Yahoo Answers is shutting down on May 4th, 2021 (Eastern Time) and beginning April 20th, 2021 (Eastern Time) the Yahoo Answers website will be in read-only mode. There will be no changes to other Yahoo properties or services, or your Yahoo account. You can find more information about the Yahoo Answers shutdown and how to download your data on this help page.
Trending News
Tiny integral with huuuuuuge integration?
Integrate ∫ (9 + 4/√x )^3/2 dx.
I have two solutions. So I am more interested in the shortest solution :)
What method do you use to integrate it and what is the final answer?
David, my first thought was x = u^2 too, but sadly it was no good at all.
I haven't tried other two methods. Exponent? Hmmm... you can try :) And this integral doesn't remind me of integration by parts but ... who knows???
Yeah, that's one I used and it is... exactly as you said :)))) but gives the answer!
to maximammal: I haven't tried hyperbolic substitutions too. So as I see it is definitely easier to integrate exponents though I wouldn't say it is much shorter.
Likely the short way doesn't exist at all.
What a naughty integral - and looks so innocent :)
Summary:
1. Standard method (we all tried it and it gives the answer) - sub x = cot^4(t) which reduces integral to the form ∫ du/(1-u²)³. Problems with expressing the answer in terms of x.
I tried integration of differential binomial with the same luck.
2. Integration using hyperbolic functions - sub x = sinh^4(t), which reduces integral to the form ∫ cosh^4(t) dt. Then there goes easy integration of the exponents, but the answer is too complicate because of all that hyperbolic and inverse hyperbolic functions.
3. To my great surprise I found out that the best way to integrate this integral is twice by parts (David's idea) and then sub x = t². Any complications at all and we get the answer in the terms of x:
http://i633.photobucket.com/albums/uu51/labasrasa/...
SO THANK YOU ALL, YOU ARE GREAT!!!
I am not able to choose the best answer cause all of you gave me ideas and insights and helped to find optimal solution.
Thanks! :)
4 Answers
- Scrander berryLv 71 decade agoFavorite Answer
∫ (9 + 4/√x )^3/2 dx
I like the suggestion of subbing x = cot^4
how abotu this
sub x = 16cot^4(u)/81
(so √x = 4cot²(u)/9, and 4/√x = 9tan²(u))
dx = -64cot²(u)csc²(u) du / 81
∫ (9 + 4/√x )^3/2 dx
= ∫ (9 + 9tan²(u))^3/2 (-64cot²(u)csc²(u) du / 81)
= -64/3 * ∫ (1 + tan²(u))^3/2 (cot²(u)csc²(u) du)
= -64/3 * ∫ (sec²(u))^3/2 cot²(u)csc²(u) du
= -64/3 * ∫ sec³(u)cot²(u)csc²(u) du
= -64/3 * ∫ cot(u)csc³(u)du
= 64/3 * ∫ csc²(u) [-cot(u)csc(u)du]
sub v = csc(u)
dv = -csc(u)cot(u)du
= 64/3 * ∫ v² dv
= 64v³/9 + C
= 64csc³(u)/9 + C
= 64csc³(u)/9 + C
x = 16cot^4(u)/81
cot^4(u) = 81x/16
cot(u) = 3x^(1/4)/2 (=A/O)
csc(u) = H/O
= √(9√x + 4) / 2
= 64√(9√x + 4)³/72 + C
Sorry, I really rushed this due to lack of time and apparently this answer is incorrect (according to online integrators) but I don't have time to fix it now and I leave the city for a week tomorrow So I can't come back and check it (although if I find a computer to use I will try)
I do believe in this method though, might be worth someone revisiting.
- 1 decade ago
I'll make a couple of substitutions, the first involving a power function and the second involving a hyperbolic function. Will skip over many details since you guys are obviously calculus experts. :)
First, let x = (16/81)t^4 then the integral becomes â« (t^2+1)^(3/2) (64/3) dt. Let's temporarily ignore the multiplicative constant in the following work for convenience.
Next, let t = sinh(p), dt = cosh(p) dp. Note that cosh²(p) = 1 + sinh²(p). So the integral now becomes ⫠(cosh^4)(p) dp. Note that by definition cosh(p) = (e^p + e^(-p))/2. So we can do a binomial expansion of (cosh^4)(p) to get [(e^(4p) + e^(-4p)) + 4(e^(2p) + e^(-2p)) + 6]/16. Each of the curve-bracketed terms can then be converted back to cosh'es i.e. [2cosh(4p) + 8cosh(2p) + 6]/16. This can now be easily integrated as sinh(4p)/32 + sinh(2p)/4 + 3p/8.
So now we just have to reverse both substitutions. Let inv_sinh be the inverse hyperbolic sine (it is a logarithm). First the hyperbolic sub: sinh(4 inv_sinh(t))/32 + sinh(2 inv_sinh(t))/4 + 3 inv_sinh(t)/8. Note that it is possible to simplify the sinh-inv_sinh terms but I won't bother here. Next the power sub: sinh(4 inv_sinh((3/2)x^(1/4)))/32 + sinh(2 inv_sinh((3/2)x^(1/4)))/4 + 3 inv_sinh((3/2)x^(1/4))/8
So the final answer should be (64/3)[sinh(4 inv_sinh((3/2)x^(1/4)))/32 + sinh(2 inv_sinh((3/2)x^(1/4)))/4 + 3 inv_sinh((3/2)x^(1/4))/8] + C
Hopefully there are no numerical errors but let me know if there are.
Source(s): http://mathworld.wolfram.com/HyperbolicFunctions.h... http://mathworld.wolfram.com/InverseHyperbolicFunc... - DavidGLv 51 decade ago
hey Tabla,
how about a Trigonometric/Hyperbolic substitution?
ie.
for simplicity I will just consider
int ( ( 1 + 1/sqrt(x) ) ^ (3/2) dx )
let x = cot^4(s) -->dx = -4cot^3(s)cosec^2(s)
Thus,
int( ( 1 + tan^2(s))^(3/2) * -4cot^3(s)cosec^2(s) ds)
= -4*int( sec^3(s) * cot^3(s)cosec^2(s) ds )
= -4*int(cosec^5(s) ds)
which is a pain in the **** with Integration by parts, but achievable
What are your thoughts?
David