I want some tricky elementary integrals. Like tricky tricky and not just regular tricky. Thank you and please.
i cooked up this hard but nice integral when one of our regular users asked a hard integeral that evaluates to her age, I forgot the answer but it should be an integer between 15 and 20 : \[\large \int_0^{\infty} \left(\frac{1944(e^{ax} - e^{bx})}{\ln((a/b)^{54x})(e^{ax}+1)(e^{bx}+1)}\right) dx = \text{your age }\]
Ok I will give these my full focus after I get a yoohoo.
And thanks.
here is another innocent looking yet very tricky integral : \[\int\limits_0^{\pi} \dfrac{dt}{1+2\sin^2t}\]
third integral i like : \[\int{\frac{x^{5}-x}{x^{8}+1}}\:dx\]
@Kainui @ikram002p might have experienced more nice/new integrals :)
if you may allow me tag few others @SithsAndGiggles @tkhunny @amistre64 @geerky42 @nincompoop @dan815 @johnweldon1993 @iambatman @Zarkon
hey @ganeshie8 that second on I'm doing the sub tan(t/2)=u which is ending up to be very long. I'm not really asking for a hint but I guess sort am when I ask is there like a neat little trick to perform to cutout all of that lengthiness. Like just a yes or no question. Don't say anything else for now. Please and thank you. :)
evaluating indefenite integral is a one liner if you divide top and bottom by *something* the tricky part is in evaluating the limits.. again im sure we will be okay if we are careful :)
Here's a favorite of mine: \[\LARGE \int\limits \frac{dx}{1+x^3}\]
I'm no good with integral, but maybe you can find some tricky integrals here: http://math.stackexchange.com/search?q=tricky+integral
@ganeshie8 I enjoyed that first one! I gave this one to a calc 2 class a few days ago and no one has given me an answer so far, everyone wants hints but, oh well, lets see if it's that tricky \[\large \int e^{2x}cos(3x)dx\]
That last one can be done by integration by parts.
I know how to do that integral 3 different ways at least. =P
What are the other two ways to do that one?
that last one is Kainui's favorite one (and mine too) and he somehow manages to work it using only matrices
You can use linear algebra or complex numbers to do it @myininaya
Matrices? Oh may I see? :D lol not very experienced using matrices in integrals!
Sure, so you can look at the derivative of this function and the corresponding sine one as spanning the space of derivatives since they only map onto scalar multiples of each other, so the anti derivative will too, to within a constant of course. So you just invert a 2x2 matrix to get the "integral" matrix! =P
If you have stewart 7e calculus textbook it has pages upon pages with some neat integral problems :P
When I'm not feeling lazy, I can take a pic/ photocopy and send it your way, if you're interested.
I have 6th edition.
Ah ok, cool :).
should have some similar stuff in it since they never really change the books :p
You're probably right, @ganeshie8 how does one approach your first integral?
@Kainui you probably have some tricky matrix way for the 1/(x^3+1) integral instead of this partial fraction/trig sub thing ?
\[\LARGE y_1=e^{2x}\cos(3x) \\ \LARGE y_2=e^{2x}\sin(3x)\] Taking the derivative with the product rule we have \[\LARGE y_1'=2y_1 -3 y_2 \\ \LARGE y_2'=3y_1+2 y_2\] So we can stick this in a matrix like this: \[\LARGE \left[\begin{matrix}2 & -3 \\ 3 & 2\end{matrix}\right] \bar y = \bar y'\] So we invert this matrix to get: \[\LARGE \bar y = \frac{1}{13}\left[\begin{matrix}2 & 3 \\ -3 & 2\end{matrix}\right] \bar y'\] now the vector y' that we're looking to integrate is \[\LARGE \bar y '=\left(\begin{matrix}1 \\ 0\end{matrix}\right)\] So that's just the first column of the matrix, so the integral appears to be \[\LARGE \frac{1}{13}(2 e^{2x}\cos(3x) - 3e^{2x}\sin(3x))\] but upon checking it looks like I messed up a negative sign somewhere since the answer is really \[\LARGE \frac{1}{13}(2 e^{2x}\cos(3x) + 3e^{2x}\sin(3x))\] but you get the idea haha.
Nope, I don't know a fast way to do that integral with 1/1+ x^3 I'm afraid. It's just deceptively easy looking.
That is neat approach for the integral of e^(2x)cos(3x)
It's really pretty quick and painless since there's a trick for inverting 2x2 matrices. It's always nice to see some new alternative too. The more math I do the more I realize how useful linear algebra is haha.
I wish I can give all of you a medal for all your fancy integrals to look at.
Here's another fun trick: \[\LARGE \int\limits \frac{dx}{1-x^2}\] but now let \[\LARGE x= i \tan \theta \\ \LARGE dx = i \sec^2 \theta d \theta\] we get \[\LARGE \int\limits \frac{ i \sec^2 \theta d \theta}{1+\tan^2 \theta}=i \int\limits d \theta= i \theta\]\[\LARGE \int\limits \frac{dx}{1-x^2}=i \tan^{-1} (-ix)\]
@ganeshie8 for some reason I want to write that first integral you wrote as two separate fractions e^(ax)-e^(bx)=e^(ax)+1-e^(bx)-1 but I don't think that is helping me
Wow! i thought of the same thing for first step @myininaya : \[\large \large \int \dfrac{1}{x(e^{bx}+1)} - \int \dfrac{1}{x(e^{bx}+1)} = F(b) - F(a) \] \]
**** \[\large \int \dfrac{1}{x(e^{bx}+1)} - \int \dfrac{1}{x(e^{ax}+1)} = F(b) - F(a)\]
yeah times that constant 36
Ahh yes and some log stuff also i think..
oh yeah over ln(a/b)
Did you just give me some weird hint with the F(b)-F(a) thing?
I didn't really mean your hint was weird if it was a hint.
or even if it wasn't a hint
The F(b)-F(a) confuses me though because why would those integrals be constants
differentiate F(b) with respect to b, ( notice that we can commute differentiation and integration : \(\large \dfrac{d}{db} \int dx = \int \dfrac{d}{db}~dx\)) do the same thing for F(a).. thats one method ^^ there is another quick method also
** spoiler ** solution using that long way worked by @ShailKumar http://assets.openstudy.com/updates/attachments/53d488bce4b0f2f7e5e971e2-shailkumar-1406447288904-diffunderintegralsign.jpg
That's crazy. (as in crazy neat)
here is a more neat+short solution :) http://openstudy.com/users/vishweshshrimali5#/updates/53d488bce4b0f2f7e5e971e2
Have fun :D
Proving the following integral identity is fairly tricky if you haven't done it before: \[\int_{-\infty}^\infty e^{-x^2}~dx=\sqrt{\pi}\]
well , using complex makes any integral solvable and able to solve hmm but i like the one that @sithangiggles mentioned
\[\int\limits_{0}^{2\pi}\frac{1}{1+2\sin^2(x)}dx=4\int\limits_{0}^{\frac{\pi}{2}}\frac{1}{2-\cos(2x)} dx \\ \tan(x)=t \\ \sec^2(x) dx=dt \\ dx=\cos^2(x) dt \\ dx=\frac{dt}{t^2+1} \\ 4 \int\limits_{0}^{\infty} \frac{1}{t^2+1}\frac{1}{2- \frac{1-t^2}{1+t^2}} dt \\ =4\int\limits_{0}^{\infty}\frac{1}{2(t^2+1)-(1-t^2)} dt \\ =4\int\limits_{0}^{\infty}\frac{1}{3t^2+1} dt \\ =4 \int\limits_{0}^{\infty}\frac{1}{(\sqrt{3}t)^2+1} dt \\ \lim_{a \rightarrow \infty} \frac{4}{\sqrt{3}} \arctan(\sqrt{3}x)|_0^{a} \\ \lim_{a \rightarrow \infty}\frac{4}{\sqrt{3}} \arctan(\sqrt{3}a)=\frac{4}{\sqrt{3}} \cdot \frac{\pi}{2}=\frac{2 \pi }{\sqrt{3}}\]
my age might change by the time i solve it though have u considered that possibility!
Ahh very nicely worked @myininaya :) I ran into weird problems when I first attempted it.. here are few more alternative methods : http://math.stackexchange.com/questions/958920/evaluate-int-limits-0-pi-fracdx12-sin2x
@dan815 that integral gives your age correctly upto an uncertainty of 100 years.. so we are fine technically speaking ;p good point though :)
:3
@ganeshie8 y'=(x^5-x)/(x^8+1) has got me stumped
And yeah that one I seen the way I first tried to work out and someone totally finished it that way but I wasn't about to because it had ugly factorization
And I manipulated that one substitution tan(x/2)=t since I had a double angle
nice thread on integrals
Join our real-time social learning platform and learn together with your friends!