Differential Equations (MA 1150) : Sukumar
Differential Equations (MA 1150) : Sukumar
Sukumar
Lecture 7
Connection:
If a, b, c ∈ R with a 6= 0, then
ay 00 + by 0 + cy = 0 (1)
Recall that solution to ODE (7) will have linear combination of the following
form
I e m1 x and e m2 x ;
I e mx and xe mx , or
I e λx cos ωx and e λx sin ωx.
Discussion
Recall
I Power series expansion of e x around 0.
I Power series expansion of sin x around 0.
I Power series expansion of cos x around 0.
Consider the following Taylor series:
∞
x
X xn
e = , −∞ < x < ∞,
n=0
n!
∞
X x 2n+1
sin x = (−1)n , −∞ < x < ∞,
n=0
(2n + 1)!
∞
X x 2n
cos x = (−1)n , −∞ < x < ∞,
n=0
(2n)!
Power Series
with positive radius of convergence R has derivatives of all orders in its open
interval of convergence, and successive derivatives can be obtained by
repeatedly differentiating term by term; that is,
∞
X
f 0 (x) = nan (x − x0 )n−1 , (2)
n=1
∞
X
f 00 (x) = n(n − 1)an (x − x0 )n−2 , (3)
n=2
..
.
∞
X
f (k) (x) = n(n − 1) · · · (n − k + 1)an (x − x0 )n−k . (4)
n=k
Hence ,
∞ 2n+1 X ∞
X d x x 2n
f 0 (x) = (−1)n = (−1)n ,
n=0
dx (2n + 1)! n=0
(2n)!
which is the series for cos x.
Discussion
ay 00 + by 0 + cy = 0 (5)
where a, b, c ∈ R with a 6= 0.
Recall The solution to ODE (7) will have linear combination of the following
form
I e m1 x and e m2 x ;
I e mx and xe mx , or
I e λx cos ωx and e λx sin ωx.
Lets slightly change the condition Consider 2nd order homogeneous ODE:
Lets slightly change the condition Consider 2nd order homogeneous ODE:
Note that y1 and y2 may be some power series (depends upon the initial
conditions).
Example
Suppose
∞
X
y (x) = an x n
n=0
Therefore (2 − x)y + 2y = 2y − xy 0 + 2y =
00 00
∞
X ∞
X ∞
X
= 2n(n − 1)an x n−2 − n(n − 1)an x n−1 + 2an x n .
n=2 n=2 n=0
Example
Shift indices in the first two so that all three series will start with n = 0; thus,
we get
∞
X
(2 − x)y 00 + 2y = [2(n + 2)(n + 1)an+2 − (n + 1)nan+1 + 2an ]x n .
n=0
(2 − x)y 00 + 2y = 0
P∞
Conclusion: The power series solution y (x) = n=0 an x n of an ODE gives rise
to recurrence relations among the coefficients an ’s.
Subsection 2
Cauchy-Euler Equations
Second Order Linear Differential Equations
This equation can be transformed into one with constant coefficients by change
of variables(x = e t ) on the interval (0, ∞). Consider the function of t given by
h(t) = y (e t ) = y (x)
Then
dh dy dx dy t
= = e .
dt dx dt dx
Therefore
dy dh −t
= e .
dx dt
Cauchy-Euler Equations
Therefore
dh dy t
= e ,
dt dx
and
d 2h d 2 y 2t dh
2
= e +
dt dx 2 dt
If h1 (t) and h2 (t) are solutions to the constant coefficient ODE, then the
general solution is given by
x 2 y 00 + 7xy 0 + 5y = 0
Putting t = ln x, we get
p(m) = m2 + 6m + 5 = (m + 1)(m + 5)
1
Therefore, h1 (t) = e −t and h2 (t) = e −5t . Hence h1 (ln x) = and
x
1
h2 (ln x) = 5 . Hence the general solution is
x
1 1
y = c1 + c2 5 .
x x
Cauchy-Euler Equations
x 2 y 00 + 7xy 0 + 9y = 0
Putting t = ln x, we get
1
Therefore, h1 (t) = e −3t and h2 (t) = te −3t . Hence h1 (ln x) = 3 and
x
1
h2 (ln x) = ln x 3 . Hence the general solution is
x
1 1
y = c1 + c2 ln x 3 .
x3 x
Cauchy-Euler Equations
x 2 y 00 + 5xy 0 + 13y = 0
Putting t = ln x, we get
y = c1 x m1 + c2 x m2 .
y = c1 x m + c2 x m ln x.
where a, b ∈ R.
Theorem: Suppose p(x), q(x) and f (x) are continuous functions on an open
interval (a, b), let x0 be any point in (a, b). Then the initial value problem
y 00 + p(x)y 0 + q(x)y = 0
y 00 + y = 1, y (0) = 2, y 0 (0) = 7.
Particular solution of non-homogeneous second order linear ODE
p(m) = m2 + 1 = 0.
y = c1 e −x + c2 e −5x + yp .
Find yp :
Solution of 2nd order non-homogeneous linear ODE
Theorem: Suppose p(x), q(x), and f (x) are continuous function on (a, b). Let
yp be a particular solution of
y = yp + c 1 y1 + c 2 y2 ,
y 00 − 2y 0 + y = −3 − x + x 2 . (17)
y 00 − 2y 0 + y = 0
is m2 − 2m + 1 = (m − 1)2 .
y = yp + c1 e x + c2 xe x ,
y 00 − 2y 0 + y = −3 − x + x 2 . (18)
Notice that y is a solution if and only if
y = yp + c1 e x + c2 xe x .
y = 1 + 3x + x 2 + e x (c1 + c2 x).
y = 1 + 3x + x 2 − e x (3 − x).
Recall: solution of 2nd order non-homogeneous linear ODE
y = yp + c 1 y1 + c 2 y2 ,
Observation
I Split the non-homogeneous equation into simpler parts,
I Find a particular solution for each part, and
I Combine their solutions to obtain a particular solution of the original
ODE.
Questions
Question:
I In the first order non-linear ODE y 0 + p(x)y = f (x), we have used
y = u(x)y1 , where y1 is a solution of complementary equation
y 0 + p(x)y = 0.
I We found yp in some different way to get the complete solution in 2nd
order non-homogeneous linear ODE.
I Then for getting solutions, what is the difference between 1st and 2nd
order ODE ? Especially, in the case of non-homogeneous linear ODE.
Subsection 1
y 00 − 7y 0 + 12y = 4e 2x . (21)
y 00 − 7y 0 + 12y = 5e 4x . (22)
that is,
u 00 + u 0 = 5.
Example (continued...)
Now if
u 00 + u 0 = 5,
then we may take u(x) = 5x as a possible candidate.
Therefore
y = 5xe 4x + c1 e 3x + c2 e 4x
is the general solution.
The method of undetermined coefficients
y 00 − 8y 0 + 16y = 2e 4x . (23)
y = e 4x (x 2 + c1 + c2 x).
The method of undetermined coefficients
ay 00 + by 0 + cy = ke αx ,
ay 00 + by 0 + cy = 0, (24)
where λ and ω are real numbers, ω 6= 0, and P(x) and Q(x) are polynomials.
If
yp = A(x) cos ωx + B(x) sin ωx
where A and B are polynomials, then
Theorem: Suppose ω > 0 and P and Q are polynomials. Let k be the larger of
the degrees of P and Q. Then the equation
where
provided that cos ωx and sin ωx are not solutions of the complementary
equation. The solutions of
yp00 − 2yp0 + yp = −4(A cos 2x + B sin 2x) − 4(−A sin 2x + B cos 2x)
+(A cos 2x + B sin 2x)
= (−3A − 4B) cos 2x + (4A − 3B) sin 2x.
yp = cos 2x − 2 sin 2x
is a particular solution
The method of undetermined coefficients - Example
so
yp00 + 4yp = −4A sin 2x + 4B cos 2x.
is a particular solution
The method of undetermined coefficients - Example
and
yp00 = (2B1 − A0 − A1 x) cos x − (2A1 + B0 + B1 x) sin x.
Therefore
yp00 + 3yp0 + 2yp = [A0 + 3A1 + 3B0 + 2B1 + (A1 + 3B1 )x] cos x
(30)
+ [B0 + 3B1 − 3A0 − 2A1 + (B1 − 3A1 )x] sin x.
The method of undetermined coefficients - Example
Comparing the coefficients of x cos x, x sin x, cos x, and sin x with the
corresponding coefficients in (29)
A1 + 3B1 = 20
−3A1 + B1 = 0
A0 + 3B0 + 3A1 + 2B1 = 16
−3A0 + B0 − 2A1 + 3B1 = 10.
Therefore
yp = (1 + 2x) cos x − (1 − 6x) sin x
is a particular solution
The method of undetermined coefficients
where λ and ω are real numbers, ω 6= 0, and P(x) and Q(x) are polynomials.
The new equation after substitution will be constant coefficient equation for u
of the P(x) cos ωx + Q(x) sin ωx.
Need to solve
Notice that cos 3x and sin 3x aren’t solutions of the complementary equation
u 00 − 7u 0 + 12u = 0,
The method of undetermined coefficients
A particular solution of
u 00 − 7u 0 + 12u = 2 cos 3x − (34 − 150x) sin 3x.
is of the form
up = (A0 + A1 x) cos 3x + (B0 + B1 x) sin 3x. (34)
Hence
up00 − 7up0 + 12up = [3A0 − 21B0 − 7A1 + 6B1 + (3A1 − 21B1 )x] cos 3x
+ [21A0 + 3B0 − 6A1 − 7B1 + (21A1 + 3B1 )x] sin 3x.
up is a solution if
3A1 − 21B1 = 0
21A1 + 3B1 = 150
(35)
3A0 − 21B0 − 7A1 + 6B1 = 2
21A0 + 3B0 − 6A1 − 7B1 = −34.
The method of undetermined coefficients
Therefore
up = (1 + 7x) cos 3x − (2 − x) sin 3x
is a particular solution of (33).
Hence
yp = e −2x [(1 + 7x) cos 3x − (2 − x) sin 3x]
is a particular solution of our original ODE.