1.

80 Variational Problems with Fixed Boundaries

10.Show that the curve, revolving about x-axis, for a given surface of revolution, containing
maximum area, is a circle.

1.15 Hamilton’s principle. Variational principle of least action (Kanpur 2011, Meerut 2009)
Consider first a single particle of mass m moving in a force field. Let the position vector from
a fixed origin to the particle at time t be denoted by r. Then, according to Newton’s laws of motion,
the actual path followed is governed by the vector equation
m (d 2 r dt 2 )  F  0 , ...(1)
where F is the force acting on the particle. Now consider any other path r  r . In what follows,
we assume only that the true and the varied path coincide at two distinct instants t  t1 and t  t2 ,
that is, that the variation r vanish at t  t1 and t  t2
 [r ]t t1  [r ]t t2  0 ...(2)

At any intermediate time t we then have to consider the true path and the varied path r  r .
Taking the scalar productof the variation into (1) and then integrating the result with respect to
b g
time over the interval t1 , t2 , we get

t2  d 2r 
t1  m dt 2 . r  F . r  dt  0
 
t t t
or m{[(d r / dt ). r ]t2   2 ( dr / dt ). ( dr / dt )dt}   2 F . r dt  0
1 t1 t1
[on integrating the first integral by parts]
t2 t
or t1 {( m / 2)  ( dr / dt ) 2 }dt   2 F. r dt  0 ...(3)
t1

[using (2) and the relation (dr / dt ) . (dr / dt )  (1/ 2) (dr / dt )2 ]

Now, T  the kinetic energy of the particle  (m 2)  (dr dt )2

  [(m / 2)  (dr / dt ) 2 ]  T ...(4)
t2
Using (4), (3) reduces to t1  T  F . r  dt  0 ...(5)
This is Hamilton’s principle in its most general form, as applied to the motion of a single
particle. However, if the force is conservative it can be re-written in a more concise form.
Let F = Xi  Y j  Zk .We recall that a force field F is conservative if
F.dr  Xdx + Ydy + Zdz
is the differential dof a single-valued function . Also, then F = grad . The function is called
the force potential and its negative, say V, is called the potential energy. It is to be noted that and
V each involve an irrelevant arbitrary additive constant.
It follows that F is conservative if there exists a single valued function such that
F . r  V ...(6)
t
Using (6), Hamilton’s principle takes the form   2 (T  V ) dt  0 ...(7)
t1
which holds when a potential function V exists i.e. when the forces acting are conservative. For
such a problem Hamilton’s principle (7) states that the motion is such that the integral of the
difference between the kinetic and potential energies is stationary for the true path.

Created with Print2PDF. To remove this line, buy a license at: http://www.software602.com/
 Variational Problems with Fixed Boundaries 1.81

t
The action ( or Hamilton’s integral) is defined as A   2 (T  V ) dt ...(8)
t1
and in general its value depends on the path along which the particle moves in passing from P1 and
P2 where P1 and P2 are the positions of the particle at times t1 and t2 respectively.
It is quite easy to give simple examples in which the actual path of a particle maximizes the
action. However, if the time interval is sufficiently short, then it can be proved that the action is
necessarily a minimum .In this form, Hamilton’s principle is known as the principle of least action
and can be loosely interpreted as saying that nature tends to equalize the kinetic and potntial
energies throughout the motion.
The energy difference = L  (T – V) is known as the kinetic potential or the Lagarangian
function. Using this function, (7) may be re-written as
t
  2 L dt  0 ...(9)
t1
When the force field is non-conservative, then thepotential energy function doesnot exist,
but (5) still holds and F . r is the work done by the force F in a small displacement r . In
particular, when the force is conservative this element of work is equivalent to   V .
The above derivation can easily be extended to a system of n particles by summation, and to
a continous system by integration. For example, if the position vector of the i th particle of mass
mi be ri and if mi is subject to a force Fi , then the total kintic energy is given by
n 2
1  dr 
T   mi  i  ...(10)
i 1 2  dt 
n
while the total work done by the forces acting is given by  Fi . ri ...(11)
i1
Finally, The principle applies equally well to a general dynamical system consisting of
particles and rigid bodies subject to inter-connections and constraints. It is to noted that the
derivation is independent of the coordinates specifying the system.

OBJECTIVE PROBLEMS ON CHEPTER 1

Write (a), (b), (c) or (d) whichever is correct:
1
1. The extremum of the functional I   ( y  2  12 xy )dx satisfying the conditions y(0)  0
0

and y(1)  1 is attained on the curve (a) y  sin2 (x/2)

(b) y  sin (x/2) (c) y  x3 (d) y  (1/2) × {x3 + sin (x/2)}. [GATE 2005]
Ans. (c). Refer Ex. 7 of Art 1.5A
1
2. (i) Extremals y  y (x) for the variational problem I[y(x)]   ( y  y ) 2 dx satisfies the
0

differential equation. (a) y   y  0 (b) y   y  0 (c ) y   y   0 (d ) y   y  0 [GATE 2003]

2
(ii) The functional I [ y( x )]   ( y 2  6 xy  x 3 )dx  0, y(0)  0, y(2)  2 can be extremized
0

on the curve (a) y  x (b) 2y  x3 (c) y  x3 – 6x (d) 2y  x3 – 2x. [GATE 2003]

Ans. (i) (b); (ii) (d)
b
3. (i) The Euler’s equation for a functional of the form a
F ( x, y )dx is
(a) Fy   C1 (b) Fy  y Fy   C1 (c) Fy  C1 (d) none of these [Kanpur 2003, 2004]

Created with Print2PDF. To remove this line, buy a license at: http://www.software602.com/