Scribd
Upload
333 views4 pages

Lecture 9 Non-Homogeneous Linear ODE, Method of Undetermined Coefficients

The document discusses the method of undetermined coefficients for finding particular solutions to non-homogeneous linear ordinary differential equations. It presents two main rules for determining the trial function used in the method of undetermined coefficients. Several examples are provided to illustrate how to apply the rules to find the particular solution based on whether terms in the non-homogeneous part are or are not solutions to the homogeneous equation. The method of annihilators is also introduced as an alternative approach.

Uploaded by

RajSarkar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
333 views4 pages

Lecture 9 Non-Homogeneous Linear ODE, Method of Undetermined Coefficients

The document discusses the method of undetermined coefficients for finding particular solutions to non-homogeneous linear ordinary differential equations. It presents two main rules for determining the trial function used in the method of undetermined coefficients. Several examples are provided to illustrate how to apply the rules to find the particular solution based on whether terms in the non-homogeneous part are or are not solutions to the homogeneous equation. The method of annihilators is also introduced as an alternative approach.

Uploaded by

RajSarkar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

S.

Ghorai

Lecture IX
Non-homogeneous linear ODE, method of undetermined coefficients

Non-homogeneous linear equation

We shall mainly consider 2nd order equations. Extension to the n-th order is straight
forward.
Consider a 2nd order linear ODE of the form
y 00 + p(x)y 0 + q(x)y = r(x),

x I,

(1)

where p, q are continuous functions. Let yp (x) is a (particular) solution to (1). Then
yp00 + p(x)yp0 + q(x)yp = r(x).
Let y be any solution to (1). Now consider Y = y yp satisfies
Y 00 + p(x)Y 0 + q(x)Y = 0.
Thus, Y satisfies a homogeneous linear 2nd order ODE. Hence, we express Y as linear
combination of two LI solutions y1 and y2 . This gives
Y = C1 y1 + C2 y2
or
y = C1 y1 + C2 y2 + yp

(2)

Thus, the general solution to (1) is given by (2). We have seen how to find y1 and y2 .
Here, we concentrate on methods to find yp .

1.1

Method of undetermined coefficients

This method works for the following nonhomogeneous linear equation:


ay 00 + by 0 + cy = r(x),

x I,

(3)

where a, b, c are constants and r(x) is a finite linear combination of products formed
from the polynomial, exponential and sines or cosines functions. Thus, r(x) is a finite
linear combination of functions of the following form:
(
x m

e x

sin x
,
cos x

where m is a nonnegative integer.


Suppose
r(x) = r1 (x) + r2 (x) + + rn (x).
If ypi (x), 1 i n is a particular solution to
ay 00 + by 0 + cy = ri (x),

S. Ghorai

then it is trivial to prove that


yp (x) = yp1 (x) + yp2 (x) + + ypn (x)
is a particular solution to
ay 00 + by 0 + cy = r(x).
Hence, we shall consider the case when r(x) is one of the ri (x). Thus, we choose r(x)
to be of the following form:
(
sin x
x m
e x
.
cos x
Rule 1: If none of the terms in r(x) is a solution of the homogeneous problem, then
for yp , choose a linear combination of r(x) and all its derivatives that form a finite set
of linearly independent functions.
Example 1. Consider
y 00 2y 0 + 2y = x sin x.
Solution: The LI solutions of the homogeneous part are ex cos x and ex sin x. Clearly,
neither x nor sin x is a solution of the homogeneous part. Hence, we choose
yp (x) = ax sin(x) + bx cos x + c cos x + d sin x.
Now substituting into the governing equation, we get
(a+2b)x sin x+(2a+b)x cos x+(2a2b2c+d) cos x+(2a2b+c+2d) sin x = x sin x.
Hence
a + 2b = 1,

2a + b = 0,

(2a 2b 2c + d) = 0,

(2a 2b + c + 2d) = 0.

Solving, we get
1
2
2
14
a= , b= , c= , d=
5
5
25
25
Hence, the general solution is
1
2
2
24
y = ex (C1 cos x + C2 sin x) + x sin(x) + x cos x +
cos x +
sin x
5
5
25
25
Aliter: (Annihilator method ) Writing D d/dx, we write
(D2 2D + 2)yp = x sin x.
Note that (D2 + 1)2 x sin x = 0. Hence, operating (D2 + 1)2 on both sides, we find
(D2 + 1)2 (D2 2D + 2)yp = 0.
The characteristic roots are found from (m2 + 1)2 (m2 2m + 2) = 0. Thus, m =
1 i and m = i, i. Now solution to this homogeneous linear ODE with constant
coefficient is
yp = ex (c1 cos x + c2 sin x) + (c3 cos x + c4 sin x) + x(c5 cos x + c6 sin x)

S. Ghorai

Since, the first two terms are the solution of the original homogeneous part and hence
contribute nothing. Thus, the form for yp must be
yp = (c3 cos x + c4 sin x) + x(c5 cos x + c6 sin x),
which conforms with previous form.
Rule 2: If r(x) contains terms that are solution of the homogeneous linear part, then to
choose the trial form of yp follow the following steps. First, choose a linear combination
of r(x) and its derivatives which are LI. Second, this linear combination is multiplied
by a power of x, say xk , where k is the smallest nonnegative integer that makes all the
new terms not to be solutions of the homogeneous problem.
Example 2. Consider
y 00 2y 0 3y = xex .
Solution: The LI solutions of the homogeneous part are ex and e3x . Clearly, ex
is a solution of the homogeneous part. Hence, we choose yp (x) = x(axex + bex ).
Substituting, we find
ex (4b + 2a 8ax) = xex
This, gives 4b + 2a = 0, 8a = 1 and thus a = 1/8, b = 1/16. Thus, the general
solution is
xex
(2x + 1)
y = C1 ex + C2 e3x
16
Aliter: (Annihilator method ) Writing D d/dx, we write
(D2 2D 3)yp = xex .
Since (D + 1)2 xex = 0, operating (D + 1)2 on both sides we find
(D + 1)2 (D2 2D 3)yp = 0
The characteristic roots are found from (m + 1)2 (m2 2m 3) = 0. Thus, m =
1, 1, 1, 3. Now solution to this homogeneous linear ODE with constant coefficient
is
yp = c1 e3x + ex (c2 + c3 x + c4 x2 )
Since, the first two terms are the solution of the original homogeneous part and hence
contribute nothing. Thus, the form for yp must be
yp = ex (c3 x + c4 x2 ),
which conforms with the previous form.
Example 3. Consider
y 00 2y 0 + y = 6xex
Solution: The LI solutions of the homogeneous part are ex and xex . Clearly, both
ex , xex are solutions of the homogeneous part. Hence, we choose yp (x) = x2 (axex +bex ).
Substituting, we find
ex (2b + 3ax) = 6xex
This, gives a = 1, b = 0. Thus, the general solution is
y = ex (C1 + C2 x + x3 )

S. Ghorai

Example 4. Consider
y 000 3y 00 + 2y 0 = 10 + 4xe2x .
Solution: The LI solutions of the homogeneous part related to the characteristic
equation
m3 3m2 + 2m = 0 m = 0, 1, 2.
Thus the LI solutions are 1, ex and e2x . Clearly, e2x and 10 are solutions of the homogeneous part. Hence, we choose yp (x) = ax + x(bxe2x + ce2x ). Substituting, we
find
2a + (6b + 2c)e2x + 4bxe2x = 10 + 4xe2x
This, gives a = 5, b = 1 and c = 3 Thus, the general solution is
y = C1 + C2 ex + C3 e2x + 5x + e2x (x2 3x)
Aliter: (Annihilator method ) Writing D d/dx, we write
(D3 3D + 2D)yp = 10 + 4xe2x .
To annihilate 10 we apply D and to annihilate xe2x , we apply (D 2)2 . Thus,
D(D 2)2 (D3 3D + 2D)yp = 0
The characteristic roots are found from m(m 2)2 (m3 3m + 2m) = 0. Thus, m =
0, 0, 1, 2, 2, 2. Now solution to this homogeneous linear ODE with constant coefficient
is
yp = c1 + c2 x + c3 ex + e2x (c4 + c5 x + c6 x2 )
The terms with c1 , c3 and c4 are the solution of the original homogeneous part and
hence contribute nothing. Thus, the form for yp must be
yp = c2 x + e2x (c5 x + c6 x2 ).
which conforms with the previous form.

You might also like