Suppose we have a differential equation M(x,y) + N(x,y) y' = 0
that is not exact. It may be that we can convert it into an
exact equation by multiplying it by some function
.
Such a function is called an integrating factor. In fact, this
is just what we did for the linear equation.
What is needed for
to be an integrating factor?
Apply the test for exact equations to
:
![]()
Unfortunately, in general it's just as hard to solve this for
as it is to solve the original differential equation.
But in some special cases it is easier. For example, suppose
depends only on x. The exactness test becomes
![]()
![]()
The left side is supposed to depend only on x, so the right side
must also depend only on x. That is the requirement for the
equation to have an integrating factor that depends only on x.
If this requirement is fulfilled, we have a first-order homogeneous
linear
equation to solve for
.
Example: Solve x2+ y3 + (x2 - 3 x y2) y' = 0.
We have M = x2+ y3, N = x2 - 3 x y2. Since
![]()

![]()
Solving this, we get
![]()
So our exact equation is

Now if the solution is F(x,y) = C, we want
![]()
![]()

K'(y) = 1
K(y) = y
Thus the solution is
![]()