Mathematics for Engineers 1

NCB10103

Chapter 4
Multivariable Functions
 Multivariable Functions
 Functions with Two Variables
 Functions of Three Variables
 Limits
 Partial Derivatives
 Chain Rules & Implicit
Differentiation
 Introduction

 In the past we have seen function of one variable, y=f(x),

where x is the independent variable and y is the dependent
variable, that is y dependent on x.
 Functions of Two Variables

 Defination:
Suppose D is a set of order pairs of real number, (x,y). A
function with 2 independent variables x and y denotes as
z = f(x,y).
 The set D is the function domain.
 The set of z-values that associates with the numbers in D is
the function’s range.
Functions of Two Variables
 Functions of Two Variables
Example
1 
For f ( x, y )  x 2  y . Find f (-2,9), f  ,0 , domain and range of f .
2 
Solution:
By substituti ng the values of x and y, we obtain
f (2,9)  (2) 2  9  7
2
1  1 1
f  ,0      0 
2  2 4
For f (x, y)  x 2  y , x can be any value, but y is defined for y  0.
Hence the domain of f
D f  ( x, y ) : x  , y  0
The value of y and x 2 are nonnegativ e. Hence the range of f is
R f   f (x, y) : f (x, y)  0
 Functions of Two Variables
Example
For f ( x, y )  x ln y  x 2 . Find f (1,1), f 2, e , domain and range of f .

Solution :
By substituti ng the values of x and y, we obtain
f (1,1)  1. ln1  12  1
f (2,e)  2 ln e  12  3
For f (x, y)  x ln y  x 2 , x can be any value, but ln y is defined only for y  0.
Hence the domain of f
D  ( x, y ) : x  , y  0
f
The value of x 2 are nonnegativ e and xln y can be any values .
Hence the range of f is
R   f (x, y) : f (x, y)  
f
 Functions of Two Variables

Determine f(1,4) , f(0,9), domain and range for the following

functions:

(i) f ( x, y)  3x2 y  1,

(ii) f (x, y)  cos( 2π x )

y
Functions of Two Variables
Find f (1,4) , f (0,9), domain and range for the following functions
(i) f ( x, y)  3x 2 y 1, (ii) f ( x, y)  cos( 2 x )
y

(i) By substituti ng the values x and y in the function, we obtain

f ( x , y)  3(1) 2 4  1  5
f ( x , y)  3(0) 2 9  1  1

The value of y is defined for y  0. Hence the domain of the function

D  ( x, y) : x , y  0 

The value of y and x 2 are nonnegativ e. Hence the range of the function
R   f ( x, y) : f ( x, y)  1
 Functions of Two Variables

(ii) By substituti ng the values x and y in the function, we obtain

2π 1 π
f ( x , y)  cos( )  cos( )  0
4 2
2π 0
f ( x , y)  cos( )  cos(0)  1
9
The value of x is defined for x  0 and the value of 1 y is defined for y  0.
Hence the domain of the function

D  ( x, y) : x  0, y  0 

The value of cosine function is between -1 and 1. Hence the range of the function
R   f ( x, y) : 1  f ( x, y)  1
 Functions with Three
Variables
Ø If we denote a function of three variables as   f ( x, y, z ),
the n x, y, and z are the independen t variable s and
 is the dependent variable.
 Functions with Three Variables
yz
Given f ( x, y, z )  . Find f (0,1,1), f  1,1,3, domain and range of f .
x2
Example Solution :
By substituti ng the values of x, y and z, we obtain
(1)(1) 1
f (0,1,1)  
02 2
(1)(3)
f (-1,1,3)  3
-1  2
yz
For f ( x, y, z )  , the value of y and z can be any value,
x2
1
the value of is defined for x  -2.
x2
Hence the domain of f
D  ( x, y, z ) : y, z  , x  2
f
yz
The value of can be any values .
x2
Hence the range of f is
R   f (x, y, z) : f (x, y, z)  
f
 Functions with Three Variables
Example
Given f ( x, y, z )  z 2  sin xy . Find f (1,/2,-2), f 0,1,2 , domain and range of f .
Solution :
By substituti ng the values of x, y and z, we obtain
f (1,/2,-2)  (2) 2  sin(1. / 2)  5
f (0,1,2)  22  sin(0.1)  4
For f ( x, y, z )  z 2  sin xy, the value of x, y and z can be any value,
Hence the domain of f
D  ( x, y, z ) : x, y, z  
f
The value of z 2 are nonnegativ e, and the value of sine function is between 1 and -1.
Hence the range of f is
R   f (x, y, z) :   1
f
 Limits
How does this extend to functions of two variables?
We can start with informal language as before:

lim f ( x, y)  A
( x, y )( a,b )

means that as (x,y) gets “closer and closer” to (a,b) , f(x,y)

gets closer and closer to A.
Limits
 Limits
Example

 Verify the following limits computations:

2 x 2  3y 2
lim
( x , y ) ( 2 , 3) 5 xy  4 y

Solution :
lim (5xy  4 y)  5.2.3  4.3  42  0
( x , y ) ( 2 , 3)

2 x  3y
2 2 lim 2 x 2  3y 235
( x , y ) ( 2 , 3)
lim  
( x , y ) ( 2 , 3) 5xy  4 y lim 5xy  4 y 42
( x , y ) ( 2 , 3)
 Limits
Example

 Verify the following limits computations:

x2  y2
lim
( x , y ) ( 0 , 0 ) x  y

Solution :
x 2  y2 ( x  y)( x  y)
lim  lim
( x , y ) ( 0 , 0 ) x  y ( x , y ) ( 0 , 0 ) xy
 lim ( x  y)  0  0  0
(x, y)(0,0)
 Limits
Example

xy4 xy4
1. lim  lim
2
(x,y )(2,2) ( x  y) 16 (x,y )(2,2) ( x  y  4)(x  y  4)

1 1
 lim 
(x,y )(2,2) ( x  y  4) 8

xy  x x( y 1)( y 1)
2. lim  lim
(x,y )(0,1) y 1 (x,y )(0,1) ( y 1)( y 1)

x( y 1)( y 1)
 lim  lim x( y 1)  0
(x,y )(0,1) y 1 (x,y )(0,1)
 Partial Derivative

 For a function of one variable, the rate of change of the function

is represente d by its derivative
 For a function of more than one variable, we are interested in the
rate of change of the function w ith respect to one of its variables while the
other vari ables remain fixed
 This leads to the concept of partial derivative s
 d
 In partial differenti ation, operator is used instead of
x dx
 Partial Derivative
z  f ( x, y )

f f ( x  x, y )  f ( x, y )
 lim
x x 0 x
f f ( x, y  y )  f ( x, y )
 lim
y y0 y
 Partial Derivatives

z  f ( x, y ) y is kept
constant
z f ( x  x, y )  f ( x, y )
 lim
x x 0 x x is kept
constant
z f ( x, y  y )  f ( x, y)
 lim
y y0 y
z f 
Notations : , f x (x, y), , f (x, y);
x x x
 Partial Derivatives
Example Find f x (0,1) , and f y (0,1) if
(i) f ( x, y )  2 x 2  3x 2 y  y 2  x  y  1

By regarding y as a constant, we differenti ate f ( x, y ) with respect

to x to obtain
f x ( x, y )  4 x  6 xy  0  1  0  0
 4 x  6 xy  1
Hence f x (0,1)  4(0)  6(0)(1)  1  1
By regarding x as a constant, we differenti ate f ( x, y ) with respect
to y to obtain
f y ( x, y )  0  3x 2  2 y  0  1  0  3x 2  2 y  1
Hence f y (0,1)  3(0) 2  2(1)  1  3
 Partial Derivatives
Example

Find the partial derivatives fx and fy for the function

f ( x, y)  5x4  x2 y2  2x3 y
Solution:

f ( x, y )  5 x 4  x 2 y 2  2 x 3 y
f x ( x, y )  20 x3  2 y 2 x  6 yx 2
f y ( x, y )  2 x 2 y  2 x3
 Partial Derivatives
Definition of Partial Derivatives of a Function of Three or
More Variables
If w = f(x,y,z), then there are three partial derivatives each
of which is formed by holding two of the variables
w f  x  x, y, z   f ( x, y, z )
 f x  x, y , z   x lim 0
x x
w f  x, y  y, z   f ( x, y, z )
 f y  x, y , z   y lim 0
y y
w f  x, y, z  z   f ( x, y, z )
 f z  x, y , z   z lim 0
z z
In general, if
w  f ( x1 , x2 ,...xn ) there are n partial derivatives
w
 f xk  x1 , x2 ,...xn  , k  1, 2,...n
xk
 Functions of three variables
2 3
f ( x, y,z)  3x yz  z y

f ( x, y,z)  fx ( x, y,z)  6xyz
x
 2 3
f ( x, y,z)  fy ( x, y,z)  3x z  z
y

f ( x, y,z)  fz ( x, y,z)  3x 2 y  3z 2 y
z
 Higher Order Partial Derivatives
Notation for Higher Order Partial Derivatives
Below are the different 2nd order partial derivatives:
  f   2 f
   2  f xx Differentiate twice with respect to x
x  x  x
  f   2 f
   2  f yy Differentiate twice with respect to y
y  y  y
  f   2 f
   f xy Differentiate first with respect to
y  x  yx
x and then with respect to y
  f   2 f
    f yx Differentiate first with respect
y  y  xy to y and then with respect to x
 1 variable Vs 2 variables:
Higher Derivatives
f ( x, y )
f ( x)  
d
| x y
dx
f ( x) f x ( x, y ) f y ( x, y )
d |    
dx x y x y
f ( x)
f xx ( x, y ) f xy ( x, y ) f yx ( x, y ) f yy ( x, y )
 Higher Order Partial Derivatives
Example
Find f xx , f xy , f yx , f yy if f ( x, y )  x 2 y 3  x 4 y

The first derivative of f are

 
f x  ( x 2 y 3  x 4 y )  2 xy3  4 x 3 y , f y  ( x 2 y 3  x 4 y )  3x 2 y 2  x 4
x y
The second derivative of f are
 
f xx  ( f x )  (2 xy3  4 x 3 y )  2 y 3  12 x 2 y
x x
 
f xy  ( f x )  (2 xy3  4 x 3 y )  6 xy2  4 x 3
y y
 
f yx  ( f y )  (3 x 2 y 2  x 4 )  6 xy2  4 x 3
x x
 
f yy  ( f y )  (3 x 2 y 2  x 4 )  6 x 2 y
y y
 Higher Order Partial Derivatives
Example
Find f xx , f xy , f yx , f yy
if given y  ( x  y )e xy


f x ( x, y )  [( x  y )e xy ]
x
 (1  0)e xy  ( x  y ) ye xy
 (1  xy  y 2 )e xy


f y ( x, y )  [( x  y )e xy ]
y
 (0  1)e xy  ( x  y ) xexy
 (1  xy  x 2 )e xy
Higher Order Partial Derivatives

 
f xx ( x, y )  ( f x )  [(1  xy  y 2 )e xy ]
x x
 (0  y  0)e xy  (1  xy  y 2 ) ye xy
 (2 y  xy2  y 3 )e xy

 
f xy ( x, y )  ( f x )  [(1  xy  y 2 )e xy ]
y y
 (0  x  2 y )e xy  (1  xy  y 2 ) xexy
 (2 x  2 y  x 2 y  xy2 )e xy
Higher Order Partial Derivatives

 
f yy ( x, y )  ( f y )  [(1  xy  x 2 )e xy ]
y y
 (0  x  0)e xy  (1  xy  x 2 ) xexy
 (2 x  x 2 y  x 3 )e xy

 
f yx ( x, y )  ( f y )  [(1  xy  x 2 )e xy ]
x x
 (0  y  2 x)e xy  (1  xy  x 2 ) ye xy
 (2 x  2 y  x 2 y  xy2 )e xy
 Higher Order Partial Derivatives
f ( x, y)  3x 2 y  2sin( xy)  y 3
Example fx ( x, y)  6xy  2cos(xy) y
fy ( x, y)  3x 2  2cos(xy) x  3y 2

fxx ( x, y)  6y  2sin( xy) y 2


fxy ( x, y)  fx ( x, y)  6x  2sin( xy) xy  2cos(xy)
y

fyx ( x, y)  fy ( x, y)  6x  2sin( xy) xy  2cos(xy)
x
fyy ( x, y)  2sin( xy) x 2  6y
Notice that fxy = fyx
 Total Differential

Definition :
Let z  f ( x, y ) with f x ( x, y ) and f y ( x, y ) exist. If δx and δy
denotes the changes in the values of x and y, then the correspond ing
change of z, δz also called the total differenti al of f is given as

δz  fx δx  fy δy
 Total Differential
Example
Let z  f ( x, y )  3x 2 y  y 2 , use partial derivative to estimate the change, δz
if the value of ( x, y ) changes from (1,2) to (1.02,1.97)
Solution :
f x ( x, y )  6xy and f y ( x, y )  3x 2  2y
The change in the values of x and y,
δx  1.02  1  0.02 δy  1.97  2  0.03
δz  fx δx  fy δy
 (6xy) x  (3x 2 - 2y) y
 6(1)( 2)(0.02)  (3(1) 2  2(2))( 0.03)
 0.27
 Total Differential
Example
Find an approximat ion for z by using partial derivative if
z  f ( x, y )  3 x 2  xy and ( x,y) changes from (1,2) to (1.01,1.98)

We have f x ( x, y )  6 x  y and f y ( x, y )   x
Hence we obtain

Substituti ng x  1, y  2, and x  0.01 and y  -0.02, we obtain

z  (6  2)0.01 - (-0.02)
 0.06
 Total Differential
Example Suppose the dimension (in cm) of a rectangula r box change from
14, 12 and 10 to 14.02, 11.97, and 10.01. Use partial derivative s to
approximat e the change in volume.
Solution :
V(x, y, z)  xyz , Vx  yz , Vy  xz, Vz  xy
V  Vx ( x, y, z )x  V y ( x, y, z )y  Vz ( x, y, z )z
 yzx  xzy  xyz
 12(10)0.02  14(10)(-0.03)  12(14)0.01
 0.12
 Chain Rule

The chain rules for functions of more than one variables is an

extension of the chain rules for functions of one variable.

Theorem : If z  f (u,v ), u  g( x,y ) and v  h ( x,y ) with partial

derivative s of f, g and h exist, then
z z u z v
  ,
x u x v x
z z u z v
  ,
y u y v y
 Chain Rule
The tree diagram can be used to help us remember t he chain rule.
The tree diagram for theore m above is as shown in figure below.

The tree diagram for function of two variables

 Chain Rule
Suppose z  u 3  v 3 , and u  xy2 , v  x 2 sin y. Use a chain rule z x
Example
and z y .

From z  u 3  v 2 , we obtain
z z
 3u 2 and  2v
u v
Similarly from u  xy 2 and v  x 2 sin y,
u u v v
 y2 ,  2 xy,  2 x sin y ,  x 2 cosy
x y x y
Hence we obtain
z z u z v
   3u 2 ( y 2 )  2v(2 x sin y )
x u x v x
 3u 2 y 2  4vx sin y
 Chain Rule
Conts
z z u z v
   3u 2 (2 xy)  2v( x 2 cos y )
y u y v y
 6u 2 xy  2vx2 cos y

Substituti ng u  xy2 and v  x 2 sin y, we obtain

z
 3u 2 y 2  4vx sin y
x
 3( xy2 ) 2 y 2  4( x 2 sin y ) x sin y
 3x 2 y 6  4 x 3 sin 2 y
z
 6u 2 xy  2vx2 cos y
y
 6( xy2 ) 2 xy  2( x 2 sin y ) x 2 cos y
 6 x 3 y 5  x 4 sin 2 y
 Chain Rule
 Chain rule can be applied to composite functions of any number of
variab les, and tree diagrams can be constructe d to help formulate these
rules.
 As an example, suppose that  is a function of u , v, and r , whereas u , v,
and r each is a function of x, y, and z.

The tree diagram for functions of three variables .

 Chain Rule
Solution: Given z  e xy , x  r - 4s, and y  r - s.
z
Determine in term of r and s.
r
z z x z y
 .  .
r x r y r
 (ye xy )(1)  ( xe xy )(1)
 (x  y)e xy
 [(r - 4s)  (r - s)]e (r -4s)(r-s)
2
r 2-5 rs 4 s
 (2r - 5s)e
 Implicit
Theorem : If F(x,y)  0 implicitly defines a differenti able function y as a
function of one variable x, then
dy F ( x, y )
 x
dx Fy ( x, y )

dy
Find , if y as a function of x is implicitly
Example dx
defined by y 5  3 y-4 x 3-5 x  1

From y 5  3 y-4 x 3-5 x  1 , we obtain

dy F ( x, y )
 x
dx Fy ( x, y )
(12 x 2  5) 12 x 2  5
 
5y  3
4
5y4  3
 Implicit
Theorem : If F(x,y, z )  0 implicitly defines a differenti able function y as a
function of 2 variables , z  f(x, y) then
dz F ( x, y, z ) dz Fy ( x, y, z )
 x , 
dx Fz ( x, y, z ) dy Fz ( x, y, z )

dz dz
Find and , if z  f(x, y)
dx dy
Example
defined by z 3  xy  yz  y 3  2

F ( x, y, z )  z 3  xy  yz  y 3  0 ,
dz F ( x, y , z )  (  y ) y
 x  2  2
dx Fz ( x, y, z ) 3z  y 3z  y
dz Fy ( x, y, z )  ( x  z  3 y 2 ) x  z  3 y 2
  
dy Fz ( x, y, z ) 3z 2  y 3z 2  y
Notes
Notes