antiderivative of rational function


The most notable real functions, which can be integrated in a closed formPlanetmathPlanetmath, are the rational functionsMathworldPlanetmath:

Theorem.  The antiderivative of a rational function is always expressible in a closed form, which only can comprise, except a rational expression summand, summands of logarithms and arcustangents of rational functions.

One can justify the theorem by using the general form of the (unique) partial fraction decomposition

R(x)= H(x)+ ∑i=1m(Ai⁢1x-ai+Ai⁢2(x-ai)2+…+Ai⁢μi(x-ai)μi)
+ ∑j=1n(Bj⁢1⁢x+Cj⁢1x2+2⁢pj⁢x+qj+Bj⁢2⁢x+Cj⁢2(x2+2⁢pj⁢x+qj)2+…+Bj⁢νj⁢x+Cj⁢νj(x2+2⁢pj⁢x+qj)νj),

of the rational function R⁢(x) ; here, H⁢(x) is a polynomialMathworldPlanetmathPlanetmathPlanetmath, the first sum expression is determined by the real zeroes ai of the denominator of R⁢(x), the second sum is determined by the real quadratic prime factorsMathworldPlanetmath x2+2⁢pj⁢x+qj of the denominator (which have no real zeroes).

The addends of the form A(x-a)r in the first sum are integrated directly, giving

∫Ax-adx=Aln|x-a|+constant  (r=1) (1)

and

∫A(x-a)rdx=-Ar-1⋅1(x-a)r-1+constant  (r>1). (2)

The remaining partial fractionsPlanetmathPlanetmath are of the form B⁢x+C(x2+2⁢p⁢x+q)s where  p2<q  and s is a positive integer.  Now we may write

x2+2⁢p⁢x+q=(x+p)2+q-p2=(q-p2)⁢[1+(x+pq-p2)2]

and make the substitution

x+pq-p2=t, (3)

i.e.  x=t⁢q-p2-p,  getting

∫B⁢x+C(x2+2⁢p⁢x+q)s⁢𝑑x=∫E⁢t+F(1+t2)s⁢𝑑t=E⁢∫t⁢d⁢t(1+t2)s+F⁢∫d⁢t(1+t2)s (4)

where E and F are certain constants.  In the case  s=1  we have

∫t⁢d⁢t1+t2=12⁢ln⁡(1+t2)+constant (5)

and in the case  s>1

∫t⁢d⁢t(1+t2)s=-12⁢(s-1)⋅1(1+t2)s-1+constant. (6)

The latter addend of the right hand side of (4) is for  s=1  got from

∫d⁢t1+t2=arctan⁡t+constant (7)

and for the cases s>1 on may first write

∫d⁢t(1+t2)s=∫(1+t2)-t2(1+t2)s⁢𝑑t=∫d⁢t(1+t2)s-1-∫t⋅t⁢d⁢t(1+t2)s.

Using integration by parts in the last integral, this equation can be converted into the reduction formula

∫d⁢t(1+t2)s=12⁢s-2⋅t(1+t2)s-1+2⁢n-32⁢n-2⁢∫d⁢t(1+t2)s-1. (8)

The assertion of the theorem follows from (1), …, (8).

Example.

∫d⁢x1+x4=14⁢2⁢ln⁡1+x⁢2+x21-x⁢2+x2+12⁢2⁢arctan⁡x⁢21-x2+C
Title antiderivative of rational function
Canonical name AntiderivativeOfRationalFunction
Date of creation 2013-03-22 19:21:38
Last modified on 2013-03-22 19:21:38
Owner pahio (2872)
Last modified by pahio (2872)
Numerical id 19
Author pahio (2872)
Entry type Theorem
Classification msc 26A36
Synonym integration of rational functions
Related topic IntegrationTechniques