0% found this document useful (0 votes)
65 views4 pages

Exercise - 11 PDF

The document provides practice exercises on finding tangent planes and normal lines to surfaces, and finding parametric equations for lines tangent to the intersection of two surfaces. It includes solutions to three exercises: 1) Finding tangent planes and normal lines for three surfaces at given points, by treating the surfaces as level surfaces and taking the gradient vector, which is perpendicular to the tangent plane. 2) Finding parametric equations for the tangent line to the intersection of two surfaces at a point, by taking the cross product of the gradient vectors of each surface, which gives the tangent vector. 3) Explaining that for a plane surface, the gradient vector provides one way to find the tangent line parametrically, while the fact that

Uploaded by

Kriti
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
0% found this document useful (0 votes)
65 views4 pages

Exercise - 11 PDF

The document provides practice exercises on finding tangent planes and normal lines to surfaces, and finding parametric equations for lines tangent to the intersection of two surfaces. It includes solutions to three exercises: 1) Finding tangent planes and normal lines for three surfaces at given points, by treating the surfaces as level surfaces and taking the gradient vector, which is perpendicular to the tangent plane. 2) Finding parametric equations for the tangent line to the intersection of two surfaces at a point, by taking the cross product of the gradient vectors of each surface, which gives the tangent vector. 3) Explaining that for a plane surface, the gradient vector provides one way to find the tangent line parametrically, while the fact that

Uploaded by

Kriti
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 4

Math 2011-Introduction to Multivariable Calculus (Edited by Dr.

Hon-Ming HO)
Practice Exercises 11: Tangent Planes and Differentials

Exercise 1:

Find equations of the tangent planes and normal lines at the given point 𝑃0 on the following surfaces.

a) cos(𝜋𝑥) − 𝑥 2 𝑦 + 𝑒 𝑥𝑧 + 𝑦𝑧 = 4 and 𝑃0 (0 ,1 ,2) .


b) 𝑥 + 𝑦 + 𝑧 = 1 and 𝑃0 (0 ,1 ,0) .
c) 𝑧 = ln |𝑥 2 + 𝑦 2 | and 𝑃0 (1 ,0 ,0) .

Solution to Exercise 1a):

Recall the following facts: (i) every plane is determined by one of its normal vector and a point on the desired plane, (ii) tangent
plane to a level surface 𝑔(𝑥, 𝑦, 𝑧) = 𝑘 at a given point (𝑎, 𝑏, 𝑐) is perpendicular to gradient vector 𝛻𝑔(𝑎, 𝑏 , 𝑐) .

1) Things to do: Treat the given surface as a level surface of some function having three variables 𝑤 = 𝑔(𝑥, 𝑦, 𝑧) at a certain
height. Find a choice of function 𝑔 . Compute gradient vector of 𝑔 at the given point. In this example, there are infinitely
many possible choices for the target function 𝑔 . One simple choice is 𝑤 = 𝑔(𝑥, 𝑦, 𝑧) ≝ cos(𝜋𝑥) − 𝑥 2 𝑦 + 𝑒 𝑥𝑧 + 𝑦𝑧 . Thus
the given surface can be regarded as a level surface of 𝑔 at height 𝑤 = 4 .
𝜕𝑔 𝜕𝑔 𝜕𝑔
𝛻𝑔(𝑥, 𝑦, 𝑧) ≝< , , > =< −𝜋 sin(𝜋𝑥) − 2𝑥𝑦 + 𝑧𝑒𝑥𝑧 , −𝑥2 + 𝑧 , 𝑥𝑒𝑥𝑧 + 𝑦 > ⇒ 𝛻𝑔(0 ,1 ,2) =< 𝟐 , 𝟐 𝟏, > .
𝜕𝑥 𝜕𝑦 𝜕𝑧
2) Things to do: Take a testing point 𝑄(𝑥, 𝑦, 𝑧) on the desired plane. Draw a vector from the point of tangency 𝑃0 to the
testing point to form a vector ⃗⃗⃗⃗⃗⃗⃗
𝑃0 𝑄 . Treat the obtained gradient vector as a free vector instead of a standard position vector.
Transport the obtained gradient vector to the point of tangency. Use the perpendicularity of the gradient vector and ⃗⃗⃗⃗⃗⃗⃗
𝑃0 𝑄 to
find the desired equation.
⃗⃗⃗⃗⃗⃗⃗
𝑃0 𝑄 ⦁ 𝛻𝑔(𝟎 , 𝟏 , 𝟐) = 0 ⇒ < 𝑥 − 𝟎 ,𝑦 − 𝟏 ,𝑧 − 𝟐 > ⦁ < 𝟐 ,𝟐 ,𝟏 > = 0 ⇒ 2𝑥 + 2𝑦 + 𝑧 − 4 = 0 .

The normal line to the level surface at 𝑃0 (0 ,1 ,2) is given by 𝑥 = 𝟎 + (𝟐)𝑡 , 𝑦 = 𝟏 + (𝟐)𝑡 , 𝑧 = 𝟐 + (𝟏)𝑡 where −∞ < 𝑡 < ∞ .

Solution to Exercise 1b):

Without using the concept of level surfaces and gradient vectors, we can conclude that the equation of the desired tangent plane
in part b) is given by 𝑥 + 𝑦 + 𝑧 = 1 . Can you explain why? The normal line is given by 𝑥 = 0 + (1)𝑡 , 𝑦 = 1 + (1)𝑡 , 𝑧 = 0 +
(1)𝑡 where −∞ < 𝑡 < ∞ .

Solution to Exercise 1c):

Recall the following facts: (i) every plane is determined by one of its normal vector and a point on the desired plane, (ii) tangent
plane to a level surface 𝑔(𝑥, 𝑦, 𝑧) = 𝑘 at a given point (𝑎, 𝑏, 𝑐) is perpendicular to gradient vector 𝛻𝑔(𝑎, 𝑏 , 𝑐) .

1) Things to do: Treat the given surface as a level surface of some function having three variables 𝑤 = 𝑔(𝑥, 𝑦, 𝑧) at a certain
height. Find a choice of function 𝑔 . Compute gradient vector of 𝑔 at the given point. In this example, there are infinitely
many possible choices for the target function 𝑔 . One simple choice is 𝑤 = 𝑔(𝑥, 𝑦, 𝑧) ≝ ln|𝒙𝟐 + 𝒚𝟐 | − 𝑧 . Thus the given
surface can be regarded as a level surface of 𝑔 at height 𝑤 = 4 .
𝜕𝑔 𝜕𝑔 𝜕𝑔 2𝑥 2𝑦
𝛻𝑔(𝑥, 𝑦, 𝑧) ≝< , , > =< 2 , 2 , −1 > ⇒ 𝛻𝑔(1 ,0 ,0) =< 𝟐 , 𝟎 , −𝟏, > .
𝜕𝑥 𝜕𝑦 𝜕𝑧 𝑥 + 𝑦 𝑥 + 𝑦2
2

2) Things to do: Take a testing point 𝑄(𝑥, 𝑦, 𝑧) on the desired plane. Draw a vector from the point of tangency 𝑃0 to the
⃗⃗⃗⃗⃗⃗⃗
testing point to form a vector 𝑃 0 𝑄 . Treat the obtained gradient vector as a free vector instead of a standard position vector.
Transport the obtained gradient vector to the point of tangency. Use the perpendicularity of the gradient vector and ⃗⃗⃗⃗⃗⃗⃗
𝑃0 𝑄 to
find the desired equation.
⃗⃗⃗⃗⃗⃗⃗
𝑃0 𝑄 ⦁ 𝛻𝑔(𝟏 , 𝟎 , 𝟎) = 0 ⇒ < 𝑥 − 𝟏 , 𝑦 − 𝟎 , 𝑧 − 𝟎 > ⦁ < 𝟐 , 𝟎 , −𝟏, > = 0 ⇒ 2𝑥 − 𝑧 − 2 = 0 .

Practice Exercises 11 (edited by Dr. Hon-Ming HO)-page 1


Exercise 2:

Find parametric equations for line tangent to the intersection curve of the surfaces at the given point.

a) Surfaces 𝑆1 ∶ 𝑥𝑦𝑧 = 1 and 𝑆2 ∶ 𝑥 2 + 2𝑦 2 + 3𝑧 2 = 6 at point 𝑃0 (1 ,1 ,1) .


b) Surfaces 𝑆1 ∶ 𝑥 = 1 and 𝑆2 ∶ 𝑥 + 𝑦 2 + 2𝑧 = 4 at point 𝑃0 (1 ,1 ,1) .

Solution to Exercise 2a:

Recall a fact about lines in 3-space: every line in 3-space is determined by a vector 𝑣 parallel to the line and a point on the line.
In this case, we may choose 𝑃0 because 𝑃0 is the point of tangency. The remaining question is to find a vector 𝑣 parallel to the
desired tangent line.

1) Things to do: Regard one of the given surfaces as a level surface of some function 𝑤 = 𝑓(𝑥, 𝑦, 𝑧) at a certain height 𝑤 =
𝑘 . There are infinitely many possible choices of 𝑓 and 𝑘 . Choose a simplest pair. Compute the gradient vector of the 1 st
surface. In this example, let 𝑤 = 𝑓(𝑥, 𝑦, 𝑧) = 𝑥𝑦𝑧 . Thus the surface 𝑆1 is a level surface of 𝑓 at height 𝑤 = 1 . The
gradient vector of 𝑓 is given by

𝜕𝑓 𝜕𝑓 𝜕𝑓
𝛻𝑓(𝑥, 𝑦, 𝑧) =< , , > =< 𝑦𝑧 , 𝑧𝑥 , 𝑥𝑦 > ⇒ 𝛻𝑓(1,1,1) =< 1 ,1 ,1 > .
𝜕𝑥 𝜕𝑦 𝜕𝑧

2) Things to do: Do the same thing on the 2nd surface. In this example, let 𝑤 = 𝑔(𝑥, 𝑦, 𝑧) = 𝑥 2 + 2𝑦 2 + 3𝑧 2 . Thus the surface
𝑆2 can be regarded as a level surface of 𝑔 at height 𝑤 = 6 . The gradient vector of 𝑔 is given by
𝜕𝑔 𝜕𝑔 𝜕𝑔
𝛻𝑔(𝑥, 𝑦, 𝑧) =< , , > =< 2𝑥 ,4𝑦 ,6𝑧 > ⇒ 𝛻𝑔(1,1,1) =< 2 ,4 ,6 > .
𝜕𝑥 𝜕𝑦 𝜕𝑧
3) Idea: We may regard vector 𝒗 ⃗ as a tangent vector to the intersecting curve 𝑬 at point 𝑷𝟎 (𝟏, 𝟏, 𝟏) . What is the
geometric relation between vectors 𝒗 ⃗ , 𝜵𝒇(𝟏, 𝟏, 𝟏) and 𝜵𝒈(𝟏, 𝟏, 𝟏) ? On the one hand, since the intersecting curve 𝐸
lies on the surface 𝑆1 ∶ 𝑥𝑦𝑧 = 1 , the desired tangent vector 𝑣⃑ must lie on the tangent plane to the surface 𝑆1 . It is known
that 𝛻𝑓(1,1,1) is perpendicular to the tangent plane. Thus 𝛻𝑓(1,1,1) ⊥ 𝑣 . On the other hand, since the intersecting
curve 𝐸 lies on the surface 𝑆2 , the tangent vector 𝑣 at 𝑃0 must lie on the tangent plane to the surface 𝑺𝟐 at 𝑷𝟎 . (For
this fact, please refer to the geometric property of tangent planes.). It is known that 𝛻𝑔(1,1,1) is perpendicular to the
tangent plane to the surface 𝑺𝟐 at 𝑷𝟎 . Thus 𝛻𝑔(1,1,1) ⊥ 𝑣 .
4) Things to do: Compute the cross-product of the two obtained gradient vectors. The cross-product must be parallel to the
desired tangent line at the given point. In this example, we have
𝑖 𝑗 ⃗
𝑘
𝜵𝒇(𝑷𝟎 ) × 𝜵𝒈(𝑷𝟎 ) = |𝟏 𝟏 ⃗ .
𝟏| = 2 𝑖 − 4 𝑗 + 2𝑘
𝟐 𝟒 𝟔

Conclusion: The parametric equations of the tangent line at 𝑃0 are given by 𝑥 = 1 + 2𝑡 , 𝑦 = 1 − 4𝑡 , 𝑧 = 1 + 2𝑡 .

Solution to Exercise 2b:

Since the surface 𝑆1 is a plane perpendicular to the 𝑥-axis, there are two possible ways to compute the desired parametric
equations for the tangent line at 𝑃0 (1 ,1 ,1) . Can you explain why?

Recall a fact about lines in 3-space: every line in 3-space is determined by a vector 𝑣 parallel to the line and a point on the line.
In this case, we may choose 𝑃0 because 𝑃0 is the point of tangency. The remaining question is to find a vector 𝑣 parallel to the
desired tangent line.

1) Things to do: Regard one of the given surfaces as a level surface of some function 𝑤 = 𝑓(𝑥, 𝑦, 𝑧) at a certain height 𝑤 =
𝑘 . There are infinitely many possible choices of 𝑓 and 𝑘 . Choose a simplest pair. Compute the gradient vector of the 1 st
surface. In this example, since the surface 𝑆1 is a plane, we may choose

𝜕𝑓 𝜕𝑓 𝜕𝑓
𝛻𝑓(𝑥, 𝑦, 𝑧) =< , , > =< 1 ,0 ,0 > ⇒ 𝛻𝑓(1,1,1) =< 1 ,0 ,0 > .
𝜕𝑥 𝜕𝑦 𝜕𝑧

2) Things to do: Do the same thing on the 2nd surface. In this example, let 𝑤 = 𝑔(𝑥, 𝑦, 𝑧) = 𝑥 + 𝑦 2 + 2𝑧 . Thus the surface 𝑆2
can be regarded as a level surface of 𝑔 at height 𝑤 = 4 . The gradient vector of 𝑔 is given by

Practice Exercises 11 (edited by Dr. Hon-Ming HO)-page 2


𝜕𝑔 𝜕𝑔 𝜕𝑔
𝛻𝑔(𝑥, 𝑦, 𝑧) =< , , > =< 1 ,2𝑦 ,2 > ⇒ 𝛻𝑔(1,1,1) =< 1 ,2 ,2 > .
𝜕𝑥 𝜕𝑦 𝜕𝑧
3) Idea: We may regard vector 𝒗 ⃗ as a tangent vector to the intersecting curve 𝑬 at point 𝑷𝟎 (𝟏, 𝟏, 𝟏) . What is the
geometric relation between vectors 𝒗 ⃗ , 𝜵𝒇(𝟏, 𝟏, 𝟏) and 𝜵𝒈(𝟏, 𝟏, 𝟏) ? On the one hand, since the intersecting curve 𝐸
lies on the surface 𝑆1 ∶ 𝑥 = 1 , the desired tangent vector 𝑣⃑ must lie on the tangent plane to the surface 𝑆1 . It is known
that 𝛻𝑓(1,1,1) is perpendicular to the tangent plane. Thus 𝛻𝑓(1,1,1) ⊥ 𝑣 . On the other hand, since the intersecting
curve 𝐸 lies on the surface 𝑆2 , the tangent vector 𝑣 at 𝑃0 must lie on the tangent plane to the surface 𝑺𝟐 at 𝑷𝟎 . (For
this fact, please refer to the geometric property of tangent planes.). It is known that 𝛻𝑔(1,1,1) is perpendicular to the
tangent plane to the surface 𝑺𝟐 at 𝑷𝟎 . Thus 𝛻𝑔(1,1,1) ⊥ 𝑣 .
4) Things to do: Compute the cross-product of the two obtained gradient vectors. The cross-product must be parallel to the
desired tangent line at the given point. In this example, we have
𝑖 𝑗 ⃗
𝑘
𝜵𝒇(𝑷𝟎 ) × 𝜵𝒈(𝑷𝟎 ) = |𝟏 𝟎 ⃗ .
𝟎| = 0 𝑖 − 2 𝑗 + 2𝑘
𝟏 𝟐 𝟐

Conclusion: The parametric equations of the tangent line at 𝑃0 are given by 𝑥 = 1 + (0)𝑡 , 𝑦 = 1 + (−2)𝑡 , 𝑧 = 1 + (2)𝑡 .

Exercise 3:

By about how much will 𝑤 = 𝑓(𝑥 , 𝑦 , 𝑧 ) = 𝑒 𝑥 cos(𝑦𝑧) change as the point 𝑃(𝑥, 𝑦 , 𝑧) moves from the origin a distance of
𝑑𝑠 = 0.1 unit in the direction of < 2 ,2 , −2 > ?

Solution to Exercise 3:

The question above is equivalent to approximate the change of the value of 𝑤 if the point 𝑃(𝑥, 𝑦 , 𝑧) moves from the origin to
another point 𝑄(0 + ∆𝑥 ,0 + ∆𝑦 ,0 + ∆𝑧) where vector 𝑂𝑄 ⃗⃗⃗⃗⃗⃗⃑ has the same direction as the vector < 2 ,2 , −2 > and |𝑂𝑄
⃗⃗⃗⃗⃗⃗⃑| =
0.1 . Recall a fact about local linear approximation.

𝒅𝒇 ≝ 𝑳(𝒙𝟎 + ∆𝒙 , 𝒚𝟎 + ∆𝒚 , 𝒛𝟎 + ∆𝒛) − 𝑳(𝒙𝟎 , 𝒚𝟎 , 𝒛𝟎 ) = ∆𝒙 ∙ 𝒇𝒙 (𝒙𝟎 , 𝒚𝟎 , 𝒛𝟎 ) + ∆𝒚 ∙ 𝒇𝒚 (𝒙𝟎 , 𝒚𝟎 , 𝒛𝟎 ) + ∆𝒛 ∙ 𝒇𝒛 (𝒙𝟎 , 𝒚𝟎 , 𝒛𝟎 ) ,

𝑑𝑓 ≈ the actual change of function 𝑓 when point (𝑥, 𝑦, 𝑧) moves from (𝑥0 , 𝑦0 , 𝑧0 ) to (𝑥0 + ∆𝑥 , 𝑦0 + ∆𝑦 , 𝑧0 + ∆𝑧) .

1) Things to do: Identify starting point (𝑥0 , 𝑦0 , 𝑧0 ) and the ending point (𝑥0 + ∆𝑥 , 𝑦0 + ∆𝑦 , 𝑧0 + ∆𝑧) . Find the three
increments ∆𝑥 , ∆𝑦 , ∆𝑧 . In this example, since the point 𝑃(𝑥, 𝑦 , 𝑧) moves from the origin to another point
𝑄(0 + ∆𝑥 ,0 + ∆𝑦 ,0 + ∆𝑧) , we have 𝒙𝟎 = 𝟎 , 𝒚𝟎 = 𝟎 , 𝒛𝟎 = 𝟎 . The three increments are determined by the following
⃗⃗⃗⃗⃗⃗⃑ = (0.1) ∙ <2 ,2 ,−2> which implies that < ∆𝑥 , ∆𝑦 , ∆𝑧 >=< 1 , 1 , −1 > .
condition 𝑂𝑄 |<2 ,2 ,−2>| 10√3 10√3 10√3
2) Things to do: To compute 𝑑𝑓 , we compute the three partial derivatives of the given function and then evaluate the partial
derivatives at the starting point (𝑥0 , 𝑦0 , 𝑧0 ) . In this example, the three partial derivatives are as follows.
𝜕𝑓 𝜕𝑓
= 𝑒 𝑥 cos(𝑦𝑧) ⟹ 𝒇𝒙 (𝟎 , 𝟎 , 𝟎) = (0 ,0 ,0) = 1 ,
𝜕𝑥 𝜕𝑥
𝜕𝑓 𝜕𝑓
= −𝑧𝑒 𝑥 sin(𝑦𝑧) ⟹ 𝒇𝒚 (𝟎 , 𝟎 , 𝟎) = (0 ,0 ,0) = 0 ,
𝜕𝑦 𝜕𝑦
𝜕𝑓 𝜕𝑓
= −𝑦𝑒 𝑥 cos(𝑦𝑧) ⟹ 𝒇𝒛 (𝟎 , 𝟎 , 𝟎) = (0 ,0 ,0) = 0 .
𝜕𝑧 𝜕𝑧
3) Things to do: Compute the total differential 𝑑𝑓 .
1 1 −1 1
𝑑𝑓 = ∆𝑥 ∙ 𝑓𝑥 (𝑥0 , 𝑦0 , 𝑧0 ) + ∆𝑦 ∙ 𝑓𝑦 (𝑥0 , 𝑦0 , 𝑧0 ) + ∆𝑧 ∙ 𝑓𝑧 (𝑥0 , 𝑦0 , 𝑧0 ) = ( ) (1) + ( ) (0) + ( ) (0) = .
10√3 10√3 10√3 10√3
1 1 −1
Conclusion: The actual change of function 𝑓 given by ∆𝑓 ≝ 𝑓 ( , , ) − 𝑓(0 ,0 ,0) is approximately equal to 𝑑𝑓 =
10√3 10√3 10√3
1
.
10√3

Practice Exercises 11 (edited by Dr. Hon-Ming HO)-page 3


Exercise 4:

By about how much will 𝑤 = 𝑓(𝑥 , 𝑦 , 𝑧 ) = ln √𝑥 2 + 𝑦 2 + 𝑧 2 change as the point 𝑃(𝑥, 𝑦 , 𝑧) moves from (3 ,4 ,12) a distance
⃗⃑ ?
of 𝑑𝑠 = 0.1 unit in the direction of 3𝑖⃑ + 6𝑗⃑ − 2𝑘

Solution to Exercise 4:

The question above is equivalent to approximate the actual change of the value of 𝑤 if the point 𝑃(𝑥, 𝑦 , 𝑧) moves from
⃗⃗⃗⃗⃗⃗⃗⃑
𝑃0 (3 ,4 ,12) to another point 𝑄(3 + ∆𝑥 ,4 + ∆𝑦 ,12 + ∆𝑧) where vector 𝑃 0 𝑄 has the same direction as the vector 3𝑖⃑ + 6𝑗⃑ −
⃗⃑ and |𝑃
2𝑘 ⃗⃗⃗⃗⃗⃗⃗⃑
0 𝑄 | = 0.1 . Recall a fact about local linear approximation.

𝒅𝒇 ≝ 𝑳(𝒙𝟎 + ∆𝒙 , 𝒚𝟎 + ∆𝒚 , 𝒛𝟎 + ∆𝒛) − 𝑳(𝒙𝟎 , 𝒚𝟎 , 𝒛𝟎 ) = ∆𝒙 ∙ 𝒇𝒙 (𝒙𝟎 , 𝒚𝟎 , 𝒛𝟎 ) + ∆𝒚 ∙ 𝒇𝒚 (𝒙𝟎 , 𝒚𝟎 , 𝒛𝟎 ) + ∆𝒛 ∙ 𝒇𝒛 (𝒙𝟎 , 𝒚𝟎 , 𝒛𝟎 ) ,

𝑑𝑓 ≈ the actual change of function 𝑓 when point (𝑥, 𝑦, 𝑧) moves from (𝑥0 , 𝑦0 , 𝑧0 ) to (𝑥0 + ∆𝑥 , 𝑦0 + ∆𝑦 , 𝑧0 + ∆𝑧) .

4) Things to do: Identify starting point (𝑥0 , 𝑦0 , 𝑧0 ) and the ending point (𝑥0 + ∆𝑥 , 𝑦0 + ∆𝑦 , 𝑧0 + ∆𝑧) . Find the three
increments ∆𝑥 , ∆𝑦 , ∆𝑧 . In this example, since the point 𝑃(𝑥, 𝑦 , 𝑧) moves from 𝑃0 (3 ,4 ,12) to another point
𝑄(3 + ∆𝑥 ,4 + ∆𝑦 ,12 + ∆𝑧) , we have 𝒙𝟎 = 𝟑 , 𝒚𝟎 = 𝟒 , 𝒛𝟎 = 𝟏𝟐 . The three increments are determined by the following
3𝑖⃑+6𝑗⃑−2𝑘⃗⃑ 3 3 −1
condition ⃗⃗⃗⃗⃗⃗⃗⃑
𝑃0 𝑄 = (0.1) ∙ |<3 which implies that < ∆𝑥 , ∆𝑦 , ∆𝑧 >=< , , > .
,6 ,−2>| 70 35 35
5) Things to do: To compute 𝑑𝑓 , we compute the three partial derivatives of the given function and then evaluate the partial
derivatives at the starting point (𝑥0 , 𝑦0 , 𝑧0 ) . In this example, the three partial derivatives are as follows.
𝜕𝑓 𝑥 𝜕𝑓 3
= 2 2 2
⟹ 𝒇𝒙 (3 ,4 ,12) = (3 ,4 ,12) = ,
𝜕𝑥 𝑥 + 𝑦 + 𝑧 𝜕𝑥 169
𝜕𝑓 𝑦 𝜕𝑓 4
= ⟹ 𝒇𝒚 (3 ,4 ,12) = (3 ,4 ,12) = ,
𝜕𝑦 𝑥 2 + 𝑦 2 + 𝑧 2 𝜕𝑦 169
𝜕𝑓 𝑧 𝜕𝑓 12
= 2 2 2
⟹ 𝒇𝒛 (3 ,4 ,12) = (3 ,4 ,12) = .
𝜕𝑧 𝑥 + 𝑦 + 𝑧 𝜕𝑧 169
6) Things to do: Compute the total differential 𝑑𝑓 .
3 3 4 3 12 −1 9
𝑑𝑓 = ∆𝑥 ∙ 𝑓𝑥 (𝑥0 , 𝑦0 , 𝑧0 ) + ∆𝑦 ∙ 𝑓𝑦 (𝑥0 , 𝑦0 , 𝑧0 ) + ∆𝑧 ∙ 𝑓𝑧 (𝑥0 , 𝑦0 , 𝑧0 ) = ( )( ) + ( )( ) + ( )( ) = .
169 70 169 35 169 35 11830
3 3 −1
Conclusion: The actual change of function 𝑓 given by ∆𝑓 ≝ 𝑓 (3 + ,4 + ,12 + ( )) − 𝑓(3 ,4 ,12) is approximately equal
70 35 35
9
to 𝑑𝑓 = .
11830

Practice Exercises 11 (edited by Dr. Hon-Ming HO)-page 4

You might also like