Newton's Method for Finding the Area of a Curve: A Tricky Homework Problem

Click For Summary

Homework Help Overview

The problem involves finding a value of k such that the area under the curve of ln(x) from 1 to e is divided in half by the line y = k. The original poster is using Newton's Method to solve for k after setting up the integral to compute half the area.

Discussion Character

  • Exploratory, Assumption checking, Problem interpretation

Approaches and Questions Raised

  • The original poster attempts to set up the integral and apply Newton's Method but expresses confusion about the relationship between k and the area. Some participants question the clarity of the problem statement and the setup of the integral.

Discussion Status

Participants are exploring different interpretations of the problem, with some providing guidance on the integral setup and the application of Newton's Method. There is a recognition of potential misunderstandings regarding the integral of ln(x) and its evaluation.

Contextual Notes

There are indications of confusion regarding the integral's evaluation and the relationship between the area and the line y = k. Some participants suggest checking the original problem statement for accuracy.

becca4
Messages
20
Reaction score
0

Homework Statement


Problem:
A (area) = [tex]\int^{e}_{1}[/tex] ln(x) dx = 1

Now we let k be such that 0 [tex]\leq[/tex] k [tex]\leq[/tex]1

Consider the line y = k.

Find k so that area computed by A is exactly one half.

Homework Equations



So, first, I found point of intersection:

k = ln (x)

e[tex]^{k}[/tex] = x​

Now I have:

1/2 = [tex]\int^{e}_{e^{k}}[/tex] ln (x) dx​



The Attempt at a Solution



I'm having a hard time grasping the big picture of this, that's pretty much why I'm stuck. I know that Newton's Method is used to find the roots of a function, but this area twist is really giving me a hard time. I'm supposed to use N.M. to solve for k after setting up integral to compute A/2. THEN I have to experiment to find 2A/3 and A/1000. Can anyone enlighten me? Where is this zero happening?? Am I supposed to use [tex]\int[/tex] ln (x) -k dx??

HELP!
 
Physics news on Phys.org
The problem, as stated, doesn't make a whole lot of sense. The "area computed by A" does not depend on k. Do you mean to say that y= k divides the area computed by A in half? That's different from saying that [itex]\int_{e^k}^e ln(x)dx= 1/2[/itex] which has nothing to do with A.

In any case, you will have to integrate [itex]\int ln(x) dx[/itex]. Can you do that?
 
I copied the problem verbatim from the handout I was given, so I'm sorry that it doesn't make a whole lot of sense to you...

Yes, I mean to say that the line y = k splits the area in half. And yes, I know how to integrate ln(x), but it doesn't do me a whole lot if I don't know what I'm doing. :o(


So I have:

[tex]\int^{e}_{e^{k}}[/tex] ln (x) dx = x ln x - x ][tex]^{e}_{e^{k}}[/tex]

= ( e ln(e) - e ) - (e[tex]^{k}[/tex] ln(e[tex]^{k}[/tex] )

= (e*1 -e) - ( e[tex]^{k}[/tex]*k)

= - ( e[tex]^{k}[/tex]*k)
Since that integral is equal to 1/2 (since the line cuts it in half)

1/2 = -k * e[tex]^{k}[/tex]

0 = -k * e[tex]^{k}[/tex] - 1/2 <--- Is this supposed to be the equation I use N.M. on?
 
becca4 said:
I copied the problem verbatim from the handout I was given, so I'm sorry that it doesn't make a whole lot of sense to you...

Yes, I mean to say that the line y = k splits the area in half. And yes, I know how to integrate ln(x), but it doesn't do me a whole lot if I don't know what I'm doing. :o(


So I have:

[tex]\int^{e}_{e^{k}}[/tex] ln (x) dx = x ln x - x ][tex]^{e}_{e^{k}}[/tex]

= ( e ln(e) - e ) - (e[tex]^{k}[/tex] ln(e[tex]^{k}[/tex] )

= (e*1 -e) - ( e[tex]^{k}[/tex]*k)​

You forgot the "-x". the integral is
[tex]-ke^k+ e^k[/tex]
= - ( e[tex]^{k}[/tex]*k)
and so this is [itex]e^k(1-k)[/itex]
Since that integral is equal to 1/2 (since the line cuts it in half)

1/2 = -k * e[tex]^{k}[/tex]

0 = -k * e[tex]^{k}[/tex] - 1/2 <--- Is this supposed to be the equation I use N.M. on?
No, because you neglected the "-x" your equation is
[tex]e^k (1-k)= 1/2[/tex]
 
Last edited by a moderator:
Ok, yeah, that makes sense. So now,

[tex]k - ( e^k (1-k) ) / d/dk ( e^k (1-k) )[/tex]​

should give me what k converges to, meaning the value of k that when 1/2 is subtracted, equals to zero, right??

So for 2/3, I'd use this equation to solve for k using Newton's method,
[tex]e^k (1-k)= 2/3[/tex]​

and for 1/1000,
[tex]e^k (1-k)= 1/1000[/tex]​

Am I on the right track?
 
Are you 100% sure that your original integral wasn't [tex]\int^e_1 \frac{1}{x} dx = 1[/tex] instead? Because [tex]\int^e_1 \log x dx = -1[/tex] in case you didn't check with the anti derivative... Also the integral of 1/x from t to 1, wrt t, is a common definition of the natural logarithm, so you might have just typed down something you were thinking about a few lines ahead. Just check up on the question.
 
Since log(x) (ln(x)) is positive for x> 1, how do you get -1 as its integral? I get 1 for the integral!
 
That's a good point...

[tex]\int^e_1 \log x dx = x(\log x -1) \right|^e_1 = e(1-1) - 1(\log 1 -1)= 0 - (\log 1 -1) = 1[/tex]..

Great, just great, i forgot the brackets on the log 1 -1 when I was doing it in my head, now I've made an ass of myself =]. Please ignore any thing i have said.
 
So I am on the right track?
 
  • #10
Yes probably, seeing as Halls is helping you
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 14 ·
Replies
14
Views
3K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 12 ·
Replies
12
Views
3K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 14 ·
Replies
14
Views
2K
Replies
9
Views
3K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 9 ·
Replies
9
Views
4K
  • · Replies 4 ·
Replies
4
Views
2K