Begin2.DVI

(Ben Green) #1
the surface. If the surface is a closed surface, then one normal is called an inward

normal and the other an outward normal.

Maximum and Minimum Values


The directional derivative of a scalar field φin the direction of a unit vector e

has been defined by the projection


ds = grad φ·e.

Define a second directional derivative of φin the direction e as the directional deriva-

tive of a directional derivative. The second directional derivative is written

d^2 φ
ds^2 = grad

[

ds

]
·e = grad [grad φ·e ]·e. (7 .73)

Higher directional derivatives are defined in a similar manner.

Example 7-23. Let φ(x, y )define a two-dimensional scalar field and let


ˆeα= cos αeˆ 1 + sin αˆe 2

represent a unit vector in an arbitrary direction α. The directional derivative at a

point (x 0 , y 0 )in the direction eˆαis given by


ds = grad φ·eˆα=

∂φ
∂x cos α+

∂φ
∂y sin α,

where it is to be understood that the derivatives are evaluated at the point (x 0 , y 0 ).

The second directional derivative is given by

d^2 φ
ds^2

= grad

(

ds

)
·ˆeα

d^2 φ
ds^2 =


∂x

(
∂φ
∂x cos α+

∂φ
∂y sin α

)
cos α+


∂y

(
∂φ
∂x cos α+

∂φ
∂y sin α

)
sin α

d^2 φ
ds^2 =

∂^2 φ
∂x^2 cos

(^2) α+ 2 ∂^2 φ
∂x∂y sin αcos α+
∂^2 φ
∂y^2 sin
(^2) α.


Directional derivatives can be used to determine the maximum and minimum

values of functions of several variables. Recall from calculus a function of a single

variable y=f(x)has a relative maximum (or relative minimum) at a point x 0 if for

any xin a neighborhood of x 0 and different from x 0 ,the inequality f(x)< f (x 0 )(or
Free download pdf