Comprehensive Exercises

I. Limits and Continuity

Problem 1

Find the limit lim⁡(x,y)→(0,0)x2y2x2+y2\lim_{(x,y)\to(0,0)} \frac{x^2 y^2}{x^2 + y^2}.

the limitthe squeeze theorempolar coordinates

Approach:

The degree of the numerator is lower than that of the denominator, so the squeeze theorem shows that the limit is zero.

Steps:

Method 1: the squeeze theorem 0≤x2y2x2+y2≤x2y22∣xy∣=∣xy∣2→00 \leq \frac{x^2 y^2}{x^2+y^2} \leq \frac{x^2 y^2}{2|xy|} = \frac{|xy|}{2} \to 0

Method 2: polar coordinates The original expression =lim⁡ρ→0ρ4cos⁡2θsin⁡2θρ2=lim⁡ρ→0ρ2cos⁡2θsin⁡2θ=0= \lim_{\rho\to0} \frac{\rho^4\cos^2\theta\sin^2\theta}{\rho^2} = \lim_{\rho\to0} \rho^2\cos^2\theta\sin^2\theta = 0

Answer: 00

Problem 2

Prove that the limit lim⁡(x,y)→(0,0)x2yx4+y2\lim_{(x,y)\to(0,0)} \frac{x^2 y}{x^4 + y^2} does not exist.

the limit does not existthe path methoda parabolic path

Approach:

Along the coordinate axis and along the parabola the limits differ, and this proves that the limit does not exist.

Steps:

  1. Along the xx-axis (y=0y=0): the limit is 00
  2. Along the parabola y=x2y=x^2: lim⁡x→0x2⋅x2x4+x4=12\lim_{x\to0} \frac{x^2 \cdot x^2}{x^4+x^4} = \frac{1}{2}
  3. The limits along the two paths differ, so the limit does not exist.

Answer: the limit does not exist.

II. Partial Derivatives and Total Differentials

Problem 3

Let z=arctan⁡x+y1−xyz = \arctan\frac{x+y}{1-xy}. Find ∂z∂x\frac{\partial z}{\partial x} and ∂z∂y\frac{\partial z}{\partial y}.

partial derivativesa composite functionthe inverse tangent

Approach:

First introduce an intermediate variable, and then apply the chain rule with respect to each of the two independent variables.

Steps:

  1. Put u=x+y1−xyu = \frac{x+y}{1-xy}, so z=arctan⁡uz = \arctan u
  2. ∂u∂x=(1−xy)−(x+y)(−y)(1−xy)2=1+y2(1−xy)2\frac{\partial u}{\partial x} = \frac{(1-xy) - (x+y)(-y)}{(1-xy)^2} = \frac{1+y^2}{(1-xy)^2}
  3. ∂z∂x=11+u2⋅∂u∂x=(1−xy)2(1−xy)2+(x+y)2⋅1+y2(1−xy)2=1+y21+x2+y2+x2y2=11+x2\frac{\partial z}{\partial x} = \frac{1}{1+u^2} \cdot \frac{\partial u}{\partial x} = \frac{(1-xy)^2}{(1-xy)^2+(x+y)^2} \cdot \frac{1+y^2}{(1-xy)^2} = \frac{1+y^2}{1+x^2+y^2+x^2y^2} = \frac{1}{1+x^2}
  4. By symmetry, ∂z∂y=11+y2\frac{\partial z}{\partial y} = \frac{1}{1+y^2}

(Note: the identity arctan⁡x+y1−xy=arctan⁡x+arctan⁡y\arctan\frac{x+y}{1-xy} = \arctan x + \arctan y gives the result directly.)

Answer: ∂z∂x=11+x2\frac{\partial z}{\partial x} = \frac{1}{1+x^2} and ∂z∂y=11+y2\frac{\partial z}{\partial y} = \frac{1}{1+y^2}

Problem 4

Let z=f(xy,xy)z = f(xy, \frac{x}{y}), where ff has continuous second-order partial derivatives. Find ∂2z∂x∂y\frac{\partial^2 z}{\partial x \partial y}.

second-order partial derivativesa composite functionan abstract function

Approach:

The outer function is a function of two variables and the inner ones are two variables, so the second mixed partial derivative requires expanding all the product terms that the chain rule produces — none of them may be dropped.

Steps:

  1. Put u=xy,v=xyu=xy, v=\frac{x}{y}, and write f1=∂f∂u,f2=∂f∂v,f11=∂2f∂u2,f12=∂2f∂u∂v,f22=∂2f∂v2f_1=\frac{\partial f}{\partial u}, f_2=\frac{\partial f}{\partial v}, f_{11}=\frac{\partial^2 f}{\partial u^2}, f_{12}=\frac{\partial^2 f}{\partial u\partial v}, f_{22}=\frac{\partial^2 f}{\partial v^2}
  2. ∂z∂x=f1⋅y+f2⋅1y\frac{\partial z}{\partial x} = f_1 \cdot y + f_2 \cdot \frac{1}{y}
  3. ∂2z∂x∂y=∂∂y(yf1+1yf2)\frac{\partial^2 z}{\partial x\partial y} = \frac{\partial}{\partial y}(yf_1 + \frac{1}{y}f_2) =f1+y(f11⋅x+f12⋅(−xy2))−1y2f2+1y(f21⋅x+f22⋅(−xy2))= f_1 + y(f_{11}\cdot x + f_{12}\cdot(-\frac{x}{y^2})) - \frac{1}{y^2}f_2 + \frac{1}{y}(f_{21}\cdot x + f_{22}\cdot(-\frac{x}{y^2})) =f1−1y2f2+xyf11−xy3f22= f_1 - \frac{1}{y^2}f_2 + xy f_{11} - \frac{x}{y^3}f_{22} (using f12=f21f_{12}=f_{21}, so the cross terms cancel)

Answer: ∂2z∂x∂y=f1−1y2f2+xyf11−xy3f22\frac{\partial^2 z}{\partial x\partial y} = f_1 - \frac{1}{y^2}f_2 + xyf_{11} - \frac{x}{y^3}f_{22}

Problem 5

Discuss the continuity, the existence of the partial derivatives and the differentiability of f(x,y)={xyx2+y2,(x,y)≠(0,0)0,(x,y)=(0,0)f(x,y) = \begin{cases} \frac{xy}{\sqrt{x^2+y^2}}, & (x,y)\neq(0,0) \\ 0, & (x,y)=(0,0) \end{cases} at (0,0)(0,0).

continuitythe partial derivativesdifferentiabilitya discussion

Approach:

Continuity, the existence of the partial derivatives and differentiability have to be verified separately: first use the squeeze theorem to prove continuity, then compute the partial derivatives, and finally use the linear representation of the increment to test differentiability.

Steps:

  1. Continuity: ∣f(x,y)∣=∣xy∣x2+y2≤x2+y22x2+y2=x2+y22→0|f(x,y)| = \frac{|xy|}{\sqrt{x^2+y^2}} \leq \frac{x^2+y^2}{2\sqrt{x^2+y^2}} = \frac{\sqrt{x^2+y^2}}{2} \to 0, so lim⁡(x,y)→(0,0)f(x,y)=0=f(0,0)\lim_{(x,y)\to(0,0)}f(x,y)=0=f(0,0); the function is continuous.
  2. The partial derivatives: fx(0,0)=lim⁡Δx→00−0Δx=0f_x(0,0)=\lim_{\Delta x\to0}\frac{0-0}{\Delta x}=0, and similarly fy(0,0)=0f_y(0,0)=0; the partial derivatives exist.
  3. Differentiability: Δz−fxΔx−fyΔy=ΔxΔy(Δx)2+(Δy)2\Delta z - f_x\Delta x - f_y\Delta y = \frac{\Delta x \Delta y}{\sqrt{(\Delta x)^2+(\Delta y)^2}} ∣ΔxΔyρ(Δx)2+(Δy)2∣=∣ΔxΔy∣ρ2≤12\left|\frac{\Delta x\Delta y}{\rho\sqrt{(\Delta x)^2+(\Delta y)^2}}\right| = \frac{|\Delta x\Delta y|}{\rho^2} \leq \frac{1}{2}, which does not tend to 0. Taking Δx=Δy=t\Delta x=\Delta y=t: t2t⋅t2=12≠0\frac{t^2}{t\cdot t\sqrt{2}} = \frac{1}{\sqrt{2}} \neq 0, so the function is not differentiable.

Answer: the function is continuous, the partial derivatives exist (both are 0), but it is not differentiable.

III. Directional Derivatives and the Gradient

Problem 6

Find the directional derivative of u=xx2+y2+z2u = \frac{x}{\sqrt{x^2+y^2+z^2}} at the point (1,2,−2)(1,2,-2) in the direction l⃗=(−1,1,0)\vec{l}=(-1,1,0).

the directional derivativethe gradienta function of three variables

Approach:

The directional derivative is the scalar product of the gradient with the unit direction vector.

Steps:

  1. Put r=x2+y2+z2r=\sqrt{x^2+y^2+z^2}, so u=xru=\frac{x}{r}
  2. ux=r−x⋅xrr2=y2+z2r3u_x = \frac{r - x\cdot\frac{x}{r}}{r^2} = \frac{y^2+z^2}{r^3}, uy=−xyr3u_y = -\frac{xy}{r^3} and uz=−xzr3u_z = -\frac{xz}{r^3}
  3. At (1,2,−2)(1,2,-2): r=3r=3, ux=4+427=827u_x=\frac{4+4}{27}=\frac{8}{27}, uy=−227u_y=-\frac{2}{27} and uz=227u_z=\frac{2}{27}
  4. ∇u=(827,−227,227)\nabla u = (\frac{8}{27}, -\frac{2}{27}, \frac{2}{27})
  5. Normalising l⃗\vec{l}: ∣l⃗∣=2|\vec{l}|=\sqrt{2} and l⃗0=(−12,12,0)\vec{l}^0=(-\frac{1}{\sqrt{2}},\frac{1}{\sqrt{2}},0)
  6. ∂u∂l=∇u⋅l⃗0=−8272−2272=−10272=−5227\frac{\partial u}{\partial l} = \nabla u \cdot \vec{l}^0 = -\frac{8}{27\sqrt{2}} - \frac{2}{27\sqrt{2}} = -\frac{10}{27\sqrt{2}} = -\frac{5\sqrt{2}}{27}

Answer: −5227-\frac{5\sqrt{2}}{27}

IV. Extrema and Constrained Extrema

Problem 7

Find the extrema of the function f(x,y)=x3−4x2+2xy−y2f(x,y) = x^3 - 4x^2 + 2xy - y^2.

an extremumthe second derivative testa stationary point

Approach:

First find the stationary points, and then use the second-derivative test to classify them as local maxima, local minima or saddle points.

Steps:

  1. fx=3x2−8x+2yf_x = 3x^2-8x+2y and fy=2x−2yf_y = 2x-2y
  2. From fy=0f_y=0: x=yx=y; substituting into fx=0f_x=0: 3x2−6x=03x^2-6x=0, so x=0x=0 or x=2x=2
  3. The stationary points: (0,0)(0,0) and (2,2)(2,2)
  4. fxx=6x−8f_{xx}=6x-8, fxy=2f_{xy}=2 and fyy=−2f_{yy}=-2
  5. (0,0)(0,0): A=−8,B=2,C=−2,Δ=16−4=12>0,A<0A=-8,B=2,C=-2,\Delta=16-4=12>0,A<0, so a local maximum with f(0,0)=0f(0,0)=0
  6. (2,2)(2,2): A=4,B=2,C=−2,Δ=−8−4=−12<0A=4,B=2,C=-2,\Delta=-8-4=-12<0, so a saddle point

Answer: the local maximum is f(0,0)=0f(0,0)=0, and (2,2)(2,2) is a saddle point.

Problem 8

Find the maximum and the minimum values of z=x2+y2z = x^2 + y^2 on the circular region (x−2)2+(y−2)2≤9(x-\sqrt{2})^2 + (y-\sqrt{2})^2 \leq 9.

the maximum and minimum valuea closed regiona constrained extremumthe Lagrange multiplier method

Approach:

On a closed region the extremum may occur at an interior stationary point or on the boundary curve, so both have to be examined in two rounds and then compared.

Steps:

  1. The interior stationary points: zx=2x=0z_x=2x=0 and zy=2y=0z_y=2y=0, so the stationary point is (0,0)(0,0). Checking (0−2)2+(0−2)2=4≤9(0-\sqrt{2})^2+(0-\sqrt{2})^2=4\leq9, it is in the interior, and z(0,0)=0z(0,0)=0.
  2. The boundary: (x−2)2+(y−2)2=9(x-\sqrt{2})^2+(y-\sqrt{2})^2=9, handled with the Lagrange multiplier method. L=x2+y2+λ[(x−2)2+(y−2)2−9]L=x^2+y^2+\lambda[(x-\sqrt{2})^2+(y-\sqrt{2})^2-9] Lx=2x+2λ(x−2)=0L_x=2x+2\lambda(x-\sqrt{2})=0 and Ly=2y+2λ(y−2)=0L_y=2y+2\lambda(y-\sqrt{2})=0 This gives x=yx=y (for λ≠−1\lambda\neq-1); substituting into the boundary: 2(x−2)2=92(x-\sqrt{2})^2=9, so x=2±32x=\sqrt{2}\pm\frac{3}{\sqrt{2}}
  3. The boundary points: (2+32,2+32)(\sqrt{2}+\frac{3}{\sqrt{2}},\sqrt{2}+\frac{3}{\sqrt{2}}) with z=2(52)2=25z=2(\frac{5}{\sqrt{2}})^2=25, and (2−32,2−32)(\sqrt{2}-\frac{3}{\sqrt{2}},\sqrt{2}-\frac{3}{\sqrt{2}}) with z=2(−12)2=1z=2(-\frac{1}{\sqrt{2}})^2=1
  4. Comparing: the maximum is 2525 and the minimum is 00.

Answer: the maximum value is 2525, and the minimum value is 00.

Problem 9

In the first octant, construct the tangent plane to the ellipsoid x2a2+y2b2+z2c2=1\frac{x^2}{a^2}+\frac{y^2}{b^2}+\frac{z^2}{c^2}=1 so that the volume of the tetrahedron enclosed by the tangent plane and the three coordinate planes is as small as possible, and find the coordinates of the point of tangency.

a constrained extremumthe tangent planethe minimum volumean application

Approach:

The volume of the tetrahedron can be written as a function of the coordinates of the point of tangency, and then the extremum condition gives the point that minimises the volume.

Steps:

  1. The tangent plane at the point of tangency (x0,y0,z0)(x_0,y_0,z_0): x0xa2+y0yb2+z0zc2=1\frac{x_0 x}{a^2}+\frac{y_0 y}{b^2}+\frac{z_0 z}{c^2}=1
  2. The intercepts: a2x0,b2y0,c2z0\frac{a^2}{x_0}, \frac{b^2}{y_0}, \frac{c^2}{z_0}, so the volume V=16⋅a2b2c2x0y0z0V=\frac{1}{6}\cdot\frac{a^2b^2c^2}{x_0y_0z_0}
  3. Minimising VV is equivalent to maximising x0y0z0x_0y_0z_0, subject to x02a2+y02b2+z02c2=1\frac{x_0^2}{a^2}+\frac{y_0^2}{b^2}+\frac{z_0^2}{c^2}=1
  4. L=xyz+λ(x2a2+y2b2+z2c2−1)L=xyz+\lambda(\frac{x^2}{a^2}+\frac{y^2}{b^2}+\frac{z^2}{c^2}-1)
  5. yz+2λxa2=0yz+\frac{2\lambda x}{a^2}=0, xz+2λyb2=0xz+\frac{2\lambda y}{b^2}=0 and xy+2λzc2=0xy+\frac{2\lambda z}{c^2}=0
  6. These give x2a2=y2b2=z2c2\frac{x^2}{a^2}=\frac{y^2}{b^2}=\frac{z^2}{c^2}; substituting into the constraint gives 3⋅x2a2=13\cdot\frac{x^2}{a^2}=1, so x=a3x=\frac{a}{\sqrt{3}}
  7. Similarly y=b3y=\frac{b}{\sqrt{3}} and z=c3z=\frac{c}{\sqrt{3}}

Answer: the coordinates of the point of tangency are (a3,b3,c3)(\frac{a}{\sqrt{3}}, \frac{b}{\sqrt{3}}, \frac{c}{\sqrt{3}}).

V. Geometric Applications

Problem 10

Find the equations of the tangent planes to the surface z=x2+y2z = x^2 + y^2 that are parallel to the plane 2x+4y−z=02x + 4y - z = 0.

the tangent planea parallel planean explicit surface

Approach:

A tangent plane parallel to a given plane must have its normal vector parallel to the normal vector of the given plane, which determines the point of tangency.

Steps:

  1. Put F=x2+y2−zF=x^2+y^2-z; the normal vector ∇F=(2x,2y,−1)\nabla F=(2x,2y,-1)
  2. The normal vector of the given plane is (2,4,−1)(2,4,-1), and the tangent plane is parallel to it, so ∇F∥(2,4,−1)\nabla F \parallel (2,4,-1)
  3. That is, 2x2=2y4=−1−1=t\frac{2x}{2}=\frac{2y}{4}=\frac{-1}{-1}=t, giving x=t,y=2t,z=tx=t, y=2t, z=t
  4. Substituting into the equation of the surface: x2+y2=zx^2+y^2=z gives t2+4t2=tt^2+4t^2=t, so 5t2=t5t^2=t and t=15t=\frac{1}{5} (the case t=0t=0 gives the origin, where the plane z=0z=0 is itself tangent but is not parallel to the given plane)
  5. The point of tangency: (15,25,15)(\frac{1}{5}, \frac{2}{5}, \frac{1}{5})
  6. The tangent plane: 2(x−15)+4(y−25)−(z−15)=02(x-\frac{1}{5})+4(y-\frac{2}{5})-(z-\frac{1}{5})=0, that is, 2x+4y−z−1=02x+4y-z-1=0

Answer: the tangent plane is 2x+4y−z−1=02x+4y-z-1=0 (the plane z=0z=0 is also tangent at the origin but is not parallel to the given plane).


Summary

This set of comprehensive exercises covers the core examination points of multivariable differential calculus:

  • Limits and continuity (2 problems): the squeeze theorem, the path method
  • Partial derivatives and total differentials (3 problems): the chain rule, second-order partial derivatives, the relation between continuity, the partial derivatives and differentiability
  • Directional derivatives and the gradient (1 problem): the directional derivative in three dimensions
  • Extrema and constrained extrema (3 problems): the second-derivative test, extrema on a closed region, the Lagrange multiplier method
  • Geometric applications (1 problem): tangent planes parallel to a given plane

The three core problem types to master are computing partial derivatives and the total differential, the second-derivative test for the extrema, and the Lagrange multiplier method.

Found an issue with this content?

Tell us what could be improved.