MAT200
Franklin College
Erich Prisner
Every rational function can be integrated. Let's start with an example. Look at the rational function
![]()
The first step would be polynomial division to write the function as a sum of a polynomial and a rational function where the degree of the numerator is less than the degree of the denominator. Since x5+x4-x2+x divided by x4-x3-x+1 equals x+2 with a remainder of 2x3+2x-2, we get
![]()
Next we factor the denominator poynomial as far as possible. We can always achieve that all factors are polynomials of degree 1 or 2. Why? Remember that every odd degree polynomial has a real zero, and therefore a linear factor. What about even degree polynomials without real zeros (linear factors?)
In our example, using the Rational Zero Theorem and again polynomial division, we can factor x4-x3-x+1 = (x2+x+1)(x-1)2. Hence

![]()
![]()
We multiply the approach equation by (x2+x+1)(x-1)2
to obtain
(Ax+B)(x-1)2+C(x2+x+1)(x-1)+D(x2+x+1)
= 2x3-x2+2x-1, or
(A+C)x3+(B-2A+D)x2+(A-2B+D)x+(B-C+D)=
2x3-x2+2x-1.
A, B, C, D are obtained by comparing coefficients---remember that
two polynomials are equal if and only if corresponding coefficients
of like terms are equal. So we solve the system of equations
A+C=2
B-2A+D=-1
A-2B+D=2
B-C+D=-1
The solution is A=2/3, B=-1/3, C=4/3, D=2/3, and we get
![]()
![]() | Since (2x+p)/(x^2+px+q) can be integrated easily, this integral can be reduced to the integral of the form ... |
![]() | |
![]() | |
![]() | |
| polynomials | obvious |
Erich Prisner, October 2003