0
y dx  y '  dx 2  y " 
d2  1 
   2 y "  0  y IV  0
dx  2 
 y ( x)  c1  c2 x  c3 x 2  c4 x 3 ……..(1)
y (0)  0  c1  0 ……..(2)
1 1
 y (1)   c2  c3  c4  ……..(3)
2 2
y '( x )  c2  2c3 x  3c4 x 2 ……..(4)
y '(0)  0  c2  0 ……..(5)
By equation (1), (2), (4) and (5) we get,
y ( x)  c3 x 2  c4 x3 …….(6)
y '( x)  2c3 x  3c4 x 2 …….(7)
Now, y '(1)  1  2c3  3c4  1 …….(8)
By equation (3) we get,
1
c3  c4  …….(9)
2
By equation (8) and (9) we get,
1
c4  0 and c3  …..(10)
2
x2
Hence the solution is y ( x) 
2
x2

  y ''  2 y '  y  2 y sin x  dx

2 2 2
Example 2. Find the extremal of the functional I [ y ( x)] 
x1

2 2 2
Solution : F  y "  2 y '  y  2 y sin x
d d2
 2 y  2sin x    4 y '  2  2 y "  0
dx dx
y  sin x  2 y " y IV  0
 y IV  2 y " y  sin x
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 21
2
D 4  2 D 2  1   D 2  1  0  D 2  1, 1
 D  i ,  i
 C.F.   c1  c2 x  cos x   c3  c4 x  sin x
1
P.I.   sin x  case fails
 D  1
2

x 1
 sin x  x  sin x case fails
3
4D  4 D 4 D  D 2  1

x2 1 x2 1 x2
   sin x   sin x   sin x
2 12 D 2  4 2  12  4  16
x2
 y ( x)   c1  c2 x  cos x   c3  c4 x  sin x  sin x
16

Exercise 4.1
1
1. Find the extremal of the functional I [ y ( x)]   1  y ''2  dx , y (0)  0, y '(0)  1, y (1)  1, y '(1)  1
0

x2

 4y  y ''2  2 x 2  dx
2
2. Find the extremal of the functional I [ y ( x)] 
x1

x2

 16 y  y ''2  x 2  dx
2
3. Find the extremal of the functional I [ y ( x)] 
x1

x2

  y ''  y  x 2  dx
2 2
4. Find the extremal of the functional
x1

5. Find the extremal of the functional


2
   
I [ y ( x)]    y ''2  y 2  x 2  dx ; y (0)  1, y '(0)  0, y    0, y '    1
0 2 2
6. Find the extremal of the functional

4
    1
I [ y ( x)]    y ''2  y 2  x 2  dx ; y (0)  0, y '(0)  1, y    y '   
0 4 4 2
x2

  360 x y  y ''2  dx
2
7. Find the extremal of the functional
x1

x2

y  2 y '2  y ''2  dx

2
8. Find the extremal of the functional
x1
 22
1
9. Find the extremal of the functional I    y '2  y ''2  dx ; y (0)  0, y '(0)  1, y (1)  sinh1, y '(1)  cosh1
0

1
 y '''2 
10. Find the extremal of the functional I [ y ( x)]    yy '''  dx ; y (1)  y (0)  y (1)  0 ;
1 
2 
y '(1)  y '(0)  y '(1)  1
0
11. Find the extremal of the functional I [ y ( x)]    480 y  y '''2  dx , y (0)  y '(0)  y ''(0)  0
1

12. Find the extremal of the functional

0

  240 y  y '''  dx ;
2
I y (1)  1, y (0)  0 ; y '(1)   4.5, y '(0)  0 ; y ''(1)  16, y ''(0)  0
1
x2

  2 xy  y '''  dx
2
13. Find the extremal of the functional I [ y ( x)] 
x1

x2

14. Find the stationary function of the functional   y '2  yy ''  dx , subject to the conditions
x1

y  x1    , y '  x1    , y  x2    , y '  x2   

2 x
 y ''' 
15. Find the extremal of the functional   y   dx .
x1 
2 
x2

  y '''  y  2 yx3  dx
2 2
16. Find the extremal of the functional I [ y ( x)] 
x1

Answers
1. y  x 2. y  c1e x 2
 c2e  x 2
  
 c3 cos x 2  c4 sin x 2 
3. y  c1e 2 x  c2e 2 x  c3 cos 2 x  c4 sin 2 x 4. y  c1e x  c2 e x  c3 cos x  c4 sin x
5. y  cos x . 6. y  sin x
x6
7. y   c1 x3  c2 x 2  c3 x  c4 8. y   c1  c2 x  e x   c3  c4 x  e  x
2
3 x5  5 x 3
9. y  sinh x 10. y  x  .
2
x6 x6 x3
11. y    c1x 4  c2 x 5  c3 x 3 12. y   x4 
3 6 6
x7
13. y  c1  c2 x  c3 x 2  c4 x3  c5 x4  c6 x 6  14. No extremal 15. No extremal
7!
x 
 x 3  x 3   2x  x 3  x 3   3
16. y  c1e x  c2 e  x  e 2 c3 cos    c4 sin     e c5 cos    c6 sin     x
  2   2     2   2  
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 23

5. Functionals dependent on the functions of several independent variables

I  z  x  , z  y     F  x, y, z , z x , z y  dx dy where x,y are two independent variables , z is dependent

D

variable depending upon x and y and z x , z y are first order partial derivatives of z w.r.t. x and y
respectively.

 
Euler’s Equation is: Fz 
x
 
Fzx   
Fz  0
y y

  2 2 2
Euler’s Equation for 2nd derivative is : Fz 
x
 
Fzx 
y
     
Fz y  2 Fz xx  2 Fz yy  2 Fz xy  0
x y x y
 
 z 2  z  2 
Example : I  z  x, y          dx dy
x y
D      
Solution : F  z x 2  z y 2
   
Fz 
x
 
Fz x 
y y
 
Fz  0  0
x
 2 z x    2 z y   0
y
2 z  2 z
  0
x 2 y 2

Exercise 5.1

Obtain the Euler equation for the following functionals :

 z  2  z  2    2 z 2   2 z  2 2
 2 z  
1. I  z          dx dy 2. I  z     2    2   2    dx dy
D  x   y   D  x   y   x y  

Answers

2 z 2 z 4 z 2 4 z 4z
1.  0 2.   0
x 2 y 2 x 4 x 2 y 2 y 4
 24

6. Boundary Value Problems

Rayleigh-Ritz Method : The solution of Euler’s differential equation along with boundary conditions
amounts to extremising a certain definite integrals. This fact provides a technique of solving a boundary
value problem approximately by assuming a trial solution satisfying the given boundary conditions and
then extremising the integral whose integrand is found from the given differential equation.

To find approximate value of the boundary value problem

a  x  y " x   b  x  y '  x   c  x  y  x   f  x  …..(1)

such that y  x1   y1 , y  x2   y2 …..(2)

then,
x2

Step (i) : Construct a functional I  y  x      P  x  y '2  Q  x  y 2  R  x  y  dx …..(3)

x1

b
c 2 f  x
where P  x   e 
dx
a
Q x  P  x R  x  P  x
a a
Step (ii) : Now approximate y  x  using polynomial expressions which satisfies boundary conditions.

n
y   ci i  x  …..(4)
i 0

where i  x  are polynomial and i ' s  C 2 . Also 0  x  satisfies boundary conditions and all other
i ' s vanish on boundary points and ci ' s need to be determined.

Step (iii) : Put value of y from equation. Now differentiate I  y  x   partially w.r.t ci to get values of
I
ci . i.e.,  0 , will give values of ci ' s .
ci

Put these values of ci ' s in equation (4) to get y  x  .

Example 1 : Solve by BVP y " y  x  0 s.t. y  0   y 1  0 .

Solution : y " y  x  0 .….(1)

Such that y  0   y 1  0 …..(2)
Here a  1, b  0, c  1, f  x    x

0
c 1.1
P  x   e  1  e0  1
. dx
Q x  P x  1
a 1
2 f  x 2 x
R  x  . P x  .1  2 x
a 1
1
I  y  x      P  x  y '2  Q  x  y 2  R   y  dx
0
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 25
1
I  y  x      y '2  y 2  2 xy  dx ….(3)
0

F d  F 
where F  x   y '2  y 2  2 xy for which Euler’s equation   0
y dx  y ' 

d
 2 y  2x   2 y '  0  y  x  y"  0  y " y   x
dx
Let y  co  c1 x  c2 x 2 be the trial solution

y 0  0  c0  0

y 1  0  c1  c2  0  c2  c1  y  c1 x 1  x 
Put y in equation (3), we have
1
2 2
I  y  x       c1 1  2 x    c12 x 2 1  x   2 xc1x 1  x   dx
0
 
1
2 2
  c12 1  2 x   c12 x 2 1  x   2c1x 2 1  x   dx
 
0
1 1 1
2 2
 c12  1  2 x  dx  c12  x 2 1  x  dx  2c1  x 2 1  x  dx
0 0 0

1 1 1
 c12  1  4 x 2  4 x  dx  c12   x 2  x 4  2 x 3  dx  2c1   x 2  x3  dx
0 0 0

1 1 1
 24 x3 4 x 2  2x
3
x5 2 x 4   x3 x 4 
 c x    c    2c  
2  0
1 
4  0
1 1
 3 3 5  3 4 0
 4  1 1 1 1 1
 c12 1   2   c12      2c1   
 3  3 5 2 3 4
1 1  1 11 1
 c12     2c1 I  c12  c1
 3 30  12 30 6
dI 11 1 1 15 5
Now 0  2c1  0  c1   
dc1 30 6 6 11 22
5
Thus, the approximate y  x  
22
 x  x2 

Example 2 : Find the approximate solution of y " xy   x, y  0   y 1  0 and

F  x, y , y '  y '2  xy 2  2 xy .
 26
1
Solution : Here I  y  x      y '2  xy 2  2 xy  dx ……(1)
0

Let y  c0  c1x  c2 x 2 be the trial solution

y  0   0  c0  0

y 1  0  c1  c2  0  c1  c2

 y  c1  x  x 2 

Put value of y in equation (1), we get

1
2
I    c12 1  2 x   c12 x  x  x 2   2c1x  x  x 2   dx
2

0

1
  c12 1  4 x 2  4 x   c12  x 3  x5  2 x 4   2c1  x 2  x 3   dx
0

1 1 1
2 4 x3 4 x2  2 x
4
x 6 2 x5   x3 x 4 
c x
1    c1      2c1  
 3 2 0  4 6 5 0  3 4 0
 4  1 1 2 1 1
 c12 1   2   c12      2c1   
 3  4 6 5 3 4
1 1 1
 c12  c12  2c1
3 60 12
19 1
I  c12  c1
60 6
dI 19 1
Now 0  2 c1   0
dc1 60 6
1 30 5
 c1   
6 19 19

5
Thus approximate solution is y
19
 x  x2 

Note : Rayleigh-Ritz method can also be used to find smallest eigen value.

Example 3 : Find the approximate smallest eigen value  of

y "  y  0, y  0   y 1  0 and F  x , y , y '   y '2   y 2
1
Solution : Here, I    y '2   y 2  dx …….(1)
0

Let y  x   c0  c1 x  c2 x 2 be the trial solution

y 0  0  c0  0
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 27

y 1  0  c1  c2  0  c1  c2

 y  c1  x  x 2 

Put value of y in equation (1), we get

1
2
I    c12 1  2 x    c12  x  x 2   dx
2
 
0

1
 c12  1  4 x 2  4 x   x 2   x 4  2 x5  dx
0

1
 4 x3 4 x 2  x3  x 5 2  x 4   4   
2
 c x       c12 1   2    
4  0
1
 3 2 3 5  3 3 5 2
1  
I  c12   
 3 30 
dI 1   1 
Now 0 2c1     0   0    10
dc1  3 30  3 30

  10 is the least approximate eigen value of the given differential equation.

Exercise 6.1
1. Solve the BVP y " y   x, such that y  0   y 1  0
2. Solve y "  1, y  0   y 1  0
1
1 
3. Find an approximate solution for the functional I  y  x      y '2  y  dx , s.t. y  0   y 1  0
0
2 
1
4. Find an approximate solution for the functional I  y  x      y '2  y 2  2 xy  dx such that
0

y  0   1, y 1  2

5. Find the least eigen value of y "  y  0, y '  0   0, y 1  0 .

Answers
5 1 2 1
1. y  x  
18
 x  x2  2. y  x  
2
 x  x 3. y  x  
2
 x  x2 

5
4. y  x   1  x 
4
 x  x2  5. least eigen value is 3.
 28

7. Weierstrass Function
x2

Let I  y  x     F  x, y , y ' dx be a functional with fixed boundaries i.e., y  x1   y1 and y  x2   y2 .

x1

Then Weierstrass function is defined as E  x, y, y ', p   f  x, y, y '   f  x, y, p    y ' p  f p  x, y , p 

dy
where p is the slope of the extremal. i.e., p  .
dx
Weierstrass function is used to check maxima /minima and strong / weak for a functional.

S. No. Sufficient Condition Nature of extremal of the given functional

1. E  0, for arbitrary y ' strong minima
2. E  0, for some y ' weak minima
3. E  0, for arbitrary y ' strong maxima
4. E  0, for some y '(close to p) weak maxima
1
 y '2 
Example 1 : Test for an extremum the functional I  y  x      x  2 y   dx ; y  0   y 1  0
0
2 

y '2
Solution : F  x, y , y '  x  2 y 
2

F d  F 
By Euler-Lagrange equation   0
y dx  y ' 

d
 2  y '  0  y " 2  0
dx
 y '  2 x  c1  y  x 2  c1x  c2

Now y  0   0  c2  0 , y 1  0  c1  1

 y  x2  x
Extremal is : y  x 2  x p  2x 1

Weierstrass function is :

E  x, y, y ', p   F  x, y, y '  F  x, y , p    y ' p  Fp  x, y , p 

y '2  p2 
E  x  2y    x  2y     y ' p  p
2  2 
1 2
E  x, y, y ', p  
2
 y '  p 2    y ' p  p
2
1  y ' p 
  y ' p  y ' p  2 p  
2 2
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 29

Now E  0 for all y '

 This extremal y  x 2  x will give minimum value of the functional. This minima will be strong
minima.
x2

Legendre Condition : Let I  y  x     F  x, y , y ' dx be a functional with fixed boundaries P1  x1 , y1 

x1

2F
and P2  x2 , y2  . Let F  x, y , y '  possess a continuous partial derivative Fy ' y ' i.e., . Let C be the
y ' y '
curve of extremal of given functional which passes through P1 and P2 . Then, the following table
provides the sufficient conditions for the nature of the extremal of the given functional.

S.No. Sufficient Condition Nature of extremal of the given functional

1. fy' y'  0, at points close to C and for some y ' weak minima
2. f y ' y '  0, at points close to C and also for arbitrary y ' strong minima
3. f y ' y '  0,at points close to C and for some y ' weak maxima.
4. f y ' y '  0, at points close to C and for arbitrary y ' strong maxima.
1
 1 
Example 2 : Test for an extremum the functional I  y  x     e x  y 2  y '2  dx
0  2 
 1 
Solution : F  x, y , y '   e x  y 2  y '2 
 2 
F d  F 
Euler-Lagrange equation is   0
y dx  y ' 
d x
 ex  2 y 
dx
 e  y '  0  e x  2 y  e x y " e x y '  0

 e x  y  y "  0  y " y  0 e x
0 
A.E. is D 2  1  0
 D  1  y  c1e x  c2e  x
Now Fy '  e x y ' Fy ' y '  e x  0 , for all y '
It follows that on the curve C i.e., C : y  c1e x  c2e  x , a strong minima is attained.
Proper field, central field and field of extremals :
Proper field : A family of curves y  y  x, c  is said to form a proper field in a given region D of the
xy plane if one and only one curve of the family passes through every point of the region D. For
example, inside the circle x 2  y 2  1 , the family of parallel lines y  x  c ( c being an arbitrary
constant) forms a proper field since through any point of the above circle there passes one and only
 30
2
one straight line of the family. On the other hand, the family of parabolas y   x  c   1 inside the

same circle x 2  y 2  1 does not form a proper field since the parabolas of this family interest inside
the circle.

Central Field : If all the curves of the family y  y  x, c  pass through a certain point  x0 , y0  , i.e., if
they form a pencil of curves, then they do not form a proper field in the region D, if the centre of the
pencil  x0 , y0  belongs to D. However, if the curves of the pencil cover the entire region D and do not
intersect anywhere in this region, with the exception of the centre of the pencil  x0 , y0  , then the
family y  y  x, c  is said to form a central field.
For example, the pencil of sinusoids y  c sin x for 0  x  a, a   forms a central field. But the
above mentioned pencil of sinusoids forms a proper field in a sufficiently small neighbourhood of the
segment of x-axis for   x  a, where   0, a   . Again the above mentioned pencil of sinusoids
does not form a proper field in a neighbourhood of the segment of x-axis, for 0  x  a1 , a1   .

Extremal Field : If a proper field or a central field is formed by a family of extremals of a given
variational problem, then it is known as an extremals field.
a

  y '  y  dx
2 2
Example 1. Consider the functional I  y  x   
0
Let it be required to include the arc of the extremal y  0 that connects the points  0,0  and  a,0 
where 0  a   in the central field of extremals.
a

Solution : Comparing the given functional with

 F  x, y, y ' dx , we have
0
2 2
F  x, y , y '   y '  y …….(1)
F d  F 
Euler’s equation is   0 …….(2)
y dx  y ' 
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 31

From (1),
F F
 2 y ,  2 y ' and
y y '
Using these values, (2) reduces to
 
2 y  2 y '  0 or D 2  1 y  0, where D  d / dx …….(3)
The general solution of (3) is y  c1 cos x  c2 sin x …….(4)
Since, the required extremal passes through  0,0  ,  4  yields, c1  0 and hence (4) yields.
y  c2 sin x …….(5)
And the curves of this pencil form as central field on the interval 0  x  a, a   including, for
c2  0, the extremal is y  0
dy  dy 
From (5),  c2 cos x so that c2    , showing that the parameter of the family of extremals
dx  dx  0,0 
dy
(5) can be taken as the value of at  0,0  . However, in the above problem, if a   , then the
dx
family of extremals (5) does not form an extremal field.

4

 y'  y 
2 2
Example 2. Find the proper and central fields of extremals for the functional  2 x 2  4 dx
0

4
2
 y2  2x2  4
Solution : Comparing the given functional with
 F  x, y, y ' dx , F  x, y, y '  y
0

F d  F  d  2 y '
 Euler’s equation   0  2 y  0
y dx  y '  dx
d2y d
dx 2  
 y  0 or D 2  1 y  0, where D 
dx
…….(1)

Auxiliary equation of (1) is D 2  1  0 so that D  1 and hence solution of (1) is given by

y  c1 cos x  c2 sin x, c1 and c2 being arbitrary constants …….(2)
(2) in the equation of the extremals.

For c2  0, (2) yields y  c1 cos x, which is a proper field of extremals in the domain 0  x  .
4
Again, for c1  0, (2) yields y  c2 sin x, which is a central field of extremals in the domain

0 x .
4
 32
2

  y '  sin x  dx, y  0   0, y  2   6 is

3 2
Example 3. Show that the extremal of the variational problem
0

included in a central field of extremals of the given functional.

2
3 2
Solution : Comparing the given functional with
 F  x, y, y ' dx, F  x, y, y '  y '  sin
0
x


F d  F 
 

f 3 y '2  0
Euler’s equation 0  0
y dx  y '  dx
dy
Integrating it, y '2  c12 or y '   c1, so that y  c1x  c2 …….(1)
dx
Where c1 and c2 are arbitrary constants.
(1) is the equation of the extremals. Using the given boundary conditions y  0   and y  2   6, (1)
gives
0  c2 and 6  2c1  c2 so that c1  3 and c2  0.
Therefore, (2),  y  3  x,  which is the extremal of the given variational problem.
For c2  0, (1) yields y  c1 x, which is a central field of extremals in the domain 0  x  2 with centre
at  0,0  . For c1  3, y  c1x reduces, to y  3x, showing that the extremal y  3 x is included in the
central field of extremals y  c1x .

Exercise 7.1
Test for an extremum the following functionals :
a
1. I  y  x     y '3 dx ; y  0   0, y  a   b, a  0, b  0
0

2
2. I  y  x      e y '  3 dx ; y  0   0, y  2   1
0
1
3. I  y  x     yy '2 dx ; y  0   4, y 1  4
0

Answers
1. Weak minima on extremal 2. strong minima 3. strong minima
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 33

8. Variational Problems with moving Boundaries

x2

Let the functional I  y  x     F  x, y , y ' dx in which the boundary points are not specified, but are to
x1

be determined along the unknown function. i.e., x1 and x2 are not specified but  x1 , y1  and  x2 , y2 
lie on the curves. Suppose the boundary point  x2 , y2  moves along the curve y    x  . Then, the
sufficient condition to determine the extremals on which functional attains extremum is given by
F
F   ' y '   0 at x  x2
y '
which is known as transversality condition.
If the boundary point  x1 , y1  moves along the curve y    x  then as above transversality condition is.
F
F   ' y '  0 at x  x1
y '
Remark : 1. If  x2 , y2  moves along a vertical line, then transversality condition is  Fy '  0
x  x2

2. If  x2 , y2  moves along horizontal line, then transversality condition is  F  y ' Fy '  x x2  0

Moving boundary problems : An elementary problem with moving boundaries.
Let F  F  x, y , y '  be a three times differentiable function of its arguments and in the xy-plane.
Let there be two curves y    x  and y    x  ……(1)

where   x   C1  a, b  and   x   C1  a, b  .

We consider the functional J  y   F  x, y , y '  dx



……(2)

Defined on the smooth curves y  y  x  , the endpoints of which A  x0 , y0  and B  x1 , y1  lie on the
given curves (1) so that y0    x0  , y1    x1  . It is required to find the extremum of the functional (2).

Theorem : Let the curve  : y  y  x  extremize the functional J  y   F  x, y , y '  dx from among all


curves of the class C 1 joining two arbitrary points of two given curves y    x  , y    x  . Then the

 F   ' y '  Fy ' |x  x0  0, 

curve  is an extremal and the transversality conditions  …..(3)
 F   ' y '  Fy ' |x  x1  0 

are fulfilled at the endpoints A  x0 , y0  and B  x1 , y1  of the curve  .

Thus, to solve an elementary problem with moving boundaries it is necessary :
 34
(i) To write down and solve the appropriate Euler equations. We then obtain a family of extremals
y  f  x, c1, c2  that is dependent on two parameters c1 and c2 .

f  x0 , c1 , c2     x0  , 
(ii) From the transversality conditions (3) and from the equations  ……(4)
f  x0 , c1 , c2     x1  
to determine the constants c1 , c2 , x0 and x1 .
(iii) To compute the extremum of the functional (2).

Example 1. Find the transversality condition for the functional

x1
1
J  y  f  x, y  e tan y'
1  y '2 dx, where f  x, y   0

x0

Solution : Let the left end of the extremal be fixed at the

Point A  x0 , y0  and let the right end B  x1 , y1  be movable along the curve y    x  . We then get

 F   ' y ' Fy ' |x  x1  0

In our case
1
F  f  x, y  e tan y'
1  y '2
1 1 y '
Fy '  f  x, y  e tan y'

1  y '2
The transversality condition is :
 1 1 1 y ' 
 f  x, y  e tan y ' 1  y '2   ' y ' f  x, y  e tan y '  | x  x1  0
 1  y '2 
From this, by virtue of the condition f  x, y   0, we get
 ' y '
 1 ……(*)
1  ' y '
Geometrically, equation (*) signifies that the extremals y  y  x  must intersect the curve

y    x  , along which the boundary point B  x1 , y1  slides, at an angle
.
4
Actually, the relation (*) may be represented thus : suppose that the tangent to the extremal at the
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 35

point B  x1 , y1  lying on the curve y    x  cuts the x-axis at an angle  , while the tangent to the
given curve y    x  cuts it at an angle  . Then tan   y ' , tan    ' and the left member of (*)
   
yields tan    x  ; but 1  tan    ; therefore      , whence     , which is what we
 4 4 4
set out to prove.
Example 2. Find the shortest distance between the point (1,0) and the ellipse
4 x 2  9 y 2  36 …..(1)

Solution : We have to find the shortest distance P1 1,0  and P2  x2 , y2  , where P2 lies on the ellipse
4 x 2  9 y 2  36 . The arc length PP
1 2 of the minimizing curve y  y  x  is given by

x2 1
I  y  x     1  y ' 
2 2
dx …..(2)
1

Where the end point P1 1,0  is fixed while the end P2  x2 , y2 

1 y  axis
lies on equation (1). Here F  1  y '2  2 . Since F is independence of
x and y. Euler’s equation is P2  x2 , y2 

F d  F  d  y' 
  0
y dx  y ' 


y'
C



0 
dx  1  y '2

y '  C 1  y '
2 2 2
0




P
o (1,0) x  axis

1  y '2
C2
 y '2  2
 c12  y '  c1
1 C
 y  c1 x  c2 ……(3)
which is a straight line along which the required shortest distance is attained.

 Equation (3) passes through P1 1,0 

 c1  c2  0  c2  c1
 Equation (3) becomes y  c1  x  1 ……(4)
Also it passes through  x2 , y2  .
 y2  c1  x2  1
2
Now, from equation (1), y  9  x 2     (say)
3
 36

2 x 2 x
 ' x  
3 9  x2 3 9  x2
Here, end point P2  x2 , y2  lies on (1), i.e., 4 x22  9 y22  36 …..(5)
Now, by using transversality condition for   x  , we get
 F   ' y '  Fy '  0
x  x2

  2 x  1 2y ' 
  1  y '2    y '    0
 2 1 y '
2 2
 3 9 x  x  x
2

2c1 x2 c12
 1  c12   0  y '  c1 
3 1  c12 9  x22 1  c12
1  c12  c12 2c1x2
  0
2
1 c 1 3 1  c12 9  x2
2c1 x2 1
   2c1 x2  3 9  x22
2 2 2
3 1 c 1 9 x 1 c 1

 4c12 x22  9  9  x22  ……(6)

Now, from equation (4) and (5), we get
2
4 x22  9c22  x2  1  36
2
 9c12  x2  1  36  4 x22
2
 9c12  x2  1  4  9  x22  ……(7)
Dividing equation (6) and (7), we get
4 x22 9 x2 9 9
2
    x2 
9  x2  1 4 x2  1 4 5
From equation (6), we get
81  81 
4c12  9 9    c12  4  c1  2
25  25 
From equation (4), we get
9  8 9 8
y2  2   1   P2  x2 , y2   P2  , 
5  5 5 5
2 2
9  8  16 64 4 5
 Required shortest distance is    1    0    
5  5  25 25 5
x1
2
Result : For the functional of the form J  y   h  x, y  1   y '  dx where h  x, y   0 at the

x0

1 1
boundary points, the transversality conditions are of the form y '  x    and y '  x    .
 ' x  ' x
That is, the transversality conditions reduced to orthogonality conditions.
J.R. INSTITUTE OF MATHEMATICS
189/35 BEHIND RAILWAY STATION, VAISH COLLEGE ROAD, ROHTAK PIN-124001 (HARYANA)
E-mail us on - [email protected], [email protected] Mob. 8607383607, 9802177766

Page 37

Exercise 8.1

1. Find the shortest distance between the parabola y  x 2 and the straight line y  x  5 .
2. Find the shortest distance from the point A  1,5  to the parabola y 2  x .

3. Find the shortest distance between the circle x 2  y 2  1 and the straight line x  y  4 .
4. Find the shortest distance between the point A  1,3 and the straight line y  1  3 x .

5. Find the shortest distance between  0,1 and y  x 2 .

6. Find the shortest distance between  0,0  and x  y  2 .

7. Find the shortest distance between y  x 2 and x  y  3 .

Answer
19 2 1 3 11
1. 2. 20 3. 2 2  1 4. 5. 6. 1 7.
8 10 2 4 2