MA103L - Engineering Mathematics I

Lecture 10

Shilpak Banerjee1

1 Assistant Professor

Department of Mathematics and Statistics

IIT Tirupati, Tirupati

August 28, 2023

Recall

1
We already discussed
• Review of calculus of real valued functions of a single real variable.
◦ limits,
◦ continuity, and
◦ differentiation
• Partial derivatives: Functions of several variables.
◦ Terminologies from topology: Interior points, boundary points, open sets,
closed sets, bounded sets, unbounded sets.
◦ Level curves, level surfaces, contour maps,
◦ Limits and continuity,
◦ Partial Derivatives, Differentiability and their relation to continuity.

Today we talk more about Chain rule.

2
Partial derivatives: The
Chain Rule

One independent variable and two intermediate However,

To remember the Chain Rule, picture the
ing equati
variables diagram below. To find dw > dt, start at w
and read down each route to t, multiplying a function
derivatives along the way. Then add the w = ƒ(x,
products. dw>dt, dx
If w = f (x, y) is differentiable and if Chain Rule 0w>0x and
x = x(t), y = y(t) are differentiable functions With that
w = f (x, y) Dependent
variable text when
of t, then the composite w = f (x(t), y(t)) is a The b
'w 'w
differentiable function of t and 'x 'y Chain Ru
y are inter
dw x y Intermediate A mo
= fx (x(t), y(t)) · x0 (t) + fy (x(t), y(t)) · y 0 (t) variables
Theorem
dt dx dy
dw ∂f dx ∂f dy dt dt
= + Independent
dt ∂x dt ∂y dt t variable
dw 'w dx 'w dy Example
= +
dt 'x dt 'y dt

Figure 1: Source: Thomas’s with resp

[We can think of this as differentiating a t = p>2?
function defined on a curve in 2D] Calculus
Solution 
4
An example
Find dw/dt where

w = x2 y, x(t) = cos t y(t) = sin t

We proceed by chain rule

dw ∂f dx ∂f dy
= +
dt ∂x dt ∂y dt
(x y d(cos t) ∂x2 y d sin t
2
= +
∂x dt ∂y dt
2
= (2xy)(− sin t) + (x )(cos t)
= −2 cos t sin2 t + cos3 t
= cos t cos 2t
5
 824 Chapter 14: Partial Derivatives

One independent variable and threeHere we have three routes from w to The p
t instead of two, but finding dw > dt is intermedi
intermediate variables still the same. Read down each route, new equat
multiplying derivatives along the way;
then add. Example
If w = f (x, y, z) is differentiable and if Chain Rule

x = x(t), y = y(t), z = z(t) are differentiable w = f (x, y, z) Dependent In this exa

variable
functions of t, then the composite changes. W
'w 'w
w = f (x(t), y(t), z(t)) is a differentiable 'x 'w 'z
Solution 
'y
function of t and x y z Intermediate
variables
dy
dw ∂f dx ∂f dy ∂f dz dx dt dz
= + + dt dt
dt ∂x dt ∂y dt ∂z dt Independent
t variable
dw 'w dx 'w dy 'w dz
= + +
dt 'x dt 'y dt 'z dt so
[We can think of this as differentiating a
function defined on a curve in 3D] Figure 2: Source: Thomas’s
Calculus
For
6 a
Two independent variable and three
intermediate variables
14.4  The Chai
If w = f (x, y, z), x = x(r, s), y = y(r, s), z =
z(r, s) are differentiable functions, then the w = f (x, y, z)
Dependent
z(r, s)) is a w
composite w = f (x(r, s), y(r, s),variable 'w
'w 'w
differentiable function with partial derivatives f 'x 'z 'x
'w 'w
'y 'y
Intermediate
x y z x y z x
∂w ∂f ∂x ∂f dy variables
∂f ∂z 'y 'y
= + +
∂r ∂x ∂r ∂y dr ∂z ∂r g h k 'x 'r 'z 'x 's
'r 'r 's
∂w ∂f ∂x ∂f ∂y Independent
∂f ∂z r, s
= + +variables r
∂s ∂x ∂s ∂y ∂s ∂z ∂s w = f( g(r, s), h(r, s), k(r, s)) 'w
=
'w 'x
+
'w 'y
+
'w 'z 'w
=
'w ' x
+
'r 'x 'r dy 'r 'z 'r 's 'x 's

(a)
Figure 3: Source:(b)Thomas’s
[We can think of this as differentiating a
Figure 14.22  Composite function and branch diagrams for Theorem 7.
Calculus
function defined on a surface in 3D]
7
An example
Find ∂w/∂r and ∂w/∂r in terms of r and s if

w = x2 + y 2 + z 2 , x = r − s, y = r + s, z=r

∂w ∂f ∂x ∂f dy ∂f ∂z
= + + = 2(r − s) · 1 + 2(r + s) · 1 + 2r · 1 = 6r
∂r ∂x ∂r ∂y dr ∂z ∂r
∂w ∂f ∂x ∂f ∂y ∂f ∂z
= + + = 2(r − s) · (−1) + 2(r + s) · 1 + 2r · 0 = 4s
∂s ∂x ∂s ∂y ∂s ∂z ∂s

8
Exercises

• Write down a chain rule for two independent variable and two
intermediate variables.
• Write down a chain rule for two independent variable and four
intermediate variables.

9
Implicit differentiation
Suppose the function F (x, y) is differentiable and the equation F (x, y) = 0
defines y implicitly as a differentiable function of x, say y = h(x). How to find
dy/dx with out solving the equation?

Put w = F (x, y), then the equation F (x, y) = 0 implies w = 0 and hence using
chain rule,
dw dx dy
0= = Fx + Fy
dx dx dx
dy
= Fx + Fy
dx
So in conclusion
dy Fx
=−
dx Fy
10
An example
Find dy/dx when
y 2 − x2 − sin xy = 0
We can use the computation in the previous slide to write
dy Fx
=−
dx Fy
−2x − y cos xy
=−
2y − x cos xy
2x + y cos xy
=
2y − x cos xy

11
Partial derivatives:
Directional Derivatives
and Gradient Vectors

12
 Directional derivative
The derivative of f at P0 (x0 , y0 ) in the
direction of the unit vector u = (u1 , u2 ) y
Line x = x 0 + su 1, y = y 0 + su 2
is the number Fi
sh
to
 
df Su
(Du f )P0 =
ds u,P0 u = u1i + u 2 j
motion along
f (x0 + su1 , y0 + su2 ) − f (x0 , y0 ) Direction of
along the lin
= lim increasing s
rate of chang
s→0 s
R u, we find th
different dire
P0(x 0, y0 )
Suppose
x P0(x0 , y0) is a
0

FigureFigure 14.27  The rate of change of ƒ

4: Source: Thomas’s Calculus parametrize
in the direction of u at a point P0 is the rate
P0 in the dire
at which ƒ changes along this line at P0.
lating dƒ > ds
13
Interpretation of directional derivatives
The vertical plane that passes through
P and P0 (x0 , y0 ) parallel to u z
intersects the graph S in a curve C. Surface S: The rat
f(x0 + su1, y0 + su2) − f(x0, y0)
The rate of change of f in the direction z = f(x, y)

of u is the slope of the tangent to C at Tangent line

P in the right-handed system formed Q
Interp
by the vectors u and z-axis. P(x0, y0, z0)
s
The eq
Can you recover the partial derivatives point P
y allel to
from the directional derivatives? C
tion of
x (x0 + su1, y0 + su2) vectors
P0(x0, y0) u = u1i + u2j Wh
u = j,
Figure
Figure 5: 14.28 
Source:The slope of the
Thomas’s trace
Calculus tive gen
curve C at P0 is lim slope (PQ); this is the
QSP any14dir
Partial derivatives as directional derivatives
14.3  Partial Derivatives 811 812 Chapter 14: Partial Derivatives

z Vertical axis The slope of the c

in the plane z plane x = x0 (Figure 14
Vertical axis in x = x0
tangent line to the curve
the plane y = y0
slope. The partial deriva
P(x0, y0, f(x0, y0)) x is held fixed at the val
z = f(x, y) Tangent line The partial derivati
The curve z = f(x, y0)
in the plane y = y0 P(x0, y0, f(x0, y0))
tive with respect to x:

Tangent line z = f(x, y)

0 Notice that we now

x0 0
y0 x0 y0
the point P(x0, y0, ƒ(x0,
face at P? We will see
x
(x0, y0) y
x section, and we will lea
(x0 + h, y0) (x0, y0) y learn more about partial
Horizontal axis in the plane y = y0 (x0, y0 + k)

Figure 14.16  The intersection of the plane y = y0 The curve z = f(x0, y)

Horizontal axis
Figure
with the6: Source:
surface Thomas’s
z = ƒ(x, y), Calculus
viewed from above the first in the plane
in the plane x = x0
x = x0
quadrant of the xy-plane. This tang
Figure 14.17  The intersection of the has slope
Figure plane
7: Source: Thomas’s
x = x with the
0 surface z = ƒ(x, Calculus
y),
Definition  The partial derivative of ƒ(x, y) with respect to x at the point viewed from above the first quadrant of The
(x0 , y0) is
15curve
the xy-plane. in the pla
Thanks
[email protected]

16