1002微微微甲甲甲01-05班班班期期期中中中考考考解解解答答答和和和評評評分分分標標標準準準 1. (16 points) Prove that the series
∞
X
n=2
(−1)n n(n − 1)
1
2n converges absolutely, and find its sum.
Solution:
Part I (6pts) Prove the series converges absolutely
let an=
∞
X
n=2
(−1)n n(n − 1)
1 2n
then the series convergent absolutely if
∞
X
n=2
|an| converges
use ratio test:
n→∞lim |an+1
an | = ( 1
(n + 1)n2n+1/ 1
(n)(n)2n) =1 2 < 1 use comparison test:
compared an to bn = 1
n2 or bn = 1 2n
show that an< bn everywhere and prove that
∞
X
n=2
|bn| converges
Part II (10pts)find its sum
f (x) = 1
1 + x = 1 − x + x2− x3+ x4.... =
∞
X
n=0
(−1)nxn
g(x) =
Z 1
1 + xdx = ln(1 + x) = x −1 2x2+1
3x3.... + c1 =
∞
X
n=1
(−1)n−11
nxn+ c2 For x = 0 → c1 = c2 = 0
h(x) = Z
ln(1 + x)dx = 1
1 × 2x2− 1
2 × 3x3+ 1
3 × 4x4+ .... + c3 =
∞
X
n=2
(−1)n 1
n(n − 1)xn+ c4
(1 + x) ln(1 + x) − (1 + x) =
∞
X
n=2
(−1)n 1
n(n − 1)xn+ c4 For x = 0 → c4 = −1
let x =1
2 ∞
X
n=2
(−1)n 1
n(n − 1)xn= 3 2ln3
2 −1 2
2. (a) (8 points) The Binomial Theorem implies that
(1 − x)−12 = 1 +
∞
X
n=1
(2n)!
kn(n!)2xn
for some constant k. Find k, and find the interval of convergence of the power series.
(b) (8 points) Estimate the error if one uses x = −1
4, and the first five non-zero terms in (a) to approximate 1
√ 5.
Solution:
Page 1 of 4
(a) (1 − x)−12 =
∞
X
n=0
(2n)!
4nn!2xn. So k = 4.(2 points) Let an = (2n)!
4nn!2. R = lim an
an+1 = 1. (2 points)
At x = −1, since an is decreasing and converge to 0,X
an(−1)n converges. (2 points) At x = 1, sinceX
an≥X
anxn= (1 − x)−12 for all x ∈ (−1, 1) and lim
x→1−
(1 − x)−12 = ∞.
Xan diverges. (2 points)
Hence, the interval of convergence is [−1, 1).
(b) 1
√5 = 1
2((1 − (−1
4))−12 = 1 2(
∞
X
n=0
(2n)!
4nn!2(−1
4)n) (1 point) Since it is an alternating series (2 points),
| 1
√5 −1 2(
∞
X
n=0
(2n)!
4nn!2(−1
4)n)| ≤ 10!
2 ∗ 410(5!)2 = 126
410(5 points)
3. (16 points) Consider the curve
r(t) = t2 i + (sin t − t cos t) j + (cos t + t sin t) k, t ≥ 0 Find T(t), N(t), B(t), the curvature κ(t) and the torsion τ (t).
Solution:
r(t) = (t2, sint − tcost, cost + tsint)
r0(t) = v(t) = (2t, cost + tsint − cost, −sint + tcost + sint) = (2t, tsint, tcost) (1% ) r00(t) = a(t) = (2, sint + tcost, cost − tsint)
r000(t) = a0(t) = (0, 2cost − tsint, −2sint − tcost)
|r0(t)| =p
4t2+ t2sin2t + t2cos2t =√
5t ( 1% ) T (t) = r0(t)
|r0(t)| = 1
√5(2, sint, cost) =
√5
5 (2, sint, cost) ( 2% ) To get full points (3%) please answer both question correctly.
T0(t) = 1
√5(0, cost, −sint) , |T0(t)| = 1
√5
pcos2t + sin2t = 1
√5 N (t) = T0(t)
|T0(t)| = B(t) · N (t) = (0, cost, −sint) ( 3% ) To get full points (3%) please answer both question correctly.
κ(t) = |T0(t)|
|r0(t)| = |v(t) × a(t)|
[v(t)]3 = 1
√5
√1 5t= 1
5t ( 3% ) To get full points (3%) please answer both question correctly.
v(t) × a(t) = (−t2, 2t2sint, 2t2cost) (1% )
|v(t) × a(t)| =p
(−t2)2+ (2t2sint)2+ (2t2cost)2=p
t4+ 4t4=√ 5t2 If your calculation is wrong, you will receive some partial credit.
B(t) = v(t) × a(t)
|v(t) × a(t)| = 1
√5(−1, 2sint, 2cost) =
√5
5 (−1, 2sint, 2cost) ( 2% ) To get full points (3%) please answer both question correctly.
τ (t) = [v(t) × a(t)] · a0(t)
|v(t) × a(t)|2 =2t2sint(2cost − tsint) + 2t2cost(−2sint − tcost) 5t4
=(4t2sint · cost − 2t3sin2t) + (−4t2sint · cost − 2t2cost2t) 5t4
=−2t3 5t4 = −2
5t ( 3% )
I will also grant partial credit for partial solutions and solutions with minor flaws. I will give no credit for wildly incorrect answers which are obviously only there in the hopes of getting partial credit.
Page 2 of 4
4. (16 points) Let F (x, y, z) = x2+ 2z + Z z
y
p3
(t2+ 7)y2dt. Find the tangent plane of the surface F (x, y, z) = 2 at the point (2, −1, −1).
Solution:
F 1 =∂F
∂x = 2x (1 points) F 2 =∂F
∂y =2 3y−13
Z z y
p3
t2+ 7dt − y23p3
y2+ 7 (5 points) F 3 =∂F
∂z = 2 + y23p3
z2+ 7 (4 points) F 1(2, −1, −1) = 4 (1 points)
F 2(2, −1, −1) = −2 (2 points) F 3(2, −1, −1) = 4 (1 points)
4(x − 2) − 2(y + 1) + 4(z + 1) = 0 (2 points)
5. Suppose that z = f (x, y) has continuous second order partial derivatives, and x = s2− t2, y = 2st. Define F (s, t) = f (s2− t2, 2st)
(a) (6 points) Express Fs, Ftin terms of fx, fy, s and t.
(b) (10 points) Show that Fss+ Ftt= h(s, t)(fxx+ fyy) for some function h(s, t). Find h(s, t) explicitly.
Solution:
5.(a)
Fs(s, t) = 2sfx(s2− t2, 2st) + 2tfy(s2− t2, 2st)...(3pts) Ft(s, t) = −2tfx(s2− t2, 2st) + 2sfy(s2− t2, 2st)...(3pts) (b)
Fss(s, t) = 2fx+ 2s[2sfxx + 2tfxy] + 2t[2sfxy+ 2tfyy]
= 2fx+ 4s2fxx+ 8stfxy + 4t2fyy...(3pts) Ftt(s, t) = −2fx+ 4t2fxx− 8stfxy+ 4s2fyy...(3pts)
Fss(s, t) + Ftt(s, t) = 4(s2 + t2)fxx+ 4(s2+ t2)fyy...(2pts)
= 4(s2 + t2)(fxx+ fyy) ⇒ h(s, t) = 4(s2+ t2)...(2pts)
6. Let f (x, y, z) = yx2+ xz2− y.
(a) (10 points) Find all critical points of f (x, y, z) and classify them.
(b) (10 points) Find the maximum and minimum of f on the region x2+ y2+ z2≤ 1.
Solution:
(a)
f1= 2xy + z2 (1 point) f2= x2− 1 (1 point) f3= 2xz (1 point)
⇒ critical points at f1= f2= f3= 0
⇒ (±1, 0, 0) (2 points)
Hessian
2y 2x 2z
2x 0 0
2z 0 2x
(1 point) ⇒
0 ±2 0
±2 0 0
0 0 ±2
(2 points)
def H 6= 0 ⇒ neither positive definite nor negative definite
⇒ both saddle points (2 points)
Page 3 of 4
(b)
x2+ y2+ z2− 1 = g(x, y, z) ≤ 0
Solve the system
∇f = λ∇g (1 point) g = 0
2xz + z2= 2λx (1 point) x2− 1 = 2λy (1 point) 2xz = 2λz (1 point) x2+ y2+ z2= 1
case 1: z = 0, then 2xy = 2λx. If x = 0, y = ±1 ⇒ (0, ±1, 0) ⇒ f (0, ±1, 0) = ∓1 If x 6= 0, y = λ ⇒ x2= 2y2+ 1 = 1 − y2 ⇒ (±1, 0, 0) ⇒ f (±1, 0, 0) = 0 (1 point)
case 2: z 6= 0, x = λ ⇒ x2= 2xy + 1 = (2x2− z2) + 1 ⇒ (0, 0, ±1) ⇒ f (0, 0, ±1) = 0 (1 point) max = 1 at (0, −1, 0) (2 points), min = −1 at (0, 1, 0) (2 points)
Page 4 of 4