You are not logged in.
Pages: 1
Hi there,
i'm getting frustrated by a lot of the integration material in my latest maths book. I feel like I understand all the topics but there are some questions which just throw me completely. I'll just list two here:
1. Show that
2. Given that
show that, for n ≥ 2,
The second question obviously involves integration by reduction, but I can't for the life of me see how to work either of them out.
Any help appreciated!
Student: "What's a corollary?"
Lecturer: "What's a corollary? It's like when a theorem has a child. And names it corollary."
Offline
for problem 1. let u = sqrt(1 + x), then u^2 = x + 1, so x + 5 must be u^2 + 4, we also have dx = 2u du
so the problem becomes
∫ (2u^2) / (u^2 + 4) du evaluated from x = 3 to 8, which means, u = 2 to 3
we have that 2u^2 = 2u^2 + 8 - 8 which is 2(u^2 + 4) - 8 so the integral becomes
∫ (2(u^2 + 4) - 8 )/ (u^2 + 4) from u = 3 to 8
= ∫ 2 - ∫ 8/(u^2 + 4) du, now its a fairly easy trig substitution problem.
for the second, we are not actually solving the integral, just expressing it in terms of another (in this case, slightly reduced) integral. This just screams INTEGRATION BY PARTS!
so lets try. Let u = (1 + x^2)^-n and dV = dx, then du = -2nx/ (1 + x^2)^(n +1), and V = x so the integral becomes
x(1 + x^2)^-n + ∫ 2nx^2/(1 + x^2)^(n+1) now i'm going to use the same trick I did in the problem above, 2nx^2 = 2nx^2 + 2n - 2n, so we have ∫ 2n(x^2 + 1)/(x^2 n + 1)^n+1 - 2n ∫1/(x^2 n + 1)^(n+1) which translates to ∫ 2n/(x^2 +1)^n - 2n ∫1/(x^2 n + 1)^(n+1).
Ugh.. PUTTING IT ALL TOGETHER
I(n) = x(1 + x^2)^-n + ∫ 2n/(x^2 +1)^n - 2n ∫1/(x^2 n + 1)^(n+1)
note the first portion, before the integrals, can be evaluated from 0 to 1 which gives us 2^-n, now if we look closely, we find that the equation becomes
I(n) = 2^-n + 2n I(n) - 2n I(n+1)
rearranging we get 2n I (n+1) = 2^-n + (2n - 1)I(n)
now note we have I(n+1) and I(n) instead of n and n-1, but n is just an arbitrary number so lets just reindex. (in otherwords, replace n+1 with n and n with n-1 then we get
2(n-1)I(n) = 2^-(n-1) + (2n - 2 - 1)I(n-1)
so
2(n-1) I(n) = 2^(1 - n) + (2n - 3)I(n-1) QED.
phew.. I need a cold soda..
Last edited by mikau (2007-08-10 07:22:54)
A logarithm is just a misspelled algorithm.
Offline
you should note that whenever you have an integral of the form:
∫ka/(a + b)^n you can always write this as ∫(ka + kb - kb)/(a + b)^n = ∫ k/(a + b)^(n-1) - ∫kb/(a + b)^n it doesn't hurt it since you are just adding 0 to the numerator and it can get you out of some tight spots. It tends to come up in trig substitution problems where you have something like
∫ 2x^2/(1 + x^2) dx just add + 2 and -2 to the numerator and you can separate it as ∫1dx + ∫2/(1 + x^2)dx = x + 2 arctan(x) + C. wasn't that simple?
A logarithm is just a misspelled algorithm.
Offline
Ah okay, cool. Just ran through that on paper and I get those now, thanks! And learnt a few little tricks there which i'll try to remember.
Hopefully i'll be able to have a better crack at some of the others now. I think the biggest problem I have is not knowing whether to try integration by parts or by substitution, and if so, what substitution to use. Hopefully it'll become easier the more questions I do.
Once again, thanks.
Student: "What's a corollary?"
Lecturer: "What's a corollary? It's like when a theorem has a child. And names it corollary."
Offline
well obviously in some cases you can use either, and in some cases you must use both!! x_x
what I reccomend is, always consider u substitution first, the best way is to look for the derivative of some inner function as a factor, then see if substitution works, if you don't see one, try another method. If you do see the derivative of an inner function as a factor, but substitution gets you nowhere, in otherwords, you end up with an integral of the form f(x)f'(x)g(x) dx then integration by parts is obviously worth a shot.
When the square root of a binomial comes in, usually its a trig substitution.
When its something of the form 4/[(x+5)(x+3)] dx, it usually involves partial fractions.
Just a few tips.
A logarithm is just a misspelled algorithm.
Offline
When its something of the form 4/[(x+5)(x+3)] dx, it usually involves partial fractions.
Well sometimes it can be avoided.
Offline
some imp tips while solving integration?
Offline
Hi akshay1124;
Welcome to the forum!
1)Practice! Practice! Practice!
2)Substitute!
3)IBP!
4)DUIS!
5)EMS!
5)Tables and Mathematica or Wolfram!
In mathematics, you don't understand things. You just get used to them.
If it ain't broke, fix it until it is.
Always satisfy the Prime Directive of getting the right answer above all else.
Offline
Hi akshay1124;
Welcome to the forum!
1)Practice! Practice! Practice!
2)Substitute!
3)IBP!
4)DUIS!
5)EMS!
5)Tables and Mathematica or Wolfram!
What are 4 and 5? Sorry, I haven't heard those terms before... or maybe I am too stupid to work out what they stand for, haha.
Hi;
Five is the Euler Mclaurin summation formula. Tables are giant books of integrals. Mathematica is a Computer Algebra System. Wolfram Alpha is Mathematica online for everyone to use.
In mathematics, you don't understand things. You just get used to them.
If it ain't broke, fix it until it is.
Always satisfy the Prime Directive of getting the right answer above all else.
Offline
But what is duis?
Here lies the reader who will never open this book. He is forever dead.
Taking a new step, uttering a new word, is what people fear most. ― Fyodor Dostoyevsky, Crime and Punishment
The knowledge of some things as a function of age is a delta function.
Offline
DUIS - Differentiation under the integral sign. You can read Richard Feynman's comments about it.
In mathematics, you don't understand things. You just get used to them.
If it ain't broke, fix it until it is.
Always satisfy the Prime Directive of getting the right answer above all else.
Offline
Pages: 1