If a and b are positive coprime integers in the following integral:
∫ 0 ∞ x 4 + 5 x 2 + 4 2 x 2 − 1 d x = b a π ,
find the value of a + b .
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
Shouldn't it be arctan(x/2) in the final solution, not arctan(2x)
You are elaborating fundamental path. However, many people would refer to standard tables to look for direct answer up to certain extend.
Log in to reply
He is proving that all the integrals are true. That's a good start to be good at solving integrals.
Log in to reply
70 × 7 is quite a great margin.
Log in to reply
@Lu Chee Ket – I have no idea what that number means in this context.
Log in to reply
@Pi Han Goh – Bible. Look for those who know about this.
EDIT: This answer uses the tools of complex analysis to get a result, Sudeep Salgia has a much simpler solution as a comment, please upvote his solution instead!
Let I = ∫ 0 ∞ x 4 + 5 x 2 + 4 2 x 2 − 1 d x = 2 1 ∫ − ∞ ∞ ( x 2 + 1 ) ( x 2 + 4 ) 2 x 2 − 1 d x The poles of the integrand are at i and 2 i , we take the upper semicircle along the real axis, centred at 0 of radius R as our contour. Hence we have that ∮ C ( z 2 + 1 ) ( z 2 + 4 ) 2 z 2 − 1 d z = ∫ C 1 ( z 2 + 1 ) ( z 2 + 4 ) 2 z 2 − 1 d z + 2 I as R → ∞ ∫ − R R ( x 2 + 1 ) ( x 2 + 4 ) 2 x 2 − 1 d x where C 1 is the circular part of the semicircle contour. We evaluate the integral covering C 1 : Note that C 1 is given by z = R e i t for t ∈ [ 0 , π ] , then d z = i R e i t d t so: ∫ C 1 ( z 2 + 1 ) ( z 2 + 4 ) 2 z 2 − 1 d z = ∫ 0 π ( ( R e i t ) 2 + 1 ) ( ( R e i t ) 2 + 4 ) i R ( 2 R e i t 2 − 1 ) d t = ∫ 0 π R 4 ( ( e i t ) 2 + 1 / R 2 ) ( ( e i t ) 2 + 4 / R 2 ) i R 2 ( 2 e i t 2 − 1 / R ) d t which clearly goes to 0 as R → ∞ . Now by the Residue Theorem: ∮ C ( z 2 + 1 ) ( z 2 + 4 ) 2 z 2 − 1 d z = 2 π i ( b i + b 2 i ) Where b 1 and b 2 are the residues at i and 2 i respectively. These are simple poles of f ( z ) = ( z 2 + 1 ) ( z 2 + 4 ) 2 z 2 − 1 which can be written as f ( z ) = z − i ϕ 1 ( z ) = z − 2 i ϕ 2 ( z ) so b i = ϕ 1 ( i ) = i / 2 and b 2 i = ϕ 2 ( 2 i ) = − 3 i / 4 hence: ∮ C ( z 2 + 1 ) ( z 2 + 4 ) 2 z 2 − 1 d z = 2 π i ( b i + b 2 i ) = π / 2 and so as R → ∞ , 2 I = π / 2 ⟹ I = π / 4
It is much easier using partial fractions. After the first step, we can write, ∫ 0 ∞ ( x 2 + 1 ) ( x 2 + 4 ) 3 ( x 2 + 1 ) − ( x 2 + 4 ) d x = 3 ∫ 0 ∞ x 2 + 4 1 d x − ∫ 0 ∞ x 2 + 1 1 d x = 4 3 π − 2 π = 4 π
Log in to reply
Much better solution!
Just a caution that t a n − 1 2 x of x → ∞ equals to 2 π
Log in to reply
That's true. But the integral ∫ x 2 + 4 1 d x = 2 1 arctan ( 2 x ) . So that factor of 4 1 . I have skipped some trivial substitution steps as that was just an alternate method.
Log in to reply
@Sudeep Salgia – Correct! Some more with a 3 for ( 2 3 − 1 ) 2 π = 2 1 2 π as a whole.
Your solution is identical to mine or perhaps vice versa. But solution by Ali is has frightened me
Log in to reply
Nothing frightening, just using results of complex analysis :)
Can't we write 2-1\x^(2)= m(1-2\x^(2))+n(1+2\x^(2)) derivatives of x+2\x and x-2/x I did by this way ... Edit-this trick has been done after multiplying and dividing the integrand by x^(2)
I am unsure about what you are trying to say, could you clarify?
Log in to reply
Yes first multiply and divide whole integrand by x^(2) then we can see in the denominator part a square of either x+(2\x) or x-(2\x) can be made and in numerator part we can see that an algebric expression of the differentiation of these two discussed expressions are written from which we can get the integrand of the form f '(x)/f(x)...try iy
Problem Loading...
Note Loading...
Set Loading...
as suggested by @Sudeep Salgia ( x 4 + 5 x 2 + 4 2 x 2 − 1 = x 4 + 4 x 2 + x 2 + 3 3 x 2 + 3 − x 2 − 4 = ( x 2 + 4 ) ( x 2 + 1 ) 3 ( x 2 + 1 ) − ( x 2 + 4 ) = x 2 + 4 3 − x 2 + 1 1 hence ∫ 0 ∞ ( x 2 + 1 ) ( x 2 + 4 ) 3 ( x 2 + 1 ) − ( x 2 + 4 ) d x = 3 ∫ 0 ∞ x 2 + 4 1 d x − ∫ 0 ∞ x 2 + 1 1 d x lets solve each integral. ∫ x 2 + 4 1 d x let ⎩ ⎪ ⎨ ⎪ ⎧ u = a r c t a n ( 2 x ) x = 2 t a n ( u ) d x = 2 s e c 2 ( x ) d u so: ∫ x 2 + 4 1 d x = ∫ 4 t a n 2 ( x ) + 4 1 2 ( s e c 2 ( x ) ) d u by a fundemental trig identity: = ∫ 4 s e c 2 ( x ) 1 2 ( s e c 2 ( x ) ) d u = ∫ 2 1 d u = 2 1 u + C = 2 a r c t a n ( 2 x ) + C great, second part: ∫ x 2 + 1 1 d x so: ⎩ ⎪ ⎨ ⎪ ⎧ u = a r c t a n ( x ) x = t a n ( u ) d x = s e c 2 ( u ) d u put this in ∫ x 2 + 1 1 d x = ∫ t a n 2 ( u ) + 1 1 s e c 2 ( u ) d u = ∫ s e c 2 ( u ) 1 s e c 2 ( u ) d u = ∫ d u = u + C = a r c t a n ( x ) + C so we are looking for l i m n → ∞ [ ( 3 ( 2 a r c t a n ( 2 x ) ) − a r c t a n ( x ) ] 0 n = l i m n → ∞ ( 3 ( 2 a r c t a n ( 2 n ) ) − a r c t a n ( n ) − 3 ∗ 0 + 0 ) = 3 2 2 π − 2 π = 4 π