α = π 1 ∫ 0 ∞ x 4 + x 2 + 1 x 3 + x + 1 3 x d x Find the value of 2 7 α 2 ( 7 − 3 α 2 ) 2 .
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.
Rem.: If someone has an idea how to transform the odd integral into a contour integral, please let me know! Also if there is another way (like integration under the integral sign), though I did not see a way there either. Perhaps it just is that ghastly :(
Rem.: If someone knows of a systematic algebraic way to verify P ( α ) = 1 , please update the solution!
Rem.: If this is the intended way to solve the integral, I'd say it deserves an upgrade to level 5 for the amount of work necessary.
Problem Loading...
Note Loading...
Set Loading...
This integral is an absolute monster. There might be a faster way of doing it, but even with complex analysis it seems really hard. For this solution, I'll only use real analysis, paired with some complex algebra to shorten the partial fraction decompositions (PFD).
Finding symmetry and reducing the denominator
Notice the rational part has an odd and an even part, while the cube root is odd. We expand by x 2 − 1 ( ∗ ) to shorten the denominator: f ( x ) : = π 1 x 4 + x 2 + 1 x 3 + x + 1 ⋅ x − 3 1 ( ∗ ) = π 1 x 6 − 1 x 5 + x 2 − x − 1 ⋅ x − 3 1 = π 1 x 6 − 1 ( x 4 − 1 ) x 3 2 + π 1 x 6 − 1 ( x 2 − 1 ) x − 3 1 = : e ( x ) + o ( x ) We factor the denominators to x 6 − 1 = ( x 3 − 1 ) ( x 3 + 1 ) . Comparing coefficients, we solve the PFDs: e ( x ) o ( x ) = 2 π 1 [ x 3 − 1 x − 1 + x 3 + 1 x + 1 ] x 3 2 = : 2 π 1 ( g 1 ( x ) + g 1 ( − x ) ) x 3 2 = 2 π 1 [ x 3 − 1 x 2 − 1 + x 3 + 1 1 − x 2 ] x − 3 1 = : 2 π 1 ( g 2 ( x ) + g 2 ( − x ) ) x − 3 1 With a substitution ( ∗ ∗ ) u : = − x , u → x for the " − x "-parts above we express the entire integral in terms of the reduced functions g 1 , 2 ( x ) : α = ∫ 0 ∞ e ( x ) + o ( x ) d x ( ∗ ∗ ) = 2 π 1 ∫ R g 1 ( x ) x 3 2 d x + 2 π 1 ∫ 0 ∞ g 2 ( x ) x − 3 1 d x + 2 π 1 ∫ 0 − ∞ g 2 ( x ) x − 3 1 d x = : α e + α o The even integral simplifies nicely (and could easily be evaluated by contour integration), while the odd integral does not. I couldn't come up with a nice complex curve to solve the odd integral, so let's just use real analysis, like before :)
Solving the even part α e
Substitute u : = x 3 1 , u ( 1 ) ( x ) = 3 u 2 1 to eliminate the cube root. The integration bounds stay the same: α e = 2 π 1 ∫ R u 9 − 1 u 3 − 1 ⋅ u 2 ⋅ 3 u 2 d u = 2 π 3 ∫ R u 9 − 1 u 3 − 1 ⋅ u 4 d u , ∣ u k : = w k = : c k + i s k , k ∈ { 1 , 2 , 4 ; 5 ; 7 ; 8 } , u 9 − k = u k ∗ , w : = e 9 2 π i We define u k as the complex conjugate pole pairs and notice they all are powers of the fundamental ninth roots of unity w . Their real- and imaginary parts are all sines and cosines we denote by s k , c k . Even though the degree of the denominator suggests otherwise, we only have six of them: The remaining three cancel against the numerator (we introduced them ourselves when we expanded by x 2 − 1 to shorten the denominator). As all poles are distinct, we have a simple PFD: α e PFD = 2 π 3 ∫ R k ∈ { 1 ; 2 ; 4 } ∑ u − u k A k + u − u k ∗ A k ∗ d u ∣ ∣ ∣ ∣ A k = u → u k lim ( u − u k ) u 9 − 1 u 3 − 1 u 4 l’H. = 9 u k 8 u k 3 − 1 u k 4 = 9 1 ( u k − 1 − u k − 4 ) = : a k + i b k , a k , b k ∈ R The antiderivative we still need to find is listed at ( ∗ ∗ ∗ ) at the end. The log-parts cancel out when we let x → ± ∞ because a 1 + a 2 + a 3 = 0 : α e = 2 π 3 ⎣ ⎡ k ∈ { 1 ; 2 ; 4 } ∑ a k lo g ∣ ( u − c k ) 2 + s k 2 ∣ − 2 b k arctan ( s k u − c k ) ⎦ ⎤ − ∞ ∞ = 2 π 3 ⋅ ( − 2 π ) ( b 1 + b 2 + b 4 ) = − 3 ⋅ 9 1 ( − s 1 + s 4 − s 2 + s 8 − s 4 + s 1 6 ) = 3 2 ( s 1 + s 2 ) s 1 , 2 , 4 > 0
Solving the odd part α o
Using the exact same steps as in the even part, we can calculate the first part of the odd integral. The only change is the new definition of A k : 2 π 1 ∫ 0 ∞ g 2 ( x ) x − 3 1 d x = 2 π 3 ∫ 0 ∞ u 9 − 1 u 6 − 1 ⋅ u d u = 2 π 3 ∫ 0 ∞ k ∈ { 1 ; 2 ; 4 } ∑ u − u k A k + u − u k ∗ A k ∗ d u = 2 π 3 ⎣ ⎡ k ∈ { 1 ; 2 ; 4 } ∑ a k lo g ∣ ( u − c k ) 2 + s k 2 ∣ − 2 b k arctan ( s k u − c k ) ⎦ ⎤ 0 ∞ = 2 π 3 k ∈ { 1 ; 2 ; 4 } ∑ − 2 b k [ 2 π − ( 9 2 π k − 2 π ) ] ∣ ∣ ∣ ∣ A k = 9 1 ( u k − 1 − u k − 7 ) = : a k + i b k , a k , b k ∈ R Like before, the log-parts cancel at infinity because a 1 + a 2 + a 3 = 0 . At zero, they cancel because s k 2 + c k 2 = 1 . The second part of the odd integral works just like the first one - only the upper border changes to − ∞ : 2 π 1 ∫ 0 − ∞ g 2 ( x ) x − 3 1 d x = 2 π 3 k ∈ { 1 ; 2 ; 4 } ∑ − 2 b k [ − 2 π − ( 9 2 π k − 2 π ) ] Adding both parts together: α o = 2 π 3 k ∈ { 1 ; 2 ; 4 } ∑ − 2 b k [ 0 − 2 ( 9 2 π k − 2 π ) ] = π 3 ⋅ 9 1 ( 9 − 5 π ( − s 1 + s 7 ) + 9 − π ( − s 2 + s 1 4 ) + 9 7 π ( − s 4 + s 2 8 ) ) = 2 7 1 ( 1 2 s 1 + 6 s 2 − 6 s 4 ) = 9 2 ( 2 s 1 + s 2 − s 4 )
Solving the task
With solutions to both the odd and even part, we can finally calculate a solution to the integral. Remember the short-hand s k = sin ( 9 2 π k ) : α = α e + α 0 = 9 2 ( 5 s 1 + 4 s 2 − s 4 ) ⇒ P ( α ) = 2 7 α 2 ( 7 − 3 α 2 ) 2 ≈ 1 ± 1 0 − 8 The last result I only checked numerically. I suspect you just have to use addition theorems repeatedly, but at this point, I'm a bit too tired to do that:)
Standard anti-derivative of a complex pole pair
Let u : = c + i s ∈ C with ∣ u ∣ = 1 , s = 0 and let A = a + i b ∈ C . Then ∫ x − u A + x − u ∗ A ∗ d x = a lo g ∣ ( x − c ) 2 + s 2 ∣ − 2 b arctan ( s x − c ) + C , C ∈ R ( ∗ ∗ ∗ )