Evaluate $$\iint_A xy\,dx\,dy$$ where $A$ is bounded by $x^2+y^2-2x=0$, $y^2=2x$ and $y=x$.

After I sketched the curves, I found these points,

enter image description here So by Integrating the above $xy$ by $dy\,dx$ I will get the area enclosed by those curves, right? there are 3 points and I'm confused how should I solve this problem.

if I consider the limits upper $(2,2)$ and lower limits $(0,0)$ is that correct?

Please help me by pointing me to the right direction

  • $\begingroup$ Try converting to polar coordinates. Then $\theta$ will be in $[\frac {\pi}{4},\frac {\pi}{2}]$ and $r$ in $[2\cos\theta, \csc\theta \cot\theta].$ Alternatively, break the region in two, one with $x$ in [0,1], and one with $x$ in $(1,2]$ $\endgroup$ – Doug M Sep 13 '17 at 18:11
  • $\begingroup$ @DougM thanks for the answer. but we weren't taught about the polar coordinates. I still didn't get your second method. let's say I divided it. what will be the limits then? and I still must use xy dxdy for the both parts? Sorry i'm not very familiar with multiple integrations as I was taught them recently $\endgroup$ – user2511652 Sep 13 '17 at 18:18


To solve the leftmost curve for $x$ note it is equivalent to $$(x-1)^2 = 1-y^2$$ and therefore, as you are above the $x$-axis, you get $$x = 1 + \sqrt{1-y^2}.$$

From your picture then, $$ \iint_A dA = \int_{y=0}^{y=1} \int_{x=1 + \sqrt{1-y^2}}^{x=y^2/2} dx\,dy + \int_{y=1}^{y=2} \int_{x=1 + \sqrt{1-y^2}}^{x=y} dx\,dy $$

Can you take it from here?

  • $\begingroup$ I understand it now sir. thank you very much for your answer $\endgroup$ – user2511652 Sep 13 '17 at 18:27
  • $\begingroup$ btw your method gives me 'indeterminate' for the answer when I solved it in using 'Microsoft Mathematics' $\endgroup$ – user2511652 Sep 14 '17 at 2:20
  • $\begingroup$ @user2511652 not sure about the software nor about how you used it... $\endgroup$ – gt6989b Sep 14 '17 at 3:07

If you have not yet learned how to transform the coordinate system in double integration...

The region bound by the line the circle and the parabola can be broken into two.

$\displaystyle \int_0^1 \int_{\frac 12y^2}^{\sqrt{1 -y^2}+1} xy\ dx\ dy + \int_1^2 \int_{\frac 12 y^2}^y xy\ dx\ dy$


$\displaystyle \int_0^1 \int_{\sqrt{2x -x^2}}^{\sqrt {2x}} xy\ dy\ dx + \int_1^2 \int_{x}^{\sqrt {2x}} xy\ dy\ dx$

and in polar coordinates

$\displaystyle \int_{\frac \pi4}^\frac\pi2 \int_{2\cos\theta}^{2\csc\theta\cot\theta} r\sin\theta\cos\theta \ dr\ d\theta$

  • $\begingroup$ thanks for the answer. I will try learn much more as possible. I will also read up on polar coordinates. thanks again $\endgroup$ – user2511652 Sep 13 '17 at 18:25
  • $\begingroup$ Finding the correct limits is usually the difficult part when performing double (triple) integration. I like to imagine a line that crosses my region (vertical or horizontal whichever appears will be easier) , and then try to find equations that would define the endpoints of that line. $\endgroup$ – Doug M Sep 13 '17 at 18:27
  • $\begingroup$ Yeah, it just confused me on this problem too. ah, i see. so it does the trick :) thank you $\endgroup$ – user2511652 Sep 13 '17 at 18:32
  • $\begingroup$ in the second part of your equation, shouldn't the limits to be from to x to squareroot of 2x ? $\endgroup$ – user2511652 Sep 14 '17 at 9:25
  • $\begingroup$ As pointed out by @user2511652, the limits in the second integral are wrong. So are the limits in the first integral. The integral goes from $\operatorname{max}\{\sqrt{2x-x^2},x\}$ to $\sqrt{2x}$ for $x\in[0,\,2]$. $\endgroup$ – Tom-Tom Sep 14 '17 at 13:08

If the portion of $A$ where $0\le x\le 1$ is broken into infinite many vertical Riemann rectangles, then the height of each rectangle is $h_1=\sqrt{2x}-\sqrt{2x-x^2}$ and its width is $dx$.

If the portion of $A$ where $1\le x\le 2$ is broken into infinite many vertical Riemann rectangles, then the height of each rectangle is $h_1=\sqrt{2x}-x$ and its width is $dx$.

Thus the total area of $A$ is

$$\begin{align} \operatorname{area}(A) &= \int_0^1 h_1 \, dx + \int_1^2 h_2 \, dx \\ &= \int_0^1 \left( \sqrt{2x}-\sqrt{2x-x^2} \right) \, dx + \int_1^2 \left( \sqrt{2x}-x \right) \, dx \\ \end{align}$$

Not that integrating $\sqrt{2x-x^2}$ is going to be easier than other approaches.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.