_
1 +
_
dy
dx
_
2
dx
!
= min. (9)
Variational Calculus
We assume that y(x) is a function of x and seek the extremum of the func-
tional
I(y) =
_
x
1
x
0
f
_
x, y,
dy
dx
_
dx =
_
x
1
x
0
f (x, y, y
) dx (10)
Now we assume that y
0
(x) is an extremum of I(y), thus I = 0 and introduce
the function y
(x) = y
0
(x) + h(x), where the function h(x) satises the
following conditions h(x
0
) = h(x
1
) = 0.
I(y
) = F() =
_
x
1
x
0
f
_
x, y
,
dy
dx
_
dx (11)
Since we assumed that y
0
(x) is an extremum of I(y), we know that
I =
_
dF
d
_
=0
= F
( = 0) = 0. (12)
3
Therefore, we can write
F
() =
_
x
1
x
0
df
d
dx =
_
x
1
x
0
_
f
y
+
f
y
_
dx
=
_
x
1
x
0
_
f
y
h(x) +
f
y
(x)
_
dx. (13)
Partial integration (
_
uv
dx = uv
_
u
() =
_
x
1
x
0
_
f
y
h(x)
d
dx
_
f
y
_
h(x)
_
dx
( = 0) =
_
x
1
x
0
_
f
y
0
h(x)
d
dx
_
f
y
0
_
h(x)
_
dx
!
= 0 (14)
Since equation (14) has to be fullled for any h(x), we obtain the Euler-
Lagrange dierential equation that has to be fullled in order that y
0
(x) is
an extremum of the functional I(y):
f
y
d
dx
_
_
f
_
y
x
_
_
_
= 0. (15)
4