Undergrad What's my mistake in this integration problem?

Click For Summary
SUMMARY

The integral ##\int_0^{2\pi} \cos^{-1}(\sin(x)) \mathrm{d}x## evaluates to ##\pi^2##, contrary to initial calculations suggesting it equals 0. The mistake arises from improper handling of the limits during substitution. The function ##\cos^{-1}(\sin(x))## is periodic with period ##2\pi##, necessitating the splitting of the integral into intervals where the function behaves consistently. The correct evaluation involves recognizing the periodic nature of both sine and arccosine functions.

PREREQUISITES
  • Understanding of integral calculus, specifically u-substitution and integration by parts.
  • Familiarity with the properties of periodic functions, particularly sine and arccosine.
  • Knowledge of how to evaluate definite integrals over multiple intervals.
  • Ability to interpret graphical representations of functions and their areas.
NEXT STEPS
  • Study the properties of periodic functions in calculus, focusing on sine and arccosine.
  • Learn about splitting integrals for functions with varying behavior over different intervals.
  • Explore advanced techniques in integration, including integration by parts and substitution methods.
  • Practice evaluating definite integrals involving inverse trigonometric functions.
USEFUL FOR

Mathematics students, educators, and professionals involved in calculus, particularly those focusing on integration techniques and the analysis of periodic functions.

murshid_islam
Messages
468
Reaction score
21
TL;DR
My calculation of the integral leads to 0 when I do u-substitution. But from the graph of the function, I can see that the area is obviously not 0.
Here's the problem: ##\int_0^{2\pi} \cos^{-1}(\sin(x)) \mathrm{d}x##
If I do the substitution ##u = \sin(x)##, both the limits of integration become 0 and the integral would result in 0. But the graph of the function tells me the area isn't 0. Where am I going wrong?
 
Last edited:
Physics news on Phys.org
murshid_islam said:
TL;DR Summary: My calculation of the integral leads to 0 when I do u-substitution. But from the graph of the function, I can see that the area is obviously not 0.

Here's the problem: ##\int_0^{2\pi} \cos^{-1}(\sin(x)) \mathrm{d}x##
If I do the substitution ##u = \sin(x)##, both the limits of integration become 0 and the integral would result in 0. But the graph of the function tells me the area isn't 0. Where am I going wrong?
The new variable in a substitution must be an interval. Otherwise, with an appropriate substitution you could make the limits equal on any integral.
 
Ok, I tried it with integration by parts with ##u = \cos^{−1}(\sin(x))## and ##\mathrm{d}v = \mathrm{d}x##, which gives us ##\mathrm{d}u = \frac{−\cos(x)}{\sqrt{1−\sin^2 x}}\mathrm{d}x = −\mathrm{d}x## and ##v=x##.
The integral becomes ##x\cos^{−1}(\sin(x)) + \int x \mathrm{d}x = x\cos^{-1}(\sin(x)) + \frac{x^2}{2} + C##
Evaluating that from 0 to ##2\pi##, we get ##\left(2\pi\cos^{−1}(0) − 0\right) + \frac{1}{2}\left(4\pi^2 − 0\right) = \pi^2 + 2\pi^2 = 3\pi^2##

But the actual answer seems to be ##\pi^2##. What am I missing?
 
That arccos is periodic, of period exactly ##2\pi## and that also sin is periodic. So you must be careful when computing the derivative of ##u## between the limits.
 
dextercioby said:
That arccos is periodic, of period exactly ##2\pi## and that also sin is periodic. So you must be careful when computing the derivative of ##u## between the limits.
The range of sin(x) is the interval [-1, 1], which is also the domain of arccos(x). I'm not being able to figure out how that will affect ##\mathrm{d}u##. What will ##\mathrm{d}u## be if ##u = \arccos(\sin x)##?
 
Last edited:
murshid_islam said:
The range of sin(x) is the interval [-1, 1], which is also the domain of arccos(x). I'm not being able to figure out how that will affect ##\mathrm{d}u##. What will ##\mathrm{d}u## be if ##u = \arccos(\sin x)##?
You have to split the integral, so that ##u## in each case takes values from a simple interval. We look at the values of ##\sin x##:

##x \in [0, \frac \pi 2]## gives ##\sin x \in [0, 1]##
##x \in [\frac \pi 2, \frac {3\pi} 2]## gives ##\sin x \in [1, -1]##
##x \in [\frac {3\pi} 2, 2\pi]## gives ##\sin x \in [-1, 0]##

To illustrate the point. Consider the integral: ##\int_0^1 x \ dx## and use the substitution ##u = x(1-x)##. This transforms the integeral to: ##\int_0^0 f(u) \ du## for some function ##f##, which you can work out. But, we've screwed the limits on the integral.
 
What about using ##\sin x = \cos(x - \frac \pi 2)##?
 
murshid_islam said:
But the actual answer seems to be ##\pi^2##. What am I missing?
I agree with the answer ##\pi^2##. As above, the integrand takes a different form across the interval of integration. So, you have to identify this and split the integral up.
 
Here's a quick way. Note that the function ##\cos^{-1}(\cos x)## has period ##2\pi##:

For ##x \in [0, \pi]## we have ##\cos^{-1}(\cos x) = x##

For ##x \in [\pi, 2\pi]## we have ##\cos^{-1}(\cos x) = 2\pi - x##

A graph of this function is a triangle of base ##2\pi## and height ##\pi##, hence area ##\pi^2##. Hence:
$$\int_0^{2\pi} \cos^{-1}(\cos x) \ dx = \pi^2$$As the function is periodic, for any ##a##, we have:
$$\int_a^{a + 2\pi} \cos^{-1}(\cos x) \ dx = \pi^2$$Finally, using ##\sin x = \cos (x - \frac \pi 2)##, we have:
$$\int_0^{2\pi} \cos^{-1}(\sin x) \ dx = \int_0^{2\pi} \cos^{-1}(\cos(x - \frac \pi 2)) \ dx$$$$= \int_{-\frac \pi 2}^{\frac {3\pi} 2} \cos^{-1}(\cos u) \ du = \pi^2$$
 
  • Like
Likes berkeman and dextercioby

Similar threads

  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 17 ·
Replies
17
Views
5K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K