Outline

I Nonhomogeneous second order linear ODEs

I Method of variation of parameters

I Method of undetermined coefficients

Nonhomogeneous second order linear ODEs

Consider the nonhomogeneous DE:

y 00 + p(x)y 0 + q(x)y = r (x),

where p, q, r are continuous on an interval I . The associated

homogeneous DE is:

y 00 + p(x)y 0 + q(x)y = 0.

Can we relate the solutions of the above two DEs?

Nonhomogeneous second order linear ODEs
Theorem
Let yp (x) be any solution of y 00 + p(x)y 0 + q(x)y = r (x), and
y1 (x), y2 (x) be a basis of the solution space of the corresponding
homogeneous DE.
Then, the set of solutions of the nonhomogeneous DE is

{c1 y1 (x) + c2 y2 (x) + yp (x) : c1 , c2 ∈ R}.

Proof: Let L(y ) := y 00 + p(x)y 0 + q(x)y , and φ(x) be any solution

of the given nonhomogeneous eq. Then,

L(φ(x) − yp (x)) = L(φ(x)) − L(yp (x)) = r (x) − r (x) = 0.

Nonhomogeneous second order linear ODEs

Thus, φ(x) − yp (x) is a solution of the homogeneous DE, which

means that

φ(x) − yp (x) = c1 y1 (x) + c2 y2 (x), c1 , c2 ∈ R.

We have:
φ(x) = c1 y1 (x) + c2 y2 (x) + yp (x).

Summary – In order to find the general solution of a

nonhomogeneous DE, we need to find:

1. a particular solution of the nonhomogeneous DE and

2. the general solution of the corresponding homogeneous DE.

Method of variation of parameters

I A method to find a particular solution of a nonhomogeneous

ODE.

I Let
yp (x) = v1 (x)y1 (x) + v2 (x)y2 (x)
.
I Then, choose these functions so that is a solution of
L(y ) = r (x).

I Now, yp0 = v1 y10 + v2 y20 + v10 y1 + v20 y2 .

I Set v10 y1 + v20 y2 = 0, we see that:

Method of variation of parameters

I Now we have yp0 = v1 y10 + v2 y20 , which will give

yp00 = v1 y100 + v10 y10 + v2 y200 + v20 y20 .

I Substituting these in the nonhomogeneous ODE, rearranging

terms
yp00 + pyp0 + qyp = r (x)
v1 (y100 + py10 + qy1 ) + v2 (y200 + py20 + qy2 ) + v10 y10 + v20 y20 = r (x).
I Now if we use the fact that y1 and y2 are solutions of
associated Homogeneous ODE, i.e

yi00 + pyi0 + qyi = 0 for i = 1, 2.

Method of variation of parameters

I We get v10 y10 + v20 y20 = r (x).

I Recall that we also required: v10 y1 + v20 y2 = 0.

I Together with above two we get

" #" # " # " # " #−1 " #
y1 y2 v10 0 v10 y1 y2 0
= =⇒ = .
y10 y20 v20 r (x) v20 y10 y20 r (x)
" # " #" #
v10 1 y20 −y2 0
= .
v20 W (y1 , y2 ) −y10 y1 r (x)
Method of variation of parameters
I Therefore
−y2 r (x) y1 r (x)
v10 = and v20 = .
W (y1 , y2 ) W (y1 , y2 )
I That is,
Z Z
y2 r (x) y1 r (x)
v1 = − dx and v2 = dx.
W (y1 , y2 ) W (y1 , y2 )
I Hence,
Z Z
y1 r (x) y2 r (x)
yp = v1 y1 + v2 y2 = y2 dx − y1 dx.
W (y1 , y2 ) W (y1 , y2 )
I Finally, general solution of the non homogeneous eq. is:

y = c1 y1 + c2 y2 + yp .
Example 1
Find a particular solution of y 00 + y = cosec x.
I General solution of associated homogeneous eq i.e y 00 + y = 0
is y = c1 sin x + c2 cos x.
I Wronskian W (y1 , y2 ) equals:

sin x cos x
= −1.
cos x − sin x

I Also,
Z Z
y2 r (x) cos x cosec x
v1 = − dx = − dx = log | sin x|,
W (y1 , y2 ) −1
Z Z
y1 r (x) sin x cosec x
v2 = dx = dx = −x.
W (y1 , y2 ) −1
I A particular solution is given by:
yp (x) = sin x log | sin x| − x cos x.
Example 2
Find the general solution of y 00 − y 0 − 2y = e −x .
General solution of associated homogeneous eq. is
y = c1 e 2x + c2 e −x .
Wronskian W (y1 , y2 ) equals:

e 2x e −x
= −3e x .
2e 2x −e −x

Also,

e −x e −x
Z Z
y2 r (x) 1
v1 = − dx = − x
dx = − e −3x ,
W (y1 , y2 ) −3e 9

e 2x e −x
Z Z
y1 r (x) 1
v2 = dx = dx = − x.
W (y1 , y2 ) −3e x 3
Example 2

A particular solution is given by:

yp (x) = − 19 e −3x e 2x − 31 xe −x = − 19 e −x − 13 xe −x .

General solution of the nonhomogeneous DE is:

1
y = c1 e 2x + c2 e −x + yp (x) = c1 e 2x + C2 e −x − xe −x .
3
Example 3

Find a particular solution of y 00 + 4y = 3 cos 2t.

I Basis of solutions of corresponding homogeneous eq. is

y1 = cos 2t, y2 = sin 2t.

I Then,
Z
sin 2t . 3 cos 2t 3
v1 = − dt = cos 4t,
2 16
Z
cos 2t . 3 cos 2t 3 3
v2 = dt = sin 4t + t.
2 16 4

I A particular solution is given by: yp = v1 y1 + v2 y2 .