Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M.

Eldesoky

CLAPTER (4)

Linear Differential Equations With Constant Coefficients

2
Let D stand for d , D 2 for d 2 ; and soon. The symbol D, D 2 etc are
dx dx
called operators. The index of D indicates the number of times the
operation of differentiation must be carried out. For example,

D 3 x 4 shows that we must differentiate x 4 three times. Thus,

D 3 x 4 = 24x. The following results are valid for such operators.

1. Dm + Dn = Dn + Dm
2. Dm Dn = Dn Dm = Dn +m
3. D(u + v ) = Du + Dv where u and v are functions of x

4. ( D −  )( D −  ) = ( D −  )( D −  ) , where  ,  are constant

Negative index of D. D −1 is equivalent to en integration. For example

2
, D x =  xdx =
−1 x . But it is important to note that the main object of
2

D −1 is to find an integral but not to complete the integral. Consequently

the arbitrary constant which arises in integration must be omitted.
It is usual to write 1/D m for D −m . It is to be remembered that

DD −1 =1 and the symbol D with negative indices also satisfy the above-
mentioned four results.

49
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

Furthermore we write
d 2y dy
2
+ a1 + a2y = ( D 2 + a1D + a2 ) y = f (D )y
dx dx

where f (D ) is the operator now. If f1(D ) and f 2 (D ) be two operators,

then f1(D ) and f 2 (D )is also an operator such that

f1(D ) f 2 (D ) = f 2 (D ) f1(D )

Linear Differential Equations With Constant Coefficients has the form

dny d n −1y d n −2y

n
+ a1 n −1 + a2 n −2
+ ..... + an y = G (x ) ---------(1)
dx dx dx

whereG (x ) is a function of x only and a1,a2 ,...an are constant, is

called a linear differential equation of nth order.

2 n
Using the symbols D, D ,.....D the equation becomes

D n y + a1D n −1y + a 2D n −2y + ..... + an y = G (x )

(D n
+ a1D n −1 + a 2D n −2 + ..... + an ) y = G (x )
f (D )y = G (x )
where f (D ) = ( D n + a1D n −1 + a 2D n −2 + ..... + an )

and now f (D ) acts as operator and operates on y to yield G (x ).

The general solution of (1) is y=C.F.+P.I. where C.F. is complementary

functions and it involves n arbitrary constants and P.I. is particular
integral which does not involve any arbitrary constant.

50
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

It should be remembered that P.I. appears due to G(x)in (1) Hence if a

linear differential equation with constant coefficients is given with
G(x)=0 then its general solution will not involve P.I. and so for such
differential equations the general solution will be given by y=C.F.

Auxiliary equation.

Consider the differential equation (1) with G(x) =0 that is,

(D n
+ a1D n −1 + a2D n −2 + ..... + an ) y = 0  f (D )y = 0 ---------(2)

assume that y = e mx is a solution of this equation, then since

y = e mx , Dy = me mx , D 2e mx = m 2e mx ,......, D n e mx = m n e mx
so (2) becomes (m n + a1m n −1 + a 2m n −2 + ..... + an ) e mx = 0

which will hold if

mn + a1mn −1 + a2mn −2 + ..... + an = 0 or f (m ) = 0 --------------(3)

this is called the auxiliary equation.

On comparing (2) and (3) we see that the auxiliary equation f (m ) = 0

will give the same values of m as the equation f (D ) = 0 gives of D.

Hence in future we shall write the auxiliary equation as f (D ) = 0 and

solve it for D.

51
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

Working rule for finding complementary function

We first rewrite the equation in a symbolic form f (D ) = G (x ). Then

we write its auxiliary equation, namely, f (D ) = 0 which will be an

algebraic equation of nth order. On solving the auxiliary equation we

shall get n roots. Three cases arise, according as the roots of the
auxiliary equation are

(i) simple real roots

(ii) complex roots   i
(iii) surds in the form  

Now, we will discuss these one by one

Case(1)
First suppose that the auxiliary equation has n distinct roots
m1,m2 ,....,mn then C.F. is given by

yc = c1e m1x + c 2e m2x + ....... + cne mn x

where c1,c 2 ,...cn are arbitrary constants.

Again, if the auxiliary equation has the real root mk occurring k times,

and if further the remaining roots of the auxiliary equation are distinct
real numbers mk +1,mk +2 ,.....,mk then C.F. is given by

52
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

(c 1 + c 2x + c 3 x 2 + .... + c k x k −1 ) e mk x + c k +1e mk +1x + .... + c ne mn x

Example 1
Solve

(D 2
− 3D + 2) y = 0

Solution

Here the auxiliary equation is (D − 3D + 2) = 0

2

Which gives D=1,2(distinct roots) hence

yc = c1e m1x + c 2e m2x

yc = c1e x + c 2e 2x
since here G(x)=0, so the general solution is

y = c1e x + c 2e 2x
Example 2
Solve

(D 3
− 3D + 2) y = 0

solution
the auxiliary equation is D 3 − 3D + 2 = 0
D 3 − D − 2D + 2 = 0
D (D 2 − 1) − 2(D − 1) = 0
or
D (D − 1)(D + 1) − 2(D − 1) = 0
(D − 1)(D 2 + D − 2) = 0
(D − 1)(D − 1)(D + 2) = 0
which gives D=1,1,-2, since 1 occurs two times, The C.F. and so the
general solution is

53
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

y = yc = (c1 + c 2x )e x + c 3e −2x
where c1,c 2 and c 3 are arbitrary constants.

Case (ii)
let   i  be a pair of complex roots.
Then the corresponding part of the C.F. may be written in the following
form

yc = e ax (c1 cos  x + c 2 sin  x )

If, however the auxiliary equation has two equal pairs of complex roots,
 + i  and  − i  say occur twice, the corresponding part of the C.F.
is written as
yc = e ax (c1 + c 2 x ) cos  x + (c 3 + c 4 x ) sin  x 

Example 3
Solve
d 4y
4
+ m 4y = 0
dx
Solution
Rewriting the given equation in the symbolic form, we have

(D 4
+ m4 )y = 0

so that its auxiliary equation is D 4 + m 4 = 0

54
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

D4 + m4 = 0
D 4 + m 4 + 2m 2D 2 − 2m 2D 2 = 0

(D + m 2 ) − 2m 2D 2 = 0
2 2

(D 2
)(
+ m 2 − 2mD D 2 + m 2 + 2mD = 0 )
(
for D 2 + m 2 + 2mD )
− 2m  (2m 2
− 4m 2 ) −m m
D= =  i
2 2 2

(
and for D 2 + m 2 − 2mD )
2m  (2m 2
− 4m 2 ) m m
D= =  i
2 2 2
since G(x)=0 in our problem , so general solution will be y=C.F.
hence the solution is given by

y = yc = e −mx / 2
(c cos (mx / 2 ) + c sin (mx / 2 ))
1 2

+e mx / 2
(c cos (mx / 2 ) + c sin (mx / 2 ))
3 4

Case (iii)
If a pair of the roots of the auxiliary equation involves surds, say
   , where  is positive, then the corresponding part of C.F.
in of the following three form

yc = e ax (c1 cosh  x + c 2 sinh  x )

55
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

it should be noted that the above result are exactly similar to those of
case (ii) except that sin and cos have been replaced by sinh and cosh
respectively.

Example 4
Solve

(D 2
+ 6D + 4 ) y = 0

Solution
Here auxiliary equation is D 2 + 6D + 4 = 0 giving D = −3  5
Hence the general solution is
yc = e −3x (c1 cosh 5x + c 2 sinh 5x )

56
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

Determination of the Particular Integral (P.I.)

Let the given differential equation be

f (D )y = G (x ) --------------------(1)
then let
1
G (x )------------------------(2)
f (D )

denote some function of x which when operated upon by f (D ) produces

G (x ). Clearly (1) is satisfied if we take

1
y= G (x )
f (D )

thus from the above definitions, f (D )  1  = G (x ), which shows that

 f (D ) 
1 is the inverse operator of f (D )
f (D )

thus,
1
P .I . = G (x ).
f (D )

Case 1: Finding P.I. when G(x) is of the form e ax

(1) P .I . = 1 ax 1 ax
e = e , f (a )  0.
f (D ) f (a )

(2) If f (a ) = 0 then f (D ) must posses a factor of the type ( D − a ) . In

r

this we apply the following formula:

57
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

1 x r ax (3)
P .I . = e ax
= e , where r = 1,2,3,...
(D − a )
r
r!

if G (x ) be a constant b, say, then it should be modified as be 0x .

Example 5
Let us find

(1) P .I . = 1 1 1
e −2x = e −2x = e −2x
D +D +5
2
(−2) − 2 + 5
2
7
1
(2) P .I . = ex
D + D − D −1
3 2

Here f (D ) = D 3 + D 2 − D − 1 = (D − 1)2 (D + 1).

1 1  1 x
P .I . = f (D ) = ex = e
 (D + 1) 
(D − 1) (D + 1)
2
(D − 1)2  
1  1 x 1
= 2 
e  = ex
(D − 1)  (1 + 1)  2(D − 1) 2

1 1 1
(3) P .I . = 5= 2 5e 0 x = 5 e 0x = 5
D + D +1
2
D + D +1 0 + 0 +1

1 1 1
(4) P .I . = k= 4 ke 0 x = k 2 2 e 0x
D +D 4
2
D +D 2
D (D + 1)
1 1 0x 1 x2
=k 2 e = k 2 .1 = k
D 0 +1 D 2!
1
where .1 means integration of one twice
D2

58
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

Example 6
d 2y dy
Solve 2
−3 + 2y = e x
dx dx
Solution
The given equation in symbolic form is

(D 2
− 3D + 2) y = e x ,

whose auxiliary equation is

(D 2
− 3D + 2) = 0
9D − 1(D − 2) = 0 so D = 1,2

then

C.F. is c1e + c 2e
x 2x

To get P.I.
1 1 1 x −x x
P .I . = ex = e = e
(D − 1)(D − 2) (D − 1)1 − 2 1!
solution is
x x
y = yc + y p = c1e x + c 2e 2x − e
1!
Example 7
Solve
(D 3 − 5D 2 + 7D − 3)y = e 2x cosh x

solution
the auxiliary equation is D 3 − 5D 2 + 7D − 3 = 0

or (D − 1)(D − 1)(D − 3) = 0

D = 1,1,3 then C.F . = (c1 + c 2x ) e x + c 3e 2x .

59
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

1
P .I . = e 2x cosh x
(D − 1) (D − 3)
2

It is noted that

e x + e −x e x − e −x
cosh x = , sinh x =
2 2

−x
2x e + e
x
1 1
P .I . = e cosh x =
2x
e
(D − 1)2 (D − 3) (D − 1)2 (D − 3) 2
1  1 3x 1 x 
= e + e 
(D − 1) (D − 3)  2
2
2 
1 1 1
Now P.I. for e x = ex
2 2 (D − 1) (D − 3)
2

1 1 1
= ex
2 (D − 1) 1 − 3
2

1 1 1 x2 x
= e =−
x
e
2(−2) (D − 1)2 4 2!

1 1 1
Now P.I. for e 3 x = e 3x
2 2 (D − 3)(D − 1)2

1 1 1 3x 1 x 3x
= e = e
2 (D − 3) 4 8 1!
Solution is
y = (c1 + c 2x ) e x + c 3e 2x − (1/8)x 2e x + (1/8)xe 3x

60
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

Case 2: Finding P.I. when G(x) is of the for cos ax or sin bx

Express f (D ) as a function of D 2 , say  (D ) and then replace D 2 by

2

−a 2 . If  (−a 2 )  0 , then we use the following result.

(1) P .I . = 1 1 1
cos ax = cos ax = cos ax
f (D ) (D 2 ) (−a 2 )

Sometimes we cannot form  (D 2 ). Then we shall try to get  (D , D 2 ),

that is, a function of D and D 2 .

(2) if  (−a ) = 0 , then we shall use the formula

2

1 x
sin ax = − cos ax
D2 + a2 2a
1 x
cos ax = sin ax
D +a
2 2
2a
Example8
Let us find
1 1
(1) P.I.= cos 2x = cos 2x
(D ) + D + 1
2
D + D +1
4 2 2 2

1 cos 2x
= cos 2x =
(-2 )
2 2
− 22 + 1 13

1 1 1
(2) P.I.= cos 3x = − cos 3x replace D2 by-32
D2 − 2D + 1 2D +4

multiply by conjugate factor (D − 4)

61
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

−1 D − 4
= cos 3x
2 D 2 − 42
−1 1 1
= (D − 4) 2 cos 3x = (D − 4)cos 3x
2 −3 − 4 2
50
1 1
= D cos 3x − 4 cos 3x  =  −3sin 3x − 4 cos 3x .
50 50
Example 9
Solve
d 4y
4
− m 4y = sin mx
dx
solution
given (D 4 − m 4 )y = sin mx

whose auxiliary equation is D 4 − m 4 = 0 giving D = m,  im

so C.F . = c1e mx + c 2e −mx + c 3 cos mx + c 4 sinmx.

1 1 1
P .I . = sin mx = 2 sin mx
(D + m )(D − m )
2 2
2 2
(D + m ) −2m 2
2

1 1 1  −x  x
= sin mx = 2 
cos x  = cos mx
−2m (D + m )
2 2 2
−2m  2m  4m
3

so the solution is
y = yc + y p
x
= c1e mx + c 2e −mx + c 3 cos mx + c 4 sin mx + cos mx .
4m 3

62
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

m
Case 3: Finding P.I. when G(x) is x or a polynomial of degree m,
m being +ve integer.

Take out the lowest degree term from f (D ) m so as to reduce it in the

form 1 + F (D )n . We now take it to numerator and get 1 + F (D ) which

−n

is then expanded with the help of binomial, namely,

n (n − 1) 2 n (n − 1)(n − 2) 3
(1 + x )n = 1 + nx + x + x + .....
2! 3!

Example 10.
Solve (D 3
− D 2 − 6D ) y = x 2 + 1

Solution
The auxiliary equation is
D(D 2 − D − 6) = 0
or D(D - 3(D + 2) = 0  D = 0,3 - 2

And C.F . = c1e 0x + c 2e 3x + c 3e −2x

63
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

1 1
P .I . = (x 2 + 1) = (x 2 + 1)
D − D − 6D
3 2
−6D (1 + D /6 − D /6)
2

−1
−1   D D 2  
= 1 +  −   (x + 1)
2

6D   6 6 

−1   D D 2   D D 2  
2

= 1 −  − +
  −  + ... (x 2 + 1)
6D   6 6  6 6  

−1  D 7 2  2
= 1 − + D + ... (x + 1)
6D  6 36
−1
 2 D 2 7 2 2 
= ( x + 1) − ( x + 1) + D ( x + 1) + ...
6D 6 36
−1  2 1 7 
= ( x + 1) − (2 x + 0) + (2 + 0) + ...
6D  6 36 
−1  2 x 25  −1  x 3
x 2
25 
= x − + =  − + x
6D  
3 18  6  3 6 18 

hence the required solution is

y = C .F . + P .I .
−2 x 1  x 3 x 2 25 
y = c1e 0x
+ c 2e 3x
+ c 3e −  − + x
6 3 6 18 

Example 11
Solve (D 3 + 8)y = x 4 + 2x + 1
Solution
The auxiliary equation is (D + 2)(D 2 − 2D + 4) = 0

So D = −2,1  i 3

(
C .F . = c1e −2x + c 2e x cos x 3 + c 3 . )
1 1
P .I . =
D +8
3 ( x 4 + 2x + 1) =
 D 3

(x 4
+ 2x + 1)
8 1 +
 8 

64
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

−1
1  D3  1  D3 
= 1 +
8 8 
( x + 2x + 1) = 8 1 − 8 + ... ( x 4 + 2x + 1)
4

 
1 4 D3 4 
= ( x + 2x + 1) − ( x + 2x + 1) + ...
8 8 
1 1
= ( x 4 + 2x + 1) − 3x  = ( x 4 − x + 1) 
8   8 

(
y = c1e −2x + c 2e x cos x 3 + c 3 + ) 1
8
( x 4 − x + 1) 


ax
Case 4: Finding P.I. when G(x) is e V , where V is any function
of x.

1 ax 1 1
e V = e ax V , where V , is calculated with the help
f (D ) f (D + a ) f (D + a )
of previous methods.
To get f (D + a ), replace D by D + a this method will be used if V is
m
cosax or sinax or x or a polynomial of degree m.

Example 12
Solve the differential equation (D 2
+ 3D + 2) y = e 2x sin x .

Solution

(
Auxiliary equation D 2 + 3D + 2 = 0 gives D=-2,-1. )
So, we have

c .F . = c1e −2x + c 2e − x

65
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

1 1
= e 2x sin x = e 2x sin x
D + 3D + 2
2
(D + 2) + 3(D + 2) + 2
2

1 1
P.I.= = e 2x 2 sin x = e 2x 2 sin x
D + 7D + 12 −1 + 7D + 12
11 − 7D 1
= e 2x sin x = e 2x (11 − 7D ) sin x
(11 − 7D )(11 + 7D ) 121 − 49D 2
1 e 2x
= e (11 − 7D )
2x
sin x = (11sin x − 7 cos x ) .
121 + 49 170

e 2x
y = c1e −2 x
+ c 2e −x
+ (11sin x − 7 cos x )
170
Example 13
Solve

(D 2
− 2D + 1) y = x 2e 3 x

Solution
Here Auxiliary equation is ( D 2 − 2D + 1) = 0  D = 1,1

So C.F . = (c1 + c 2x ) e x

1 1 1 3x 1
P .I . = x 2e 3 x = e 3 x x2 = e x2
( D − 1) ( D + 3 − 1) (1 + D /2)
2 2 2
4
1 3x 1
e (1 + D /2) x 2 = e 3 x 1 − 2(D /2) + 3(D /2)2  x 2
−2
=
4 4
1 1
= e 3 x  x 2 − Dx 2 + (3/4)x 2  = e 3 x  x 2 − 2x + 3/2 .
4 4
then the solution is
1 3x 2
y = (c1 + c 2x ) e x + e  x − 2x + 3/2 .
4

66
Chapter 4: Linear D.E. With Constant Coefficient Associate Prof. Islam M. Eldesoky

Exercise No. (4)

Solve the following differential equations

1. (D 3 − 4D 2 + 5D − 2)y = 0
2. (D 3 + 6D 2 + 12D + 8)y = 0
3. (D 4 + D 2 + 1)y = 0
d 3y d 2y dy
4. − 4 + 5 − 2y = 0
dx 3 dx 2 dx
5. (D 4 − 81)y = 0
6.(D 4 − t 4 )y = 0
d 2y dy
7. 2
− 3 + 2y = e x
dx dx
4
d y d 3y d 2y dy
8. 4 − 2 3 + 5 2 − 8 + 4y = e x
dx dx dx dx
9. (D − 1)y = cosh x
2

10. (D 3 + 1)y = (e x + 1)2

11. (D 2 − 2D + 1)y = cos 3x
12. (D 2 + 9)y = cos 2x + sin 2x
13. (D 4 − a 4 )y = x 4
14. (D 3 − 8)y = x 3
15. (D 2 − 1)y = cosh x cos
16. (D 4 − 1)y = e x cos x
17. D 2y = e x cos x
18. (D 2 − 4D + 1) = e 2x sin x
19. (D 2
− 2D + 1) y = x 2e 3 x

67