UNIT I : LINEAR DIFFERENTIAL EQUATIONS (LDE)

CHAPTER-1
LINEAR DIFFERENTIAL EQUATIONS WITH CONSTANT COEFFICIENTS

1.1 INTRODUCTION
Differential equations are widely used in fields of Engineering and Applied Sciences. Mathematical formulations of most of the
physical problems are in the forms of differential equations. Use of differential equations is most prominent in subjects like
Circuit Analysis, Theory of Structures, Vibrations, Heat Transfer, Fluid Mechanics etc. Differential equations are of two types :
Ordinary and Partial Differential Equations. In ordinary equations, there is one dependent variable depending for its value on one
independent variable. Partial differential equations will have more than one independent variables.
In what follows, we shall discuss ordinary and partial differential equations, which are of common occurrence in engineering
fields. Applications to some areas will also be dealt.
1.2 PRELIMINARIES
I. Second Degree Polynomials and Their Factorization
(a) (i) D2 – 2D – 3 = (D + 1) (D – 3) (ii) D2 + 5D + 6 = (D + 2) (D + 3)
(iii) D2 + 2D + 1 = (D + 1)2 (iv) D2 – 5D + 6 = (D – 2) (D – 3)
2
(v) D + 3D + 2 = (D + 2) (D + 1) (vi) D2 – D – 2 = (D – 2) (D + 1)
(vii) D2 – 4D + 4 = (D – 2)2 (viii) D2 – a2 = (D – a) (D + a)
2 2
(ix) D + a = (D + ia) (D – ia)
–b± b2 – 4ac
(b) The roots of ax2 + bx + c = 0 are x = , these roots are imaginary if b2 – 4ac < 0.
2a
–2± 4–8
(i) D2 + 2D + 2 = 0 ⇒ D = = –1±i
2
–1± 1–4 –1 3
(ii) D2 + D + 1 = 0 ⇒ D = = ± i
2 2 2
–1 3 1 3
If D = ± i = α ± iβ then α = – , β = , β is always positive; α may be positive, negative or zero.
2 2 2 2
(iii) D2 + 1 = 0 ⇒ D2 = – 1 i.e. D=±i ∴ α = 0, β = 1.
2
(iv) D + 4 = 0 ⇒ D 2
= –4 i.e. D = ± 2i ∴ α = 0, β = 2.
II. Third Degree Polynomials and Their Factorization
(a) (i) D3 – a3 = (D – a) (D2 + aD + a2) (iii) D3 + a3 = (D + a) (D2 – aD + a2)
3 2 3
(ii) D + 3D + 3D + 1 = (D + 1) (iv) D3 – 3D2 + 3D – 1 = (D – 1)3
(b) Use of Synthetic Division :
(i) f(D) = D3 – 7D – 6 = 0; for D = – 1, f(–1) = 0
∴ (D + 1) is one of the factors. –1 1 0 –7 –6
∴ D3 – 7D – 6 = 0 ⇒ (D + 1) (D2 – D – 6) = 0
(D + 1) (D – 3) (D + 2) = 0 ⇒ D = – 1, – 2, 3. –1 1 6
1 –1 –6 0

(ii) For D3 – 2D + 4 = 0; D = – 2
∴ f(–2) = 0 ∴ (D + 2) is one of the factors. –2 1 0 –2 4
∴ D3 – 2D + 4 = 0 ⇒ (D + 2) (D2 – 2D + 2) = 0 –2 4 –4
D = – 2 and D = 1 ± i, α = 1, β = 1. 1 –2 2 0

(1.1)
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.2) LINEAR DIFFERENTIAL EQUATIONS…
…

III. Fourth Degree Polynomials and Their Factorization

(a) D4 – a4 = (D2 – a2) (D2 + a2) = (D – a) (D + a) (D + ia) (D – ia)
(b) Making a Perfect Square by Introducing a Middle Term :
(i) For D4 + a4 = 0 ; consider (D2 + a2)2 = D4 + 2a2 D2 + a4
2
D4 + a4 = (D4 + 2a2 D2 + a4) – (2a2 D2) = (D2 + a2)2 – ( 2 a D)

D4 + a4 = (D 2
– 2 a D + a2) (D 2
+ 2 a D + a2)
2
(ii) For D4 + 1 = D4 + 2D2 + 1 – 2D2 = (D2 + 1)2 – ( 2 D)

D4 + 1 = (D 2
– 2 D + 1) (D 2
+ 2 D + 1)

(c) D4 + 8D2 + 16 = (D2 + 4)2, D4 + 2D2 + 1 = (D2 + 1)2 = (D + i)2 (D – i)2

D4 + 10D2 + 9 = (D2 + 9) (D2 + 1) = (D + 3i) (D – 3i) (D + i) (D – i)
(d) (i) f(D) = D4 – 2D3 – 3D2 + 4D + 4 = 0,
for D = – 1, f (–1) = 0 –1 1 –2 –3 4 4
2 2
∴ Factors are (D + 1) (D – 2) = 0. –1 3 0 –4
–1 1 –3 0 4 0
–1 4 –4
2 1 –4 4 0
2 –4
1 –2 0

On a similar line,
(ii) D4 – D3 – 9D2 – 11 D – 4 = (D + 1)3 (D – 4)
(e) Perfect square of the type (a + b + c)2
(i) D4 + 2D3 + 3D2 + 2D + 1 = (D2)2 + 2 · D2 · D + D2 + 2D2 + 2D + 1 = (D2 + D)2 + 2 (D2 + D) + 1
= [(D2 + D) + 1]2 = (D2 + D + 1)2
(ii) D4 – 4D3 + 8D2 – 8D + 4 = (D2)2 – 2D2 · 2D + (2D)2 + 4D2 – 8D + 4
= (D2 – 2D)2 + 4 (D2 – 2D) + 4
= [(D2 – 2D) + 2]2 = (D2 – 2D + 2)2
IV. Fifth Degree Polynomials and Their Factorization
(i) D5 – D4 + 2D3 – 2D2 + D – 1 = D4 (D – 1) + 2D2 (D – 1) + 1 (D – 1) = (D4 + 2D2 + 1) (D – 1) = (D – 1) (D2 + 1)2
= (D – 1) (D + i)2 (D – i)2

1.3 THE nth ORDER LINEAR DIFFERENTIAL EQUATION WITH CONSTANT COEFFICIENTS
A differential equation which contains the differential coefficients and the dependent variable in the first degree, does not involve
the product of a derivative with another derivative or with dependent variable, and in which the coefficients are constants is
called a linear differential equation with constant coefficients.
The general form of such a differential equation of order "n" is

dny dn–1 y dn–2 y dy

a0 n + a1 n–1 + a2 + … + an–1 + an y = f(x) … (1)
dx dx dxn–2 dx

Here a0, a1, a2 … are constants. Equation (1) is a nth order linear differential equation with constant coefficients.
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.3) LINEAR DIFFERENTIAL EQUATIONS…
…

e.g. Put n = 3 in equation (1), we get

d3y d2y dy
a0 3 + a1 + a2 + a3 y = f(x) which is a 3rd order linear differential equation with constant coefficients.
dx dx2 dx
d dy 2 d2y dn y
Using the differential operator D to stand for i.e. Dy = ; D y = 2 , … Dny = , the equation (1) will take the form
dx dx dx dx
a0 Dn y + a1 Dn–1 y + a2 Dn–2 y + … + an–1 Dy + an y = f(x)
OR (a0 Dn + a1 Dn–1 + a2 Dn–2 + … + an-1 D + an) y = f(x) … (2)
in which each term in the parenthesis is operating on y and the results are added.
Let φ(D) ≡ a0 Dn + a1 Dn–1 + a2 Dn–2 + … + an–1 D + an , φ(D) is called as nth order polynomial in D.

∴ Equation (2) can be written as φ(D) y = f(x) … (3)

Note : In equation (1), if a0, a1, … an are functions of x then it is called nth order linear differential equation.
1.4 THE NATURE OF DIFFERENTIAL OPERATOR "D"
d
It is convenient to introduce the symbol D to represent the operation of differentiation with respect to x i.e. D ≡ , so that
dx
dy d2 y 2 d3 y 3 dn y dy
= Dy; 2 = D y; 3 = D y; ……; = Dny and + ay = (D + a) y
dx dx dx dxn dx
The differential operator D or (Dn) obeys the laws of Algebra.
Properties of the Operator D :
If y1 and y2 are differentiable functions of x and "a" is a constant and m, n are positive integer then
(i) Dm (Dn) y = Dn (Dm) y = Dm+n y (ii) (D – m1) (D – m2) y = (D – m2) (D – m1) y
2
(iii) (D – m1) (D – m2) y = [D – (m1 + m2) D + m1 m2] y (iv) D (au) = a · D(u); Dn (au) = a · Dn (u)
(v) D (y1 + y2) = D (y1) + D(y2); Dn (y1 + y2) = Dn (y1) + Dn (y2).

1.5 LINEAR DIFFERENTIAL EQUATION φ (D) y = 0

Consider, (D) y = 0 … (4)
n n–1 n–2 n–3 th
where, φ(D) = a0 D + a1 D + a2 D + a3 D + … + an–1 D + an is n order polynomial in D and D obeys the laws of algebra,
we can in general factorise φ(D) in n linear factors as φ(D) = (D – m1) (D – m2) (D – m3) … (D – mn) where m1, m2, m3, … mn are the
roots of the algebraic equation φ(D) = 0
∴ Equation (4) can be written as;
φ(D) y = (D – m1) (D – m2) (D – m3) … (D – mn) y = 0 … (5)
Note : These factors can be taken in any sequence.

1.6 AUXILIARY EQUATION (A.E.)

The equation φ(D) = 0 is called as an auxiliary equation (A.E.) for equations (3), (4).
d2 y dy
e.g. –5 + 6y = 0
dx2 dx
d
By using operator D for , we have (D2 – 5D + 6) y = 0
dx
φ(D) = D2 – 5D + 6 = 0 is the A.E.
∴ (D2 – 5D + 6) y = (D – 3) (D – 2) y = (D – 2) (D – 3) y.
1.7 SOLUTION OF φ(D) y = 0
Being nth order DE, equation (4) or (5) will have exactly n arbitrary constants in its general solution.
The equation (5) will be satisfied by the solution of the equation (D – mn) y = 0
dy
i.e. – mn y = 0
dx
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.4) LINEAR DIFFERENTIAL EQUATIONS…
…

On solving this 1st order 1st degree DE by separting variables, we get y = cn emnx, where, cn is an arbitrary constant.
Similarly, since the factors in equation (5) can be taken in any order, the equation will be satisfied by the solution of each of the
mx mx
equations (D – m1) y = 0, (D – m2) y = 0 … etc., that is by y = c1 e 1 , y = c2 e 2 ………… etc.
It can, therefore, easily be proved that the sum of these individual solutions, i.e.
mx mx mx
y = c1 e 1 + c2 e 2 + … + cn e n … (6)
also satisfies the equation (5) and as it contains n arbitrary constants, and the equation (4) is of the nth order, (6) constitutes the
general solution of the equation (4).
∴ The general solution of the equation φ(D) y = 0 is
mx mx mx
y = c1 e 1 + c2 e 2 + … + cn e n
where m1, m2, … mn are the roots of the auxiliary equation φ(D) = 0.
d3 y d2 y dy
Ex. 1 : Solve 3 – 6 + 11 – 6y = 0.
dx dx2 dx
d
Sol. : Let D stand for and the given equation can be written as
dx
(D3 – 6D2 + 11D – 6) y = 0.
Here auxiliary equation is
D3 – 6D2 + 11D – 6 = 0
i.e. (D – 1) (D – 2) (D – 3) = 0 ⇒ m1 = 1, m2 = 2, m3 = 3, are roots of AE.
∴ The general solution is y = c1 ex + c2 e2x + c3 e3x .
Ex. 2 : For (4D2 – 8D + 1) y = 0.
3
Sol. : Here auxiliary equation is 4D2 − 8D + 1 = 0 i.e. D = 1 ±
2
 3  3
1 + x 1 – 2  x
y = c1 e 2  + c2 e  .
1.8 DIFFERENT CASES DEPENDING UPON THE NATURE OF ROOTS OF THE AUXILIARY EQUATION
φ(D) = 0
A. The Case of Real and Different Roots
If roots of φ (D) = 0 be m1, m2, m3 … mn, all are real and different, then the solution of φ (D) y = 0 will be
m1x mx mx mx
y = c1 e + c2 e 2 + c3 e 3 + … + cn e n
B. The Case of Real and Repeated Roots (The Case of Multiple Roots)
Let m1 = m2, m3, m4 … mn be the roots of φ(D) = 0, then the part of solution corresponding to m1 and m2 will look like
mx mx mx mx
c1 e 1 + c2 e 1 (m1 = m2) = (c1 + c2) e 1 = c'e 1
But this means that number of arbitrary constants now in the solution will be n – 1 instead of n. Hence it is no longer the
general solution. The anomaly can be rectified as under.
Pertaining to m1 = m2, the part of the equation will be (D – m1) (D – m1) y = 0
Put (D – m1) y = z, temporarily, then we have (D – m1) z = 0 ∴ z = c1 em1x
Hence putting value of z in (D – m1) y = z, we have
mx dy mx
(D – m1) y = c1 e 1 or – m1 y = c1 e 1
dx
– ∫ m1dx – m1x
which is a linear differential equation. Its I.F. = e =e and hence solution is

⌠
(
y e
– m1 x
) = ⌡ c1 e
m1x
·e
–m1x
dx + c2 = c1 x + c2

mx
∴ y = (c1 x + c2) e 1
If m1 = m2 are real, and the remaining roots m3, m4, m5, …, mn are real and different then solution of φ(D) y = 0 is
mx mx mx mx
y = (c1 x + c2) e 1 + c3 e 3 + c4 e 4 + … + cn e n
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.5) LINEAR DIFFERENTIAL EQUATIONS…
…

Similarly, when three roots are repeated. i.e. if m1 = m2 = m3 are real, and the remaining roots m4, m5, … mn are real and
different then solution of φ(D) y = 0 is
mx mx mx
y = (c1 x2 + c2 x + c3) e 1 + c4 e 4 + … + cn e n

If m1 = m2 = m3 = … = mn i.e. n roots are real and equal then solution of φ(D) y = 0 is

m1x
y = (c1 xn–1 + c2 xn–2 + … + cn–1 x + cn) e
Ex. 1. For (D2 – 6D + 9) y = 0 A.E. = (D – 3)2 = 0 and solution is y = (c1 x + c2) e3x
2. For (D – 1)3 (D + 1) y = 0, solution is y = (c1 x2 + c2 x + c3) ex + c4 e–x
3. For (D – 1)2 (D + 1)2 y = 0, solution is y = (c1 x + c2) ex + (c3 x + c4) e–x.
C. The Case of Imaginary (Complex) Roots
For practical problems in engineering, this case has special importance. Since the coefficients of the auxiliary equation
are real, the imaginary roots (if exists) will occur in conjugate pairs. Let α ± iβ be one such pair. Therefore m1 = α + iβ,
m2 = α – iβ
The corresponding part of the solution of the equation φ(D) y = 0, then takes the form
(α + iβ) x (α – iβ) x
y = Ae + Be
y = eαx [Aeiβx + Be−iβx]
y = eαx [A (cos βx + i sin βx) + B (cos βx – i sin βx)]
y = eαx [(A + B) cos βx + i (A – B) sin βx]
y = eαx [c1 cos β x + c2 sin β x]
where, c1 = A + B and c2 = i (A – B) are arbitrary constants.
Using c1 = C cos θ, c2 = – sin θ, this can also be put sometimes into the form as given below (recall SHM).
y = C eαx cos (β
β x + θ) where C,, θ are arbitrary constants.

ILLUSTRATIONS
2
Ex. 1 : Solve (D + 2D + 5) y = 0.
Sol. : The auxiliary equation is D2 + 2D + 5 = 0 whose roots are D = – 1 ± 2i which are both imaginary. Here α = – 1, β = 2.
Hence the solution is
y = e–x [A cos 2x + B sin 2x]
4 2
dy dy dy
Ex. 2 : Solve – 5 2 + 12 + 28y = 0.
dx4 dx dx
Sol. : The auxiliary equation is D4 – 5D2 + 12D + 28 = 0 having roots D = – 2, –2, 2 ± 3 i. (Here α = 2, β = 3).
Hence the solution is
y = (c1 x + c2 ) e–2x + e2x [A cos 3 x + B sin 3 x]
2
Ex. 3 : For (D + 4)y = 0.
Sol. : The auxiliary equation is D2 + 4 = 0 having roots D = 0 ± 2i (Here α = 0, β = 2).
Hence the solution is
y = A cos 2x + B sin 2x.
D. The Case of Repeated Imaginary Roots
If the imaginary roots m1 = α + iβ and m2 = α – iβ occur twice, then the part of solution of φ (D) y = 0 will be
mx mx
y = (A x + B) e 1 + (C x + D ) e 2 … (by using case B)
(α + iβ) x
= (A x + B) e + (C x + D) e(α – iβ) x
= eαx [(A x + B) eiβx + (C x + D) e–iβx]
= eαx [(A x + B) {cos βx + i sin βx} + (C x + D) {cos βx – i sin βx}]
= eαx [(A x + B + C x + D) cos βx + i (A x + B – C x – D) sin βx]
∴ y = eαx [(c1 x + c2) cos β x + (c3 x + c4) sin β x]
with proper changes in the constants c1, c2, c3 and c4.
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.6) LINEAR DIFFERENTIAL EQUATIONS…
…

ILLUSTRATIONS
6 4 2
dy dy dy
Ex. 1 : Solve + 6 4 +9 = 0.
dx6 dx dx2
Sol. : The auxiliary equation D6 + 6D4 + 9D2 = 0 has roots D = 0, 0, ± i 3,±i 3 where the imaginary roots ± i 3 are
repeated. Hence the solution is
y = c1 x + c2 + (c3 x + c4) cos 3 x + (c5 x + c6) sin 3 x
4 2
Ex. 2 : (D + 2D + 1) y = 0.
Sol. : The auxiliary equation D4 + 2D2 + 1 = 0 has roots D = ± i, ± i, repeated imaginary roots.
Hence the solution is
y = (c1x + c2) cos x + (c3x + c4) sin x
Now we will summarise the four cases for ready reference.
Case 1 : Real & Distinct Roots : A.E. ⇒ (D – m1) (D – m2) (D – m3) …(D – mn) = 0
mx mx mx
Solution is y = c1 e 1 + c2 e 2 + c3 e 3 + … + cn emnx
Case 2 : Repeated Real Roots
(i) For m1 = m2 ⇒ A.E. ⇒ (D – m1) (D – m1) (D – m3) …… (D – mn) = 0
m1x mx
Solution is y = (c1 x + c2) e + c3 e 3 + … + cn emnx
(ii) For m1 = m2 = m3 ⇒ A.E. ⇒ (D – m1) (D – m1) (D – m1) (D – m4) … (D – mn) = 0
m1x mx mx
Solution is y = (c1 x2 + c2 x + c3) e + c4 e 4 + … + cn e n
Case 3 : Imaginary Roots : For D = α ± i β
Solution is y = eαx [c1 cos β x + c2 sin β x]
Case 4 : Repeated Imaginary Roots : For D = α ± iβ be repeated twice
Solution is y = eαx [(c1 x + c2) cos β x + (c3 x + c4) sin β x]

ILLUSTRATIONS
2
d x
Ex.1 : Solve + 4x = 0.
dt2
d
Sol. : Let D stand for .
dt
∴ A.E. : D2 + 4 = 0 ⇒ D = 0 ± 2i
∴ The solution is x = c1 cos 2t + c2 sin 2t.
d4 y
Ex. 2 : Solve – 16y = 0.
dz2
d
Sol. : Let D stand for .
dz
∴ A.E. : D4 – 16 = 0, (D – 2) (D + 2) (D2 + 4) = 0.
∴ The solution is y = c1 e2z + c2 e–2z + c3 cos 2z + c4 sin 2z.
Special Case : If the two real roots of φ(D) y = 0 be m and – m [e.g. D2 – m2 = 0], then the corresponding part of the solution is
y = A emx + B e–mx
OR y = A (cosh mx + sinh mx) + B (cosh mx – sinh mx)
OR y = (A + B) cosh mx + (A – B) sinh mx
∴ y = c1 cosh mx + c2 sinh mx
We note here that (in some particular cases) solution of D2 – m2 = 0 can be written as
y = c1 emx + c2 e–mx or y = c1 cosh mx + c2 sinh mx.
e.g. 1. (D – 1) y = 0 ⇒ y = c1 cosh x + c2 sinh x.
2

2. (D2 – 4) y = 0 ⇒ y = c1 cosh 2x + c2 sinh 2x.

ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.7) LINEAR DIFFERENTIAL EQUATIONS…
…

EXERCISE 1.1
Solve the following Differential Equations
d2 y dy d2y dy
1. –5 – 6y = 0. 2. 2 – – 10y = 0.
dx2 dx dx2 dx
Ans. y = c1 e–x + c2 e6x Ans. y = c1 e–2x + c2 e(5/2) x
d3 y d2y dy
3. 3 + 2 + = 0. 4. (D4 – 2D3 + D2) y = 0.
dx dx2 dx
Ans. y = c1 + e–x (c2 x + c3) Ans. y = c1 x + c2 + (c3 x + c4) ex
5. (D6 – 6D5 + 12D4 – 6D3 – 9D2 + 12D – 4) y = 0. 6. (D3 + 6D2 + 11D + 6) y = 0 .
Ans. y = (c1x2 + c2x + c3) ex + (c4x + c5) e2x + c6e–x Ans. y = c1e–x + c2e–2x + c3e–3x
d2x dx
7. 4y" – 8y' + 7y = 0. 8. +2 + 5x = 0, x (0) = 2, x'(0) = 0.
dt2 dt
 
3   3 
Ans. y = ex A cos  x + B sin  x Ans. x = e–t (2 cos 2t + sin 2t)
 
2   2 
d2 s ds ds
9. = – 16 – 64 s, s = 0, = – 4 when t = 0. 10. (D3 + D2 – 2D + 12) y = 0.
dt2 dt dt
Ans. s = – 4e8t t Ans. y = c1 e–3x + ex [A cos 3 x + B sin 3x
]
11. (D2 + 1)3 (D2 + D + 1)2 y = 0.
  3   3 
Ans. y = (c1 + c2 x + c3 x2) cos x + (c4 + c5 x + c6 x2) sin x + e–x/2 (c7 + c8 x) cos  x + (c9 + c10 x) sin  x
  2   2 
d4 y
12. + m4y = 0.
dx4
(mx/ 2) A cos mx + B sin mx + e– (mx/ 2) C cos mx + D sin mx
Ans. y = e   2  2   2  2
         
d2s 3t 3t
13. 4 = – 9s. Ans. s = c1 sin + c2 cos
dt2 2 2
d2y
14. The equation for the bending of a strut is EI + Py = 0. If y = 0 when
dx2
P
a sin x
l EI
x = 0 and y = a when x = , find y. Ans. y =
2 P l
sin ·
EI 2

1.9 THE GENERAL SOLUTION OF THE LINEAR DIFFERENTIAL EQUATION φ(D) y = f(x)
The general solution of the equation φ(D)y = f(x) can be written as y = yc + yp where,

1. yc is the solution of the given equation with f(x) = 0, that is of equation φ(D) y = 0 (which is known as Associated
equation or Reduced equation) and is called the complimentary function (C.F.). It involves n arbitrary constants and is
denoted by C.F. then φ(D) yc = 0 .

2. yp is any function of x, which satisfies the equation φ(D) y = f(x), so that φ(D) yp = f(x) yp is called the particular
integral and is denoted by P.I. It does not contain any arbitrary constant.
Thus, on substituting y = yc + yp in φ(D) y,
φ(D) [yc + yp] = φ(D) yc + φ(D) yp = 0 + f(x) = f(x)
∴ y = yc + yp satisfies the equation φ(D) y = f(x) and as it contains exactly n arbitrary constants, is the general (or complete)
solution of the equation.
Note : 1. The complete solution of φ(D) y = f(x) is y = C.F. + P.I. = yc + yp.
2. The general solution of φ(D) y = f(x) has arbitrary constants equal in number to the order of the differential equation.
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.8) LINEAR DIFFERENTIAL EQUATIONS…
…

1
1.10 THE INVERSE OPERATOR AND THE SYMBOLIC EXPRESSION FOR THE PARTICULAR INTEGRAL
φ(D)
1
We define f(x) as that function of x which, when acted upon by the differential operator φ(D) gives f(x).
φ(D)
 1   1 
Thus by this definition, φ(D)  f(x) = f(x) and so  f(x) satisfies the equation φ(D) y = f(x) and so is the P.I. of the
φ(D)  φ(D) 
equation φ(D) y = f(x).
Thus the P. I. of the equation φ(D) y = f(x) is symbolically given by;

1
P.I. = yp = f(x)
φ(D)

1
e.g. 1. (D2 – 1) y = x2 ∴ yp = x2
D2 – 1
1
2. (D2 – 3D + 2) y = sin ex ∴ yp = sin ex.
D2 – 3D + 2

1.11 METHODS OF OBTAINING PARTICULAR INTEGRAL

1
There are three methods to evaluate the particular integral yp = f(x).
φ(D)
(A) General method
(B) Short-cut methods
(C) Method of variation of parameters.
Now we will discuss these methods in detail.
(A) General Method
This method is useful when the short-cut methods given in (B) are not applicable. This method involves integration.
1 1
(i) f(x) : By definition of the P. I., f(x) will be the P.I. of the equation (D – m) y = f(x) i.e. the part in the solution of
D–m D–m
dy
this equation which does not contain the arbitrary constant. We have, – my = f(x) (linear)
dx

I.F. = e–mx and the general solution is

y e–mx = ⌠
⌡ f(x) · e–mx dx + c1

emx ⌠ 
∴ y = (c1 emx) +
 ⌡ e–mx f(x) · dx

i.e. y = yc + yp

Here c1 emx is the C.F. and emx ⌠ –mx

⌡ e f(x) dx must be the P.I.

∴ yp = P. I. =
1 ⌠
f(x) = emx ⌡ e–mx f(x) dx
D–m

1
f(x) = e–mx
⌠ emx f(x) dx
Similarly, yp = P. I. =
D+m ⌡

Put m = 0 yp =
1 ⌠
f(x) = ⌡ f(x) dx
D
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.9) LINEAR DIFFERENTIAL EQUATIONS…
…

Also, yp =
1
f(x) =
1  1 f(x)
D2 D D 
⌠  ⌠ 
⌡ f(x) dx = ⌠
1
= ⌡ ⌡ f(x) dx dx
D    
1
⌠  
∴ yp =
D2
f(x) = ⌡ ⌠
⌡ f(x) dx dx


1 ⌠ ⌠ ⌠  
Similarly, yp = f(x) = ⌡ ⌡ ⌡ f(x) dx dx dx … and so on.
D3    
1
(ii) f(x) :
(D – m1) (D – m2)

1 1 mx ⌠ – m2x
yp =
(D – m1) (D – m2)
f(x) =
(D – m1)
e 2 ⌡ e f(x) dx

m1x ⌠ –m1x  mx⌠ – m2 x 

yp = e ⌡ e e 2 ⌡ e f(x) dx dx
 
(iii) Use of Partial Fraction :
1
yp = f(x)
(D – m1) (D – m2)

=
1  1 – 1  f(x)
(m1 – m2) D – m1 D – m2
1  1 1 
=  f(x) – f(x)
m1 – m2  D – m1 D – m2 
1  mx⌠ m x ⌠ –m x 
yp = e 1 ⌡ e 1 f(x) dx – e 2 ⌡ e 2 f(x) dx
–m x
m1 – m2  
ILLUSTRATIONS ON GENERAL METHOD

d2y dy x
Ex. 1 : Solve 2 + 3 + 2y = ee (Dec. 2010, 2016, 2017; May 2011)
dx dx

Sol. : For C.F., A.E. is D2 + 3D + 2 = 0 ⇒ (D + 2) (D + 1) = 0. Hence D = –1, –2

and C.F. = c1 e–x + c2 e–2x

1 x
Here P. I. = yp = (ee )
(D + 2) (D + 1)

=
1  1 eex
D + 2 D + 1 
 ⌠ 
1 e–x  x 
ex ee dx
=
D+2  ⌡  [put ex = t ∴ ex dx = dt]

 ⌠ 
=
1 e–x  t
e dt

D+2  ⌡ 
x
1
=
D+2 [e –x
ee]=e
x –2x
⌠ 2x –x e
⌡e e e dx
x x
⌠ ex ee dx = e–2x ee
P. I. = e–2x ⌡

Hence the complete solution will be

x
y = c1 e–x + c2 e–2x + e–2x ee
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.10) LINEAR DIFFERENTIAL EQUATIONS…
…

d2y dy 1
Ex. 2 : Solve + = . (Dec. 2008, 2010, 2012; May 2012)
dx2 dx 1 + ex
1 d
Sol. : We have (D2 + D) y = , here D ≡
1 + ex dx
For C.F., A.E. ⇒ D (D + 1) = 0 ∴ D = 0, – 1.
∴ C.F. = yc = c1 + c2 e–x

P.I. =
1  1 
D (D + 1) 1 + ex

=
 1 – 1   1 x by partial fraction
D D + 1 1 + e 
1  1  1  1 
= –
D 1 + ex D + 1 1 + ex

⌠ 1 dx
⌠ ex 1 + ex
dx – e–x ⌡
= ⌡ 1 + ex

⌠ ex dx x
– e–x ⌠ x dx putx 1 + e = t
= ⌡ e (1 + ex)
x ⌡ e 1 + ex  e dx = dt 
⌠ dt
– e–x ⌠
dt
= ⌡ t (t – 1) ⌡ t

⌠  1 – 1 dt – e–x log (ex + 1)

= ⌡ t – 1 t 
= log (t – 1) – log t – e–x log (ex + 1)
= log (ex) – log (1 + ex) – e–x log (ex + 1)
= x – log (1 + ex) – e–x log (ex + 1)
Hence the complete solution is
y = c1 + c2 e–x + x – log (1 + ex) – e–x log (1 + ex)
Ex. 3 : Solve (D + 5D + 6) y = e sec2 x (1 + 2 tan x)
2 –2x
(Dec. 2005, May 2008, 2011)
Sol. : A.E. is D2 + 5D + 6 = 0 gives (D + 2) (D + 3) = 0 ⇒ D = – 2, – 3
C.F. = c1 e–2x + c2 e–3x
1
P.I. = [e–2x sec2 x (1 + 2 tan x)]
(D + 3) (D + 2)
1  –2x ⌠ 
= e ⌡ e2x · e–2x sec2 x (1 + 2 tan x) dx
D+3  
1  –2x ⌠ 
= e ⌡ sec2 x (1 + 2 tan x) dx put tan x = t, sec2 x dx = dt
D+3  
1  –2x ⌠  1
= e ⌡ (1 + 2t) dt = [e–2x (t + t2)]
D+3   D+3
1
= [e–2x (tan x + tan2 x)]
D+3

= e–3x ⌠
⌡ e3x · e–2x [(tan x – 1) + sec2 x] dx

⌠ ex [(tan x – 1) + sec2 x] dx
= e–3x ⌡

= e–3x [ex (tan x – 1)] ⌠

‡⌡ ex [f(x) + f'(x)] dx = ex f(x)

= e–2x (tan x – 1)
Hence the complete solution is
y = c2 e–3x + e–2x [c1 + tan x – 1]
= c2 e–3x + e–2x [c3 + tan x]
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.11) LINEAR DIFFERENTIAL EQUATIONS…
…

d2y
Ex. 4 : Solve + 9y = sec 3x (May 2008)
dx2
Sol. : A.E. is D2 + 9 = 0, or D = ± 3i
C.F. = c1 cos 3x + c2 sin 3x and
1 1
P.I. = (sec 3x) = sec 3x
D2 + 9 (D + 3i) (D – 3i)

=
1  1 – 1  sec 3x
6i D – 3i D + 3i
1 1 1 1
= sec 3x – sec 3x … (1)
6i D – 3i 6i D + 3i

1
sec 3x = e3ix
⌠ e–3ix sec 3x dx = e3ix
⌠ cos 3x – i sin 3x
Now,
D – 3i ⌡ ⌡ cos 3x
dx

= e3ix
⌠
⌡ [1 – i tan 3x] dx

= e3ix x +
 i
log (cos 3x)

 3 
Changing i to – i in this, we have

(sec 3x) = e–3ix x – log (cos 3x)

1 i
D + 3i  3 
Putting values in (1), we have

P.I. =
1 e3ix x + i log (cos 3x) – e–3ix x – i log (cos 3x)
6i   3   3 
x 3ix e3ix log (cos 3x) x e–3ix e–3ix log (cos 3x)
= ·e + – +
6i 18 6i 18
Combining the like terms, we get
e
3ix
x – e–3ix 1 e
3ix
+ e–3ix
= + log (cos 3x)
3  2i  9  2 
x 1
P.I. = sin 3x + cos 3x log (cos 3x)
3 9
Hence the general solution will be
x 1
y = c1 cos 3x + c2 sin 3x + sin 3x + cos 3x · log (cos 3x)
3 9
d2 y dy 1 1
Ex. 5 : Solve – – 2y = 2 log x + + 2 (May 2006, Dec. 2012)
dx2 dx x x
1 1
Sol. : (D2 – D – 2) y = 2 log x + + 2
x x
A.E. : D2 – D – 2 = 0 ∴ (D – 2) (D + 1) = 0
∴ yc = c1 e2x + c2 e–x

yp =
1 2 log x + 1 + 12
(D – 2) (D + 1)  x x
1 e–x ⌠  1 1 
 ⌡
= ex 2 log x + + dx
D–2  x x2 
1 e–x ⌠  2 1 1 
 ⌡
= ex 2 log x + – +  dx
D–2  x x x2 
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.12) LINEAR DIFFERENTIAL EQUATIONS…
…

 –x ⌠ x  
e ⌡ e 2 log x –  +  + 2 dx
1 1 2 1
=
D–2   x   x x  
=
1 
e–x · ex 2 log x –
1  = 1 2 log x – 1
D–2  x  D – 2  x
–2x  1
= e2x ⌠
⌡ e 2 log x – x  dx
 1 
= e2x ⌠
⌡ 2 log x · e dx – ⌠
–2x
⌡ e
–2x
dx
 x 
 e–2x 2 e–2x 
= e2x 2 log x   – ⌠
1
⌡ · dx – ⌠
⌡ e–2x · dx
  –2  x  –2  x 
 1 1 
⌠ e–2x
= e2x – log x · e–2x + ⌡
x
⌠ e–2x x dx
dx – ⌡
 
= e2x {– log x e–2x} = – log x
∴ y = C.F. + P.I. = yc + yp
y = c1 e2x + c2 e–x – log x
Ex. 6 : Solve (D2 – 1) y = e–x sin e–x + cos e–x.
Sol. : AE : D2 – 1 = 0 or (D – 1) (D + 1) = 0 ⇒ D = – 1, + 1
∴ yc = c1 ex + c2 e–x
1
yp = (e–x sin e–x + cos e–x)
(D – 1) (D + 1)

1  –x ⌠   
= e ⌡ ex (cos e–x + e–x sin e–x) dx Use ⌠ x x
⌡ e [f + f'] dx = e ·f
D–1    
1 1
= {e–x · ex cos e–x} = · cos e–x
D–1 D–1

= ex ⌠
⌡ e cos e dx = – e ⌠
–x –x x –x –x
⌡ cos e (– e dx) {Use e–x = t}
= – ex sin e–x
∴ y = c1 ex + c2 e–x – ex sin e–x .
Ex. 7 : Solve (D2 – 1) y = (1 + e–x)–2.
Sol. : A.E. : D2 – 1 = 0
C.F. = c1 ex + c2 e–x
1
P.I. = (1 + e–x)–2
(D + 1) (D – 1)
1
=
D+1
ex ⌠ –x –x –2
⌡ e (1 + e ) dx
1 –1
= (– ex) ⌠ –x –2 –x x –x –1
⌡ (1 + e ) (– e dx) = D + 1 [– e (1 + e ) ]
D+1
ex · ex
= e–x ⌠
⌡ 1 + e–x
dx

e2x · (ex dx)

= e–x ⌠
⌡ 1 + ex
(1 + ex = t)

(t – 1)2 2
t – 2t + log t
= e–x ⌠
⌡ t
dt = e–x
2 
(1 + e )
x 2
= e–x – 2 (1 + ex) + log (1 + ex)

 2 
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.13) LINEAR DIFFERENTIAL EQUATIONS…
…

e–x
∴ y = c1 ex + c2 e–x + (1 + ex)2 + e–x log (1 + ex) – 2e–x – 2
2
x 2
or y = A ex + B e–x + e–x (1 + e ) + log (1 + ex) – 2
 2 
x
Ex. 8 : Solve (D2 + 3D + 2) y = e + cos ex .
e
(Dec. 2007)
2
Sol. : A.E. : D + 3D + 2 = (D + 2) (D + 1) = 0
C.F. = c1 e–2x + c2 e–x
1 x
P.I. = (ee + cos ex)
(D + 2) (D + 1)
1 x
= e–x ⌠ x e x
⌡ e (e + cos e ) dx
D+2
1 x
= e–x (ee + sin ex)
D+2
x
= e–2x ⌠ 2x –x e x
⌡ e e (e + sin e ) dx
x
= e–2x ⌠ x e x
⌡ e (e + sin e ) dx
x
= e–2x (ee – cos ex)
x
∴ y = c1 e–2x + c2 e–x + e–2x (ee – cos ex)
Ex. 9 : Solve (D2 + 3D + 2) y = sin ex. (May 2012, Dec. 2012)
Sol. : A.E. : D2 + 3D + 2 = (D + 2) (D + 1) = 0 ⇒ D = – 2, – 1.
C.F. = c1 e–2x + c2 e–x
1 1
P.I. = sin ex = e–x ⌠
⌡ ex sin ex dx
(D + 2) (D + 1) D+2
1
= e–x (– cos ex) = – e–2x ⌠
⌡ e cos e dx
x x
D+2
= – e–2x sin ex
∴ y = c1 e–2x + c2 e–x – e–2x sin ex
d2y
Ex. 10 : Solve + y = cosec x.
dx2
Sol. : A.E. : D2 + 1 = (D + i) (D – i) = 0 ⇒ D = ± i.
C.F. = c1 cos x + c2 sin x
1 1  1 1 
P.I. = 2 cosec x = – cosec x
D +1 2i D – i D + i
1  1
cosec x
1
= cosec x –
2i D – i D+i 
1  ix 
= e ⌠
⌡ e–ix cosec x dx – e–ix ⌠
⌡ e cosec x dx
ix
2i  
1  ix 
= e ⌠
⌡ ⌠ (cos x + i sin x) cosec x dx
(cos x – i sin x) cosec x dx – e–ix ⌡
2i  
1  ix 
e ⌠ (cot x – i) dx – e–ix ⌠
2i  ⌡ ⌡ (cot x + i) dx
=

1
= [eix (log sin x – ix) – e–ix (log sin x + ix)]
2i
1
= [log sin x) (eix – e–ix) – ix (eix + e–ix)]
2i
= sin x log sin x – x cos x
∴ y = c1 cos x + c2 sin x + sin x log sin x – x cos x
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.14) LINEAR DIFFERENTIAL EQUATIONS…
…

(B) Short-cut Methods for Finding P.I. in Certain Standard Cases

Although the general method (A) discussed in the previous article will always work in the theory, it many a times leads to
laborious and difficult integration. To avoid this, short methods of finding P.I. without actual integration are developed
depending upon the particular form of function f(x).
Case I : P.I. when f(x) = eax, a is any constant.
1
To obtain yp = eax, we have D eax = a eax, D2 eax = a2 eax …… Dn ean = an eax
φ(D)
∴ (a0 Dn + a1 Dn–1 + …… + an) eax = (a0 an + a1 an–1 + …… + an) eax
or φ (D) eax = φ (a) eax
1
Operating on both sides by , we have
φ (D)
1 1
[φ (D) eax] = [φ (a) eax]
φ (D) φ (D)
1 1
or eax = φ (a) (eax) , (‡ is a linear operator)
φ (D) φ (D)
Dividing by φ (a), we have the formula
1 1
eax = eax provided φ (a) ≠ 0 … (A)
φ (D) φ (a)

Case of Failure : If φ (a) = 0, above rule fails and we proceed as under.

Since φ (a) = 0, D – a must be a factor of φ (D) (by Factor Theorem).

Let φ (D) = (D – a) ψ (D) where, ψ (a) ≠ 0. Then

1 1 1
(eax) = eax
φ (D) D – a ψ (D)
1 eax
= … from (A)
D – a ψ (a)
1 1
= eax
ψ (a) D – a
1
=
ψ (a)
⌠ e–ax eax dx
eax ⌡ … (refer 1.11-A (i))

1
= eax ⌠
⌡ dx
ψ (a)

1
= x· eax, where ψ (a) = φ'(a) ≠ 0.
ψ (a)

1 1
i.e. eax = x · eax provided φ'(a) ≠ 0 … (B)
φ (D) φ' (a)

If φ' (a) = 0 then we shall apply (B) again to get

1 1
(eax) = x2 eax, provided φ"(a) ≠ 0 , and so on.
φ (D) φ" (a)

Remark 1 : Since φ (D) = (D – a) ψ (D)

φ'(D) = (D – a) ψ'(D) + ψ(D)

∴ φ'(a) = 0 + ψ (a)

or φ'(a) = ψ (a)
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.15) LINEAR DIFFERENTIAL EQUATIONS…
…

Remark 2 : It can also be established that

1 1 xr ax
r eax = e , provided ψ(a) ≠ 0.
(D – a) ψ(D) ψ(a) r!

Remark 3 : Any constant k can be expressed as k = k · e0x

1 1 1
∴ yp = (k) = k · e0x = k · e0x
φ(D) φ(D) φ(D)
1
= k· , φ(0) ≠ 0
φ(0)
Remark 4 : If f(x) = ax then we use ax = ex log a
1 1
∴ yp = ax = ex log a
φ(D) φ(D)
1
= ax Replace D with log a.
φ (log a)
–x –x x log 1/a x (–log a)
If f(x) = a then we use a = e =e
1 1
∴ yp = a–x = ex (–log a)
φ(D) φ(D)
1
= a–x. Replace D with – log a.
φ(– log a)
Formulae for Ready Reference :
1 1 x2 ax 1 x3 ax 1 xr ax
1. eax = x · eax 2. 2 e
ax
= e 3. 3 e
ax
= e 4. r e
ax
= e
D–a (D – a) 2! (D – a) 3! (D – a) r!
1 1 1 1 xr ax
5. r eax = r e
ax
= e , ψ(a) ≠ 0
(D – a) ψ (D) ψ(a) (D – a) ψ(a) r!

ILLUSTRATIONS

Ex. 1 : Find the Particular Integral of (D2 – 5D + 6) y = 3 e5x.

3 3 e5x e5x
Sol. : P.I. = 2 (e5x) = 2 =
D – 5D + 6 5 – 5.5 + 6 2

d2y dy
Ex. 2 : Find the Particular Integral of +4 + 3y = e–3x
dx2 dx
1
Sol. : Here P.I. = (e–3x) , φ (D) = D2 + 4D + 3
D2 + 4D + 3

But φ (– 3) = 9 – 12 + 3 = 0 hence φ (–3) = 0 and case I fails.

x e–3x 1
∴ P.I. = = x· · e–3x, D → a = – 3
φ' (a) 2D + 4
x e–3x x e–3x
= =
2 (–3) + 4 –2

3
Ex. 3 : Find the Particular Integral of (D – 1)3 y = ex + 2x – .
2
1 1 3 1
Sol. : yp = ex + 2x – e0x
(D – 1)3 (D – 1)3 2 (D – 1)3
x3 x 1 3 1
= e + (log 2 – 1)3 2x –
3! 2 (0 – 1)3

x3 x 1 3
= e + 2x +
6 (log 2 – 1)3 2
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.16) LINEAR DIFFERENTIAL EQUATIONS…
…

Ex. 4 : Find the Particular Integral of (D – 2)2 (D + 1) y = e2x + 2–x

1 1
Sol. : yp = e2x + 2–x
(D – 2)2 (D + 1) (D – 2)2 (D + 1)
1 1 2–x
= 2 · e2x +
(D – 2) (2 + 1) (– log 2 – 2)2 (– log 2 + 1)
1 x2 2x 2–x
= · e +
3 2! (– log 2 – 2)2 (– log 2 + 1)
Case II : P.I. when f(x) = sin (ax + b) or cos (ax + b).
1 1
To obtain yp = sin (ax + b) or cos (ax + b), we have
φ(D2) φ(D2)
D sin (ax + b) = a cos (ax + b)
D2 sin (ax + b) = – a2 sin (ax + b)
D3 sin (ax + b) = – a3 cos (ax + b)
D4 sin (ax + b) = a4 sin (ax + b)
or (D2)2 sin (ax + b) = (–a2)2 sin (ax + b)
Similarly (D2)p sin (ax + b) = (– a2)p sin (ax + b)
and we may generalise that
φ (D2) sin (ax + b) = φ (– a2) sin (ax + b)
1
Operating on both sides by , we have
φ (D2)
1 1
[φ (D2) sin (ax + b)] = [φ (–a2) sin (ax + b)]
φ (D2) φ (D2)
1
sin (ax + b) = φ (–a2) sin (ax + b)
φ (D2)
Dividing now by φ (–a2), we have
1 1
sin (ax + b) = sin (ax + b), provided φ (–a2) ≠ 0
φ (D2) φ (–a2)

Case of Failure : But if φ (–a2) = 0, above rule fails and we proceed as under :
We know by Euler's Theorem that cos (ax + b) + i sin (ax + b) = ei (ax + b) hence
1 1
sin (ax + b) = Imag. Part of ei (ax + b)
φ (D2) φ (D2)
1
= I.P. of ei (ax + b)
φ (D2)
1
= I.P. of x ei (ax + b) , (D2 = – a2)
φ'(D2)

1 1
Hence sin (ax + b) = x sin (ax + b) provided φ' (–a2) ≠ 0
φ (D2) φ' (–a2)

Add if φ' (– a2) ≠ 0, we have

1 1
sin (ax + b) = x2 sin (ax + b), provided φ"(–a2) ≠ 0
φ(D2)
φ"(–a )2

Similarly formulae for cos (ax + b) viz.

1 1
cos (ax + b) = cos (ax + b), provided φ (–a2) ≠ 0
φ (D2) φ (–a2)
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.17) LINEAR DIFFERENTIAL EQUATIONS…
…

But if φ (–a2) = 0, we have

1 1
cos (ax + b) = x cos (ax + b), provided φ' (–a2 ) ≠ 0
φ (D2) φ' (–a2)

And if φ' (–a2) = 0, we have

1 1
cos (ax + b) = x2 cos (ax + b), provided φ" (–a2) ≠ 0
φ (D2) φ" (–a2)

and so on and so forth.

Additional Results :
1 1
sin ax = sin ax , φ(–a2) ≠ 0 (Replace D2 with – a2)
φ(D2) φ(–a2)

1 1
cos ax = cos ax , φ(– a2) ≠ 0, (Replace D2 with – a2)
φ(D2) φ(–a2)

For the case of failure, it can also be established that

1 x
sin (ax + b) = – cos (ax + b)
D2 + a2 2a

1 x
cos (ax + b) = sin (ax + b)
D2 + a2 2a

1
sin (ax + b) =
 –x r 1 
sin ax + b +
rπ
(D2 + a2)r 2a r!  2

r
1
cos (ax + b) = – x 1
cos ax + b + 
rπ
(D + a2)r
2
 2a  r!  2

Useful Formulae :
1 – cos 2x eox cos 2x
sin2 x = = –
2 2 2

1 + cos 2x eox cos 2x

cos2 x = = +
2 2 2

1
sin A sin B = [cos (A – B) – cos (A + B)]
2

1
sin A cos B = [sin (A + B) + sin (A – B)]
2

1
cos A cos B = [cos (A + B) + cos (A – B)]
2

x x
sin x = 2 sin cos
2 2

sin 2x = 2 sin x cos x.

Note : Write D3 = D2 · D ; D4 = (D2)2 ; D5 = (D2)2 · D. Always replace D2 by – a2 and keep D as it is. To get D2 in the denominator,
rationalise the denominator and then replace D2 by – a2. Now numerator will contain an operator in D, therefore open the
bracket.
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.18) LINEAR DIFFERENTIAL EQUATIONS…
…

ILLUSTRATIONS

Ex. 1 : Solve (D2 + 2D + 1) y = 4 sin 2x.

Sol. : A.E. is D2 + 2D + 1 = 0 ⇒ D = – 1, – 1.

C.F. = (c1 x + c2) e–x

1
P.I. = (4 sin 2x)
D2 + 2D + 1

1
= (4 sin 2x) (putting D2 = – 22 = – 4)
– 4 + 2D + 1

4
= (sin 2x)
2D–3
4 (2D + 3)
= (sin 2x) [(Multiply numerator and denominator by (2D + 3)]
4 D2 – 9
4 (2D + 3)
= (sin 2x) (replace D2 with – 4)
4 (– 4) – 9
4
= – (2D + 3) (sin 2x)
25
4
= – [4 cos 2x + 3 sin 2x]
25
∴ General solution is
4
y = (c1 x + c2 ) e–x – [4 cos 2x + 3 sin 2x]
25
d3y dy
Ex. 2 : Solve + 4 = sin 2x. (Dec. 2010)
dx3 dx
Sol. : A.E. will be D3 + 4D = 0 ⇒ D (D2 + 4) = 0. Hence D = 0 and D = ± 2 i.
C.F. = Complementary Function = c1 + c2 cos 2x + c3 sin 2x
1
P.I. = (sin 2x) [‡ D2 + 4 = 0, for D2 = – 22 = – 4 ]
D (D2 + 4)

= x⋅
1
(sin 2x),  d (D3 + 4D) = 3D2 + 4‚ then put D2 = – 4
3 D2 + 4 dD 
1
= x· (sin 2x)
3 (–4) + 4
x
= – sin 2x.
8
x sin 2x
Hence the solution is y = c1 + c2 cos 2x + c3 sin 2x –
8

Ex. 3 : Solve (D2 + 1) y = sin x sin 2x. (Dec. 2008)

Sol. : A.E. is D2 + 1 = 0 ⇒ D = ± i

∴ C.F. = c1 cos x + c2 sin x

1
We have P.I. = (sin x sin 2x)
D2 + 1

=
1 1 
(cos x – cos 3x)
D2 + 1 2 
1 1 1 1
= cos x – cos 3x
2 D2 + 1 2 D2 + 1
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.19) LINEAR DIFFERENTIAL EQUATIONS…
…

(D2 → – 9 in 2nd term, case fails for 1st term)

1 1 1 1
= x· cos x – cos 3x
2 2D 2 –9+1
1 D 1
= x· cos x + cos 3x, (D2 → –1)
4 D2 16
1 D (cos x) 1
= x· + cos 3x
4 –1 16
1 1
= x sin x + cos 3x
4 16
1 1
Hence the solution is y = c1 cos x + c2 sin x + x sin x + cos 3x
4 16

Case III : P.I when f(x) = cosh (ax + b) or sinh (ax + b).
1 1
To find yp = cosh (ax + b) or sinh (ax + b)
φ(D2) φ(D2)
As earlier on the similar line, we can prove that
1 1
cosh (ax + b) = cosh (ax + b), φ(a2) ≠ 0
φ (D2) φ (a2)
1 1
and sinh (ax + b) = sinh (ax + b), φ(a2) ≠ 0
φ (D2) φ (a2)

ILLUSTRATION
3
dy dy
Ex. 1 : Solve –4 = 2 cosh 2x
dx3 dx
Sol. : A.E. is D3 – 4D = 0 ⇒ D (D2 – 4) = 0, D = 0, ± 2
Hence C.F. = c1 + c2 e2x + c3 e–2x

P.I. =
1  1 (2 cosh 2x)
(D2 – 4) D 
1 ⌠
=
D2 – 4 ⌡ 2 cosh 2x dx

=
2 sinh 2x
D2 – 4  2 
1
= (sinh 2x) [case of failure, hence differentiate φ (D)]
D2 – 4
x (sinh 2x) x D (sinh 2x)
= =
2D 2D2
x D (sinh 2x) x
= = D (sinh 2x)
2 (4) 8
x
= cosh 2x
4
x
∴ Solution is y = c1 + c2 e2x + c3 e–2x + cosh 2x
4

Case IV : P.I. when f(x) = xm

1 m 1
To find yp = x , we write (xm) = [φ (D)]–1 xm.
φ(D) φ (D)
We shall now expand [φ (D)]–1 in ascending powers of D as far as the term in Dm and operate on xm term-by-term. Since (m + 1)th
and higher derivatives of xm will be zero, we need not consider terms beyond Dm.
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.20) LINEAR DIFFERENTIAL EQUATIONS…
…

Important Formulae :
1
= (1 + x)–1 = 1 – x + x2 – x3 + ……
1+x
1
= (1 – x)–1 = 1 + x + x2 + x3 + ……
1–x
n(n – 1) 2
(1 + x)n = 1 + nx + x + ………
2!
Also note that Dn (xn) = n! and Dn+1 (xn) = 0.
1
Note : To find yp = xm
φ (D)
(i) We always take constant term common from the denominator and use the formulae (1 + x)–1, (1 – x)–1, (1 + x)n, (1 – x)n.
(ii) If constant term is absent in the denominator then the minimum power of D is taken common from the denominator.
1 1
for example. xm = xm
D2 – 3D – 2  D2 – 3D
–2 1–
  2 
1 1
xm = xm
D2 – 3D + 3  D 2
– 3D
3 1+
  3 
1 1
xm = xm
D3 – 3D2 + 2D  D2 – 3D
2D 1 +
  2 
ILLUSTRATION
2
d y dy
Ex. 1 : Find the particular solution of – + y = x3 – 3x2 + 1.
dx2 dx
Sol. : It can be put as (D2 – D + 1) y = x3 – 3x2 + 1.
1
P.I. = (x3 – 3x2 + 1)
(1 – D + D2)
= [1 – (D – D2)]–1 (x3 – 3x2 + 1)
Expanding by Binomial theorem upto D3 terms
= [1 + (D – D2 ) + (D – D2)2 + (D – D2)3 + …] (x3 – 3x2 + 1)
= [1 + D – D2 + D2 – 2D3 + … + D3 + …] (x3 – 3x2 + 1)
= (1 + D – D3) (x3 – 3x2 + 1)
= x3 – 6x – 5
Hence P.I. = x3 – 6x – 5.
Case V : P.I. when f(x) = eax V, where V is any function of x.
1 ax
To find yp = e V, we have
f(D)

D (eax V) = eax DV + aeax V = eax (D + a) V

and D2 (eax V) = eax D2 V + 2a eax DV + a2 eax V
= eax (D + a)2 V
and proceeding similarly, we may have in general

Dn (eax V) = eax (D + a)n V

Hence φ (D) (eax V) = eax φ (D + a) V … (I)

1
Now, let φ (D + a) V = V1 ⇒ V = V
φ (D + a) 1
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.21) LINEAR DIFFERENTIAL EQUATIONS…
…

If we put value of V in (I), we have

φ (D) eax
 1 
V = eax V1
 φ (D + a) 1
1
Operating on both sides by now, we get
φ (D)
1 1
eax V = (eax V1)
φ (D + a) 1 φ (D)
Here V1 is any function of x, and hence, we have the formula
1 1
(eax V) = eax (V)
φ (D) φ (D + a)

ILLUSTRATION

Ex. 1 : Solve (D2 – 4D + 3) y = x3 e2x.

Sol. : A.E. = D2 – 4D + 3 = (D – 1) (D – 3) ⇒ D = 1, 3.
Hence C.F. = c1 ex + c2 e3x
1
P.I. = (x3 e2x)
D2 – 4 D + 3
1
= e2x (x3) … (D → D + 2)
(D + 2) 2 – 4 (D + 2) + 3
1
= e2x (x3)
D2 + 4 D + 4 – 4 D – 8 + 3
1
= e2x (x3) = – e2x (1 – D2)–1 (x3) (by case IV)
D2 – 1
= – e2x [1 + D2 + D4 + ……] (x3) = – e2x [x3 + 6x]
Hence solution is y = c1 ex + c2 e3x – e2x (x3 + 6x)
Case VI : P.I. when f(x) = xm sin ax‚ or xm cos ax.
1 m 1 m
To find yp = x sin ax or x cos ax, we have
f(D) f(D)
1 1
xm [cos ax + i sin ax] = xm eiax
φ (D) φ (D)
1
= eiax xm
φ (D + ia)
1
Now xm, can be evaluated by method of case IV and equating the Real and Imaginary parts, we get the required
φ (D + ia)
results.

ILLUSTRATION

Ex. 1 : Solve (D4 + 2 D2 + 1) y = x2 cos x.

Sol. : A.E. is (D2 + 1)2 = 0 ⇒ D = ± i, ± i.
∴ C.F. = (c1 x + c2) cos x + (c3 x + c4) sin x
For Particular Integral, we have
1 1
[x2 (cos x + i sin x)] = x2 · eix
(D2 + 1)2 (D2 + 1)2
1 1
= eix (x2) = eix 2 (x2)
[(D + i)2 + 1]2 (D + 2iD)2

= eix
1
(x2)
.. . 1
=–i

2 1 – iD2  i 
–4D
 2
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.22) LINEAR DIFFERENTIAL EQUATIONS…
…

–2
eix 1 1 – iD
= – (x2)
4 D2  2
eix 1 1 + iD – 3  (x2)
= – D2 + …
4 D2  4 
eix 1 x2 + 2ix – 3
= –
4 D2  2
eix  x + ix
4 3
3 2
= – – x Integrating twice
4 12 3 4 
4 3
= –
1
[cos x + i sin x]  x + ix –
3 2
x
4 12 3 4 

Equating the real parts on both sides,

4
1
(x2 cos x) = –
1  x – 3 x2 cos x +
1 3
x sin x
(D + 1)2
2
4 12 4  12

Hence the general solution is

x3 sin x (x4 – 9x2)
y = (c1 x + c2) cos x + (c3 x + c4) sin x + – cos x
12 48

Case VII : P.I. when f(x) = x V, V being any function of x.

1
To find xV we have by successive differentiations
f(D)

D (xV) = x DV + V
D2 (xV) = x D2V + 2DV
D3 (xV) = x D3V + 3D2V
and so on, we may have
Dn (xV) = x Dn V + n Dn–1 (V)
d
Or Dn (xV) = x Dn V + (Dn) V … (A)
dD
d
Since φ(D) is a polynomial in D, we may write in general from (A) using φ'(D) = φ (D)
dD
φ (D) (xV) = x φ (D) V + φ' (D) V … (B)
1
Now, put φ (D) V = V1 so that V = V in equation (B), we have
φ (D) 1


φ (D) x
1
V
 = x V + φ' (D) 1 V
 φ (D) 1 1
φ (D) 1
1
Operating on both sides by , we get
φ (D)
1 1 1 1
x· V = [xV1] + φ' (D) V
φ (D) 1 φ (D) φ (D) φ (D) 1

and if we adjust the terms on both sides, we get

1
[xV1] =
x – 1 φ' (D) 1
V
φ (D)  φ (D)  φ (D) 1

But here V1 is any function of x, hence we have the formula

1
[xV] = x – 1 φ' (D) 1 V
φ (D)  φ (D)  φ (D)
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.23) LINEAR DIFFERENTIAL EQUATIONS…
…

Remark :
1. The rule xV is applied if
(i) power of x is one
1
(ii) V is not a case of failure.
φ(D)
1
2. If power of x is one and V is a case of failure then do not apply xV rule. In this case, apply rule given by case (VI).
φ(D)
1
For example, yp = 2 x sin x
D +1
1
Here 2 sin x is a case of failure. Therefore use case (VI) method.
D +1

ILLUSTRATIONS
2
dy
Ex. 1 : Solve + 4y = x sin x.
dx2
Sol. : A.E. : D2 + 4 = 0 ⇒ D = ± 2i
∴ C.F. = c1 cos 2x + c2 sin 2x,
1
and P.I. = (x sin x)
D2 + 4

= x – 22D  2 1 (sin x) [by case (VII)]

 D + 4 D + 4
=
x – 22D  1 (sin x)
 D + 4 – 1 + 4
1  2D  1  2D 
= x– 2 sin x = x sin x – 2 sin x
3  D + 4 3  D +4 
1  2D (sin x) 1  2 
= x sin x – = x sin x – (cos x)
3  –1+4  3  3 
1 2
= x sin x – cos x.
3 9
Hence the complete solution is
x sin x 2
y = c1 cos 2x + c2 sin 2x + – cos x
3 9
Ex. 2 : Solve (D2 – 2D + 1) y = x ex sin x (Dec. 2004)
2
Sol. : A.E. : D – 2D + 1 = 0
⇒ (D – 1)2 = 0, D = 1, 1.
Hence C.F. = (c1 x + c2) ex
1
P.I. = [x ex sin x]
(D – 1)2
1
= ex (x sin x) (by case V)
(D + 1 – 1)2
1
= ex (x sin x)
2

= ex x –
2D 1
(sin x) (by case VII)
 D2  D2
=
 2  (– sin x) = – ex x sin x – 2 sin x
ex x –
 D  D 
= – ex [x sin x + 2 cos x]
Hence the complete solution is
y = (c1 x + c2) ex – ex [x sin x + 2 cos x]
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.24) LINEAR DIFFERENTIAL EQUATIONS…
…

Now, we will summarise the short-cut methods of P.I. and the corresponding formulae :
1 eax
Case I : eax = , φ(a) ≠ 0
φ(D) φ(a)
1 1
Case of failure : If φ(a) = 0, eax = x · eax, φ'(a) ≠ 0
φ(D) φ'(a)
1 ax xr ax 1 1
r e = e ; (k) = k · , φ(0) ≠ 0
(D – a) r! φ(D) φ(0)
1 ax
ax =
φ(D) φ(log a)
1 1
Case II : sin (ax + b) = sin (ax + b), φ(–a2) ≠ 0
φ(D2) φ(– a2)
1 1
cos (ax + b) = cos (ax + b), φ (–a2) ≠ 0
φ(D2) φ (–a2)
1 1
Case of failure : If φ(a2) = 0, sin (ax + b) = x · sin (ax + b), φ'(–a)2 ≠ 0
φ(D2) φ'(–a)2
1 1
If φ(a2) = 0, cos (ax + b) = x · cos (ax + b), φ'(–a2) ≠ 0
φ(D2 ) φ'(–a2)
Case of failure formulae :
1 x 1 x
sin ax = – cos ax; cos ax = sin ax
D2 + a2 2a D2 + a2 2a
r
1
sin (ax + b) =
– x  1
sin ax + b + r  and
π
(D + a2)r
2
 2a r!  2
r
1
cos (ax + b) = – x  1
cos ax + b + r 
π
(D + a2)r
2
 2a r!  2
1 1
Case III : sinh ax = sinh ax, φ(a2) ≠ 0 and
φ(D2) φ(a2)
1 1
cosh ax = cosh ax, φ(a2) ≠ 0
φ(D2) φ(a2)
1
Case IV : xm = [φ(D)]–1 xm, expand by using Binomial theorem.
φ(D)
1 1
Case V : eax V = eax V
φ(D) φ (D + a)
1 1 1
Case VI : xm sin ax = I.P. of xm eiax = I.P. of eiax φ(D + ia) xm
φ(D) φ(D)
1 1 1
xm cos ax = R.P. of xm eiax = R.P. of eiax φ(D + ia) xm
φ(D) φ(D)

Case VII :
1
xV =
x – φ'(D) 1
V
φ(D)  φ(D)  φ(D)

ILLUSTRATIONS ON SHORT-CUT METHODS

Ex. 1 : Solve (D2 + 2D + 1) y = 2 cos x + 3x + 2 + 3ex.

Sol. : Here AE is (D + 1)2 = 0 ⇒ D = – 1, – 1
∴ C.F. = (c1 x + c2) e–x
1 1 1
and P.I. = 2 cos x + (3x + 2) + 3 ex
D2 + 2D + 1 [1 + (2D + D2)] (D + 1)2
1 3ex
= 2 cos x + [1 + (2D + D2)]–1 (3x + 2) +
–1 + 2D + 1 4
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.25) LINEAR DIFFERENTIAL EQUATIONS…
…

⌠ cos x dx + [1 – 2D – D2 + …] (3x + 2) +
3 x
= ⌡ 4
e

3ex
= sin x + 3x + 2 – 6 +
4
3ex
= + sin x + 3x – 4
4
Hence the complete solution is
3ex
y = (c1 x + c2) e–x + + sin x + 3x – 4
4
d2y 2 a2R dy
Ex. 2 : Solve 2 + a y = (l – x) where a, R, p and l are constants, subject to the conditions y = 0, = 0 at x = 0.
dx p dx
Sol. : Given equation is
a2R
(D2 + a2 ) y = (l – x)
p
A.E. = D2 + a2 = 0 or D = ± ia
C.F. = c1 cos ax + c2 sin ax
1 a2 R a2 R 1 1
P.I. = 22 (l – x) = · (l – x)
D +a p p a2 1 + D2 
2

 a
2 –1
1 + D2  1 – D2 
2
R R R
= (l – x) = (l – x) = (l – x)
p  a p  a p
Hence the general solution is
R
y = c1 cos ax + c2 sin ax + (l – x) … (1)
p
For initial conditions, now put y = 0 when x = 0 in (1), we get
R – Rl
0 = c1 + l ⇒ c1 =
p p
If we differentiate equation (1),
dy R
= – ac1 sin ax + ac2 cos ax –
dx p
dy
Putting x = 0 and = 0 in this, we get
dx
R R
0 = ac2 – , hence c2 =
p ap
Now put values of c1 and c2 in A, then the required particular solution is
R sin ax 
y = – l cos ax + l – x
p  a 
Ex. 3 : Solve (D3 – 1) y = (1 + ex)2.
1 3
Sol. : A.E. : D3 – 1 = 0 or (D – 1) (D2 + D + 1) = 0 ∴ D = 1, – ±i
2 2
yc = c1 ex + e(–1/2) x [c2 cos ( 3/2) x + c3 sin ( 3/2) x]
1 1
yp = 3 (1 + ex)2 = 3 (1 + 2ex + e2x)
D –1 D –1
1 1 1
= 3 eox + 2 3 ex + 3 e2x
D –1 D –1 D –1
2 1
= – 1 + xex + e2x
3 7
2 x 1 2x
∴ y = c1ex + e(–1/2) x [c2 cos ( 3/2) x + c3 sin ( 3/2) x] – 1 + xe + e
3 7
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.26) LINEAR DIFFERENTIAL EQUATIONS…
…

x
Ex. 4 : Solve (D – 1)2 (D2 + 1)2 y = sin2 .
2
Sol. : A.E. : (D – 1)2 (D2 + 1)2 = 0, D = 1, 1, ± i, ± i.
yc = (c1x + c2) ex + (c3x + c4) cos x + (c5x + c6) sin x.

yp =
1 x
sin2 =
1 1 – cos x
(D – 1)2 (D2 + 1)2 2 (D – 1)2 (D2 + 1)2  2 
1 1 1 
= eox – cos x
2 (D – 1)2 (D2 + 1)2 (D – 1)2 (D2 + 1)2 
1 1  1 1 + 1 2 1 2 sin x
= 1– 2 cos x =
2 (D + 1)2 (– 1 – 2D + 1)  2  2 (D + 1) 
1  d2 
sin x  2 (D2 + 1)2 =
1 1 d
= 1 + x2 2 (D2 + 1) 2D = 4 (3D2 + 1), then put D2 = –1
2 2 –8  dD dD 
1 1 2
= – x sin x
2 32
1 1 2
∴ y = (c1x + c2) ex + (c3x + c4) cos x + (c5x + c6) sin x + – x sin x
2 32
Ex. 5 : Solve (D4 – 2D3 – 3D2 + 4D + 4) y = x2 ex.
Sol. : A.E. : (D – 2)2 (D + 1)2 = 0
yc = (c1 x + c2) e2x + (c3 x + c4) e–x
1 1
yp = ex · x2 = ex x2
(D2 – D – 2)2 [(D + 1)2 – (D + 1) – 2]2
1 ex 1
= ex 2
2
2 x = 2 x2
(D + D – 2) 4
1 – D + D
2

  2 
ex 1 + (D2 + D) + 3 (D2 + D)2 + …  x2
=
4  4 
e 
x
7  e  2
x
7
= 1 + D + D2 + … x2 = x + 2x +
4  4  4  2
x
y = (c1 x + c2) e2x + (c3 x + c4) e–x +
e x2 + 2x + 7 .
4  2
∴

Ex. 6 : Solve (D4 – 1) y = cos x cosh x (Dec. 2006)

Sol. : A.E. : (D – 1) (D + 1) (D + i) (D – i) = 0
yc = c1 ex + c2 e–x + c3 cos x + c4 sin x.
ex + e–x
cos x 
1
yp = 4
D –1  2 
1 1 1 1
= ex cos x + · 4 e–x cos x
2 D4 – 1 2 D –1
ex 1 e–x 1
= 4 cos x + cos x
2 (D + 1) – 1 2 (D – 1)4 – 1
ex 1 e–x 1
= · 4 3 2 cos x + cos x
2 D + 4D + 6D + 4D + 1 – 1 2 D4 – 4D3 + 6D2 – 4D + 1 – 1
ex 1 e–x 1
= cos x +
2 1 – 4D – 6 + 4D 2 1 + 4D – 6 – 4D
ex cos x e–x cos x cos x
= · + = cosh x
2 –5 2 –5 –5
cos x cosh x
∴ y = c1 ex + c2 e–x + c3 cos x + c4 sin x –
5
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.27) LINEAR DIFFERENTIAL EQUATIONS…
…

Ex. 7 : Solve (D4 + 1) y = 2 sinh x sin x. (Dec. 2004)

4 4 2 2 2 2 2
Sol. : A.E. : D + 1 = 0 ∴ D + 2D + 1 – 2D = 0 or (D + 1) – ( 2 D) = 0
2 2
or (D – 2 D + 1) (D + 2 D + 1) = 0
1 1 –1 1
∴ D = ± i, D = ± i
2 2 2 2
x/ 2 c cos x x + e– x/ 2 c cos x x 
yc = e 1 + c2 sin  3 + c4 sin 
 2 2  2 2
1
yp = 2 sinh x sin x
D4 + 1
1
= (ex – e–x) sin x
D4 + 1
1 1
= ex sin x – 4 e–x sin x
D4 + 1 D +1
1 1
= ex sin x – e–x sin x
(D + 1)4 + 1 (D – 1)4 + 1
1 1
= ex · sin x – e–x 4 sin x
D4 + 4D3 + 6D2 + 4D + 2 D – 4D3 + 6D2 – 4D + 2
1 1
= ex sin x – e–x sin x
(–1)2 + 4D (–1) + 6 (–1) + 4D + 2 (–1)2 – 4D (–1) + 6 (–1) – 4D + 2

= ex
sin x sin x
– e–x
–3 –3
e – e–x
x
sin x · 
–2 2
= = – sin x sinh x
3  2  3
∴ y = C.F. + P.I.
x/ 2 c cos x x  + e– x/ 2 c cos x x  2
y = e 1 + c2 sin  3 + c4 sin  – sin x sinh x
 2 2  2 2 3

d3y dy
Ex. 8 : Solve –7 – 6y = e2x (1 + x). (Dec. 2004)
dx3 dx
Sol. : Given D.E. is written as
d
(D3 – 7D – 6) y = e2x (1 + x) where D ≡
dx

A.E. : D3 – 7D – 6 = 0 ∴ (D + 1) (D + 2) (D – 3) = 0

∴ C.F. = c1 e–x + c2 e–2x + c3 e3x

1
P.I. = e2x (1 + x)
D3 – 7D – 6
1
= e2x (1 + x), by D → D + 2
(D + 2)3 – 7 (D + 2) – 6
1
= e2x (1 + x)
D3 + 6D2 + 5D – 12
–1
–e2x 1 – D
3
+ 6D2 + 5D – e2x 1 + 5D + … 
= (1 + x) = (1 + x)
12  12  12  12 
–e2x 1 + x + 5  –e2x x + 17
= =
12  12 12  12
e2x x + 17
y = C.F. + P.I. = c1e–x + c2 e–2x + c3 e3x –
 12
∴
12
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.28) LINEAR DIFFERENTIAL EQUATIONS…
…

Ex. 9 : Solve (D2 – 1) y = x sin x + (1 + x2) ex. (Dec. 2010)

2
Sol. : A.E. : D – 1 = 0 or (D – 1) (D + 1) = 0
∴ C.F. = c1 ex + c2 e–x
1 1
P.I. = x sin x + 2 ex (1 + x2)
D2 – 1 D –1

1 2D 1
= x sin x – 2 sin x + ex (1 + x2)
D2 – 1 (D – 1)2 (D + 1)2 – 1

x sin x 2D 1
= – sin x + ex 2 (1 + x2)
–2 4 D + 2D

1 – D + D
2
= –
x
sin x –
cos x
+ ex
1 
+ … (1 + x2)
2 2 2D  2 4 
x cos x ex 1 1 + x2 – x + 1
= – sin x – +
2 2 2 D  2

x cos x ex x
3
x2 3x
= – sin x – + – +
2 2 2 3 2 2

1 ex
∴ y = c1 ex + c2 e–x – (x sin x + cos x) + (2x3 – 3x2 + 9x)
2 12

Ex. 10 : (D2 – 4D + 4) y = ex cos2 x. (Dec. 2015)

2
Sol. : A.E. is D – 4D + 4 = 0 ∴ D = 2, 2
yc = (c1x + c2) e2x
1 1
yp = ex cos2 x = ex cos2 x
(D – 2)2 (D – 1)2

= ex
1 1 + cos 2x = 0
(D – 1)2  2 
ex  1 2 eox + 2 1
= cos 2x
2 (D – 1) D – 2D + 1 
ex  1  e x
1 – 2 – 3) cos 2x
(2D
= 1– cos 2x =
2  (2D + 3)  2  4D – 9 
e 
x
1  e 1 – 1 (4 sin 2x + 3 cos 2x)
x
= 1+ (2D – 3) cos 2x =
2  25  2  25 
x
y = (c1x + c2) e2x +
e 1 – (4 sin 2x + 3 cos 2x)
1
2  
∴
25
Ex. 11 : Solve (D2 + 1) y = x2 sin 2x.
Sol. : A.E. is D2 + 1 = 0 ∴ D = ± i.
yc = c1 cos x + c2 sin x
1 1
yp = x2 sin 2x = I.P. of 2 ei2x x2
D2 + 1 D +1
1 1
= I.P. of ei2x x2 = I.P. of ei2x 2 x2
(D + 2i)2 + 1 D + 4iD – 4 + 1
–1
ei2x 1 ei2x 1 – 1 (4iD + D2)
= I.P. of x2 = I.P. of x2
(–3)  1  (–3)  3 
1 – (4iD + D2)
 3 
ei2x 1 + 1 (4iD + D2) + 1 (–16D2 + 8iD3 + D4) + … x2
= I.P. of
(–3)  3 9 
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.29) LINEAR DIFFERENTIAL EQUATIONS…
…

1 + 4 iD – 13 D2 + … x2
ei2x
= I.P. of
 3
(–3) 9 
(cos 2x + i sin 2x)  2 26 8 
= I.P. of x – +i x
(–3)  9 3 
1  2 26 8
= – x – sin 2x – x cos 2x
3  9 9

c1 cos x + c2 sin x – x2 –  sin 2x – x cos 2x

1 26 8
y =
3 9
∴
9
Ex. 12 : (D2 + D + 1) y = x sin x. (May 2011)
1 3
Sol. : A.E. is D2 + D + 1 = 0 ∴ D = – ±i .
2 2
yc = e(–1/2) x [c1 cos ( 3/2) x + c2 sin ( 3/2) x]

yp =
1
x sin x =
x – 22D + 1  2 1 sin x
D2 + D + 1  D + D + 1 D + D + 1

=
x – 22D + 1  1
sin x =
x – 22D + 1  (– cos x)
 D + D + 1 D  D + D + 1
1
= – x cos x + (2D + 1) cos x
D
= – x cos x + (2D + 1) sin x
= – x cos x + 2 cos x + sin x
∴ y = e(–1/2) x [c1 cos ( 3/2) x + c2 sin ( 3/2) x] – x cos x + 2 cos x + sin x

Ex. 13 : (D2 + 2D + 1) y = x e–x cos x.

Sol. : A.E. is D2 + 2D + 1 = 0 ∴ D = –1, 1.
yc = (c1x + c2) e–x
1 1
yp = e–x x cos x = e–x 2 x cos x
(D + 1)2 D

= e–x
x – 2D2  1  2  (– cos x)
cos x = e–x x –
 D D2  D
= e–x (– x cos x + 2 sin x)
∴ y = (c1x + c2) e–x + e–x (– x cos x + 2 sin x)

Ex. 14 : Solve (D2 + 4) y = x sin2 x.

Sol. : A.E. :D2 + 4 = 0 ∴ D = ± 2i
yc = c1 cos 2x + c2 sin 2x

yp =
1
x·
1 – cos 2x = 1 · 2 1 x – 1 2 1 x cos 2x.
D2 + 4  2  2 D +4 2 D +4
= yp + y p
1 2

1 1 1 1
yp = · x= x
1 2 D2 + 4 8 D2
1+
4

1 – D
2
=
1
+ ……
 x =
x
8  4  8
1 1
yp = – · x cos 2x.
2 2 D2 + 4
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.30) LINEAR DIFFERENTIAL EQUATIONS…
…

1
Here we can not apply "xV" rule (case VII) because cos 2x is a case of failure.
D2 + 4
1 1
∴ 2 x ei2x = ei2x (D + 2i)2 + 4 x
D +4
1 ei2x  1 
= ei2x D2 + 4 i D x =
4iD  Di x
1 – 4 
 
ei2x i  iD ei2x i  i
= – 1+ x = − x+
4D  4 4D  4
ei2x i x2 ix
= – +
4  2 4
1
= – (cos 2x + i sin 2x) (– x + i 2x2)
16
Taking real parts on both sides,
1 1
x cos 2x = – (– x cos 2x – 2x2 sin 2x)
D2 + 4 16
1 1 
yp = – (x cos 2x + 2x2 sin 2x)
2 16 
∴
2
x 1
yp = – (x cos 2x + 2x2 sin 2x)
8 32
x 1
∴ y = c1 cos 2x + c2 sin 2x + – (x cos 2x + 2x2 sin 2x)
8 32

EXERCISE 1.2

Solve the following Differential Equations :

(A) On General Method
x d2y
1. (D2 + 5D + 6) y = ee . 2. + a2y = tan ax
dx2
x 1
Ans. y = c1 e–2x + c2 e–3x + (e–2x – 2e–3x) ee . Ans. y =c1 cos ax + c2 sin ax – cos ax log [sec ax + tan ax]
a2

3. (D2 – 3D + 2) y =
1 1
+ cos x (Dec. 2006, 2008, 2010) Ans. y = c1 e2x + c2 ex + e2x [e–e
–x
]
– cos (e–x)

ee–x e 
–3x
e6x e–3x
4. (D2 – 9D + 18) y = ee Ans. y = c1 e6x + c2 e3x + e
9
e3x
5. (D2 – 2D – 3) y = 3e–3x sin (e–3x) + cos (e–3x) (Nov. 2015) Ans. y = c1 e3x + c2 e–x – sin e–3x
3
(B) On Short Methods :
d2 y dy e2x
1. –7 + 6y = e2x . Ans. y = c1 e6x + c2 ex –
dx2 dx 4
d2 y 2 x xe2x
2. – 4y = (1 + ex)2 + 3 Ans. y = c1 e2x + c2 e–2x – 1 – e +
dx2 3 4
e2x x2 e–x 1
3. (D3 – 5D2 + 8D – 4) y = e2x + 2ex + 3e–x + 2 Ans. y = c1ex + (c2 + c3x) e2x + + 2xex – –
2 6 2
1
4. (D4 – 4D3 + 6D2 – 4D + 1) y = ex + 2x + . (Dec. 2005, May 2007)
3
x4 x 1 1
Ans. y = (c1x3 + c2x2 + c3x + c4) ex + e + 2x +
24 (log 2 – 1)4 3
d2 y 1 x sin 2x 1 1
5. + 4y = cos x · cos 2x · cos 3x (May 2014) Ans. y = A cos 2x + B sin 2x + + – cos 4x – cos 6x
dx2 16 16 48 128
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.31) LINEAR DIFFERENTIAL EQUATIONS…
…

6. (D5 – D4 + 2D3 – 2D2 + D – 1) y = cos x (May 2007, 2008)

1
Ans. y = c1ex + (c2 x + c3) cos x + (c4 x + c5) sin x + [(x2 + 2x) cos x – x2 sin x]
16
x
7. (D4 – m4 ) y = sin mx Ans. y = c1 emx + c2 e–mx + c3 cos mx + c4 sin mx + cos mx
4m3
x cos x
8. (D3 + D) y = cos x (Dec. 2008) Ans. c1 + c2 cos x + c3 sin x –
2
d4 y
9. cosec x + y cosec x = sin 2x
dx4
x x
2 c cos x x  + e– 2 c cos x x + 1  cos x cos 3x
Ans. y = e 1 + c2 sin  3 + c4 sin   2
–
82 
 2 2  2 2 2
2
+ 9x = 4 cos  + t, given that x = 0 at t = 0 and x = 2 at t = . cos 3t + 2 sin 3t + cos  + t
dx π π 1 1 π
10. Ans. x =
dt2 3  6 4 2 3 
d2 y dy 1 1
11. +2 + 5y = sin2 t Ans. y = e–t [A cos 2t + b sin 2t] + – [4 sin 2t + cos 2t]
dt2 dt 10 34
d2 y dy dy 1
12. +2 + 2y = sin 2x – 2 cos 2x, given that y = 0 and = 0 when x = 0. Ans. y = e–x sin x – sin 2x
dx2 dx dx 2
d2 y dy
13. + n2y = h sin px, where h, p and n are constants satisfying the condition y = 0, = b for x = 0.
dx2 dx

Ans. y = a cos nx +  –  sin nx + h sin

b ph px
n n (n2 – p2) (n2 – p2)
x
14. (D3 + 1) y = cos (2x – 1) – cos2
2

 3 3  1 1 1
Ans. y = c1 e–x + ex/2 c2 cos x + c3 sin x + [cos (2x – 1) – 8 sin (2x – 1)] – – (cos x – sin x)
 2 2  65 2 4
d2 y dy
15. –2 + 5y = 10 sin x. Ans. y = ex (A cos x + B sin x) + 2 sin x + cos x
dx2 dx

16. (D4 + 10D2 + 9) y = 96 sin 2x cos x . Given that at x = 0, y = 0, y' = –2, y" = – 8, y"' = – 18.

Ans. y = cos 3x – cos x + x (cos 3x – 3 cos x)

4 2 2
17. (D + 6D + 8) y = sin x cos 2x
x sin 2x 1 cos 4x
Ans. y = c1 cos 2x + c2 sin 2x + c3 cos 2 x + c4 sin 2 x– – –
16 32 672
cosh 5x cosh x
18. (D3 + 3D) y = cosh 2x sinh 3x. Ans. y = c1 + (c2 cos 3 x + c3 sin 3 x) +
280
+
8
x 1
19. (D3 – 25D) y = cosh 2x sinh 3x. Ans. y = c1 + c2 e5x + c3 e–5x + sinh 5x – cosh x
100 48
1
20. (D4 – 1) y = cosh x sinh x Ans. y = c1 ex + c2 e–x + c3 cos x + c4 sin x + sinh 2x
30
x –4x 1
21. (D2 + 13D + 36) y = e–4x + sinh x. Ans. y = c1e–9x + c2e–4x + e – (13 cosh x – 37 sinh x)
5 1200
22. (D3 + 1) y = sin (2x + 3) + e–x + 2x.
1 x 2x
Ans. y = c1e–x + e(1/2) x [c2 cos ( 3/2) x + c3 sin ( 3/2) x] + [sin (2x + 3) + 8 cos (2x + 3)] + e–x +
65 3 (log 2)3 + 1
d2 y dy dy
23. +6 + 10y = 50 x with y = 0, = 1 at x = 0 Ans. y = 5x – 3 + e–3x (3 cos x + 5 sin x)
dx2 dx dx
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.32) LINEAR DIFFERENTIAL EQUATIONS…
…

2
24. (D2 – 2D + 5) y = 25x2. Ans. y = ex [c1 cos 2x + c2 sin 2x] + 5x2 + 4x –
5
25. (D4 + D2 + 1) y = 53x2 + 17 (Dec. 2008)
 3 3   3 3 
Ans. y = e–x/2 c1 cos x + c2 sin x + ex/2 c3 cos x + c4 sin x + 53x2 – 89
 2 2   2 2 
1  2 9x 23
26. (D2 + 5D + 4) y = x2 + 7x + 9. (May 2006) Ans. y = c1e–4x + c2e–x + x + +
4 2 8
27. (D4 + 6D2 + 25) y = x4 + x2 + 1.

Ans. y = ex [c1 cos 2x + c2 sin 2x] + e–x [c3 cos 2x + c4 sin 2x] +
1 x4 – 47 x2 + 589
25  25 625
 3 3 
28. (D2 – D + 1) y = x3 – 3x2 + 1 Ans. y = ex/2 c1 cos x + c2 sin x + x3 – 6x – 5
 2 2 
29. (D3 – 3D2 + 3D – 1) y = 2x3 – 3x2 + 1. Ans. y = (c1x2 + c2x + c3) ex – (2x3 + 15x2 + 54x + 85)
1
30. (D3 – 2D + 4) y = 3x2 – 5x + 2. (Dec. 2012) Ans. c1e–2x + ex (c2 cos x + c3 sin x) + (3x2 – 2x + 1)
4
d3 y 1 4
31. + 8y = x4 + 2x + 1. Ans. y = c1 e–2x + ex [A cos 3 x + B sin 3 x] + (x – x + 1)
dx3 8

1 2 7 1 3
32. (D2 – 3D + 2) y = x2 + sin x. Ans. y = c1ex + c2e2x + (x + 3x + + sin x + cos x
2 2 10 10

3
33. (D3 + 3D2 – 4) y = 6e–2x + 4x2. Ans. y = c1ex + (c2x + c3) e–2x – x2 e–2x – x2 –
2

34. (D3 + 6D2 + 12D + 8) y = e–2x + x2 + 3x + cos 2x.

x3 –2x 1 2 1 1
Ans. y = (c1x + c2x + c3) e–2x + e + (x – 3x + 3) + 3x + (sin 2x – cos 2x)
6 8 (log 3 + 2)3 32

Ans. y = (c1x + c2) e2x + 4x2 e2x + cos 2x + 2 x2 + 2x +

3
35. (D2 – 4D + 4) y = 8 (e2x + sin 2x + x2). (May 2006)
 2

36. (D5 – D) y = 12 ex + 8 sin x – 2x Ans. y = c1 + c2 e–x + c3 ex + A cos x + B sin x + 3x ex + 2x sin x + x2

1 x
37. (D2 – 1) y = ex + x3. (May 2017) Ans. y = c1ex + c2e–x + xe – x3 – 6x
2

1 2 2x 1 1
38. (D2 – 4D + 4) y = e2x + x3 + cos 2x (Dec. 2010) Ans. y = (c1 + c2 x) e2x + x e – sin 2x + [2x3 + 6x2 + 9x + 6]
2 8 8

39. (D5 – D) y = 12ex + 85mx + 2x (May 2018)

x2 2x
Ans. y = c1 + c2 ex + c3 e–x + c4 cos x + c5 sin x + 3x ex – 35 m +
2 (log 2)5 – log 2
e3x
40. (D2 – 4) y = e3x x2. Ans. y = c1e2x + c2e–2x + (125x2 – 60x + 62)
125
d3 y dy e2x 169+ x2 + 5x
41. –7 – 6y = e2x (1 + x2) Ans. y = c1 e–x + c2 e–2x + c3 e3x –
dx3 dx 12  72 6
8ex x7/2
42. (D3 – 3D2 + 3D – 1) y = x ex. Ans. y = (c1 x2 + c2 x + c3) ex +
105
1 2x
43. (D2 – 4D + 4) y = e2x sin 3x Ans. y = (c1 + c2 x) e2x – e sin 3x
9
ex
44. (D3 – D2 + 3D + 5) y = ex cos 3x Ans. y = c1 e–x + ex (c2 cos 2x + c3 sin 2x) – (3 sin 3x + 2 cos 3x)
65
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.33) LINEAR DIFFERENTIAL EQUATIONS…
…

e–x
45. (D2 + 2D + 1) y = (Dec. 2007) Ans. y = (c1 + c2 x) e–x – e–x [x log (x + 2) + 2 log (x + 2) – x]
x+2
1 –3x e–3x
46. (D2 + 6D + 9) y = e (Dec. 2014, 2017, May 2018) Ans. y = (c1 x + c2) e–3x +
x3 2x
ex 2 7
47. (D4 – 3D3 – 2D2 + 4D + 4) y = x2 ex. Ans. y = (c1x + c2) e–x + (c3x + c4) e2x + (x + 2x + )
4 2

x + x + x
5 4 3
x2
48. (D3 – 3D – 2) y = 540 x3 e–x. Ans. y = (c1x + c2) e–x + c3 e2x – 180e–x +
20 12 9 9
d2 y dy  sin x 
49. +2 + 2y = e–x sec3 x Ans. y = e–x c1 cos x + c2 sin x + tan x
dx2 dx  2 
e–x x2
50. (D2 + 2D + 1) y = e–x log x. (May 2015) Ans. y = (c1x + c2) e–x + (2 log x – 3x2)
4

 3 
51. (D4 + D2 + 1) y = e–x/2 cos  x
 2 

 3 3   3 3  1  3 3
Ans. y = ex/2 c1 cos x + c2 sin x + e–x/2 c3 cos x + c4 sin x + x e–x/2 sin x + 3 cos x 
 2 2   2 2  4 3  2 2 

ex e–x
52. (D3 – D2 – D + 1) y = cosh x sin x. Ans. y = (c1x + c2) ex + c3 e–x + (cos x – 2 sin x) – (3 cos x – 4 sin x)
10 50
d2 y 1
53. – y = cosh x cos x (Dec. 2016) Ans. y = c1 ex + c2 e–x + (2 sinh x sin x – cosh x cos x)
dx2 5
3 1
54. (D2 + 40D + 8) y = 12e–2x sin x sin 3x. (Dec. 2004) Ans. y = e–2x (c1 cos 2x + c2 sin 2x) + x e–2x sin 2x + e–2x cos 4x
2 2
ex x2 3  1 1
55. (D3 – 6D2 + 11D – 6) y = ex x + sin x + cos x. Ans. y = c1ex + c2e2x + c3e3x + + x – cos x + sin x
2  3 2  10 10
d2 y dy
56. +5 + 6y = e–2x sin 2x + 4x2 ex x (May 2011)
dx2 dx
e–2x ex x2 – 7 x + 37
Ans. y = c1 e–2x + c2 e–3x – (cos 2x + 2 sin 2x) +
10 3  6 72
d3y d2y x2 ex x3 3x2
57. – = 3x + x ex. Ans. y = c1 + c2 x + c3 ex – 2x ex + – –
dx3 dx2 2 2 2
d2 y x 3
Ans. y = c2 ex + c1 e2x + e3x 2 – 4 +
dy 1
58. –3 + 2y = x e3x + sin 2x. (3 cos 2x – sin 2x)
dx2 dx   20
3x
2e sin 4x 2x
59. (D2 – 6D + 13) y = 8 e3x sin 4x + 2x Ans. y = e3x (A cos 2x + B sin 2x) – + 2
3 (log 2) – 6 log 2 + 13

60. (D4 + D2 + 1) y = ax2 + be–x sin 2x. Ans. y = e(–1/2) x [c1 cos ( 3/2) x + c2 sin ( 3/2) x] + e(1/2) x [c3 cos ( 3/2) x
b –x
+ c4 sin ( 3/2 x] + a (x2 – 2) – e (20 cos 2x + 9 sin 2x)
481
1 2
61. (D2 – 4) y = x sinh x (May 2006) Ans. y = c1e2x + c2 e–2x – [x sinh x + cosh x]
3 3

62. (D2 – 2D + 1) y = x2 ex sin x. Ans. y = (c1x + c2) ex – ex [4x cos x + (x2 – 6) sin x]
2
dy dy
63. –4 + 4y = 8x2 · e2x sin 2x (Dec. 2004) Ans. y = e2x [c1 + c2 x + 3 sin 2x – 2x2 sin 2x – 4x cos 2x]
dx2 dx
1
64. (D2 + 2D + 1) y = x cos x Ans. y = (c1 x + c2) e–x + (x sin x + cos x – sin x)
2
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.34) LINEAR DIFFERENTIAL EQUATIONS…
…

d2 y dy 7 – 30x 12 – 5x

65. +3 + 2y = x sin 2x Ans. y = c1 e–2x + c2 e–x + cos 2x + sin 2x
dx2 dx  200   100 
3 2
x x
66. (D4 + 2D2 + 1) y = x cos x. (May 2012, Dec. 2012) Ans. y = (c1x + c2) cos x + (c3x + c4) sin x – cos x + sin x
24 2
67. (D2 + 1)2 y = 24x cos x. Ans. y = (c1x + c2) cos x + (c3x + c4) sin x – x3 cos x + 3x2 sin x
2 2 –x
68. (D + 2D + 5) y = x e cos 2x. (Dec. 2018)
e–x  3
(x – x2) cos 2x – x3 sin 2x
2
Ans. y = e–x [(c1x + c2) cos 2x + (c3x + c4) sin 2x] –
32  3 
69. (D2 – 2D + 4)2 y = xex cos [ 3 x + α] (Dec. 2006; May 2007, 2008)
ex x 
3
x2
Ans. y = ex [(c1 + c2 x) cos 3 x + (c3 + c4 x) sin 3 x] – 6 cos ( 3 x + α) + sin ( 3 x + α)
12  2 3 
e2x
70. (D2 – 4D + 4) y = x e2x sin 2x. (Dec. 2005, 2008) Ans. y = (c1x + c2) e2x – [x sin 2x + cos 2x]
4
(C) Method of Variation of Parameters
When the short-cut methods (Art. 1.13) fail to determine the particular integral then one has to make use of general method. But
this method involves laborious integration and in such cases other methods are available. One such method is the method of
variation of parameters. This method is due to a great Mathematician named Lagrange. To explain the rigours of this method, let
us start with a simple differential equation
d2 y
+ y = tan x … (1)
dx2
Here C.F. is very simple but the P.I. will be difficult to obtain even by general method because it is not of those special cases
discussed before.
The complementary function is;
A cos x + B sin x … (2)
where, A and B are Arbitrary constants. Here Lagrange has shown his ingenuity by evolving the Particular Integral from this C.F.
only by assuming that (temporarily) the constants A and B are some functions of x say A (x) and B (x) (of course it looks
ridiculous).
Since the method assumes that the quantities A and B vary, this method is called The Method of Variation of Parameters or
Variation of Constants.
Since two functions A (x) and B (x) are to be determined, they must satisfy two conditions. First is that the assumed solution (P.I.)
y = A (x) cos x + B (x) sin x … (3)
must satisfy the differential equation. When determined, (3) actually will deliver to us the Particular Integral. The second
condition is at our disposal and we shall choose it at proper time so as to evaluate A (x) and B (x) and thereby solving the
equation.
If we differentiate equation (3), we get
y' = – A (x) sin x + B (x) cos x + A'(x) cos x + B' (x) sin x … (4)
Since further differentiation will involve higher differentials of unknown functions A(x) and B(x), we apply our choice of second
condition here only and that is what we assume
A'(x) cos x + B'(x) sin x = 0 … (5)
and then (4) becomes simpler as
y' = – A (x) sin x + B (x) cos x … (6)
One further differentiation will give
y" = – A(x) cos x – B(x) sin x – A' (x) sin x + B'(x) cos x … (7)
Substituting from (3) and (7) in the given differential equation, we find that
– A' (x) sin x + B' (x) cos x = tan x … (8)
Now, if we solve equations (5) and (8) simultaneously, we get
sin2 x
A'(x) = – and B' (x) = sin x
cos x
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.35) LINEAR DIFFERENTIAL EQUATIONS…
…

and hence by integration, we get

⌠ cos2 x – 1 ⌠
A (x) =  dx =  (cos x – sec x) dx
⌡ cos x ⌡
= sin x – log (sec x + tan x)
and B (x) = – cos x
We are not using here constants of integration because it is P.I. part.
Now we frame our P.I. as follows :
y = A (x) cos x + B (x) sin x
P.I. = cos x [sin x – log (sec x + tan x)] – sin x cos x = – [log (sec x + tan x)] cos x
Hence the complete solution is
y = yc + yp
y = c1 cos x + c2 sin x – cos x log (sec x + tan x)
Note : Lagrange's method may be extended to higher order linear differential equations too, as may be seen by further exercises.
Second Method of Variation of Parameters
d2y dy
When we have to solve equation of the type a +b + cy = X
dx2 dx
where, a, b, c are constants and X, any function of x, we also have an alternative method of variation of parameters.
Let the complementary function = Ay1 + By2 then the particular integral = uy1 + vy2

⌠ – y2 X ⌠ y1 X
where u = ⌡ W
dx, v =⌡
W
dx

y1 y2
where W = = called "WRONSKIAN" = (y1 y ' – y ' y2)
y' y' 2 1
1 2

ILLUSTRATIONS ON METHOD OF VARIATION OF PARAMETERS

Ex. 1 : Solve the equation (D2 + 4) y = sec 2x by the method of variation of parameters. (Dec. 2012)
Sol. : C.F. = A cos 2x + B sin 2x … (1)
Let P.I. = y = A (x) cos 2x + B (x) sin 2x … (2)
Differentiating (2), we have
y' = – 2 A (x) sin 2x + 2B (x) cos 2x + A'(x) cos 2x + B' (x) sin 2x … (3)
Assume here that
A' (x) cos 2x + B'(x) sin 2x = 0 … (4)
Then equation (3) will become
y' = – 2A (x) sin 2x + 2B (x) cos 2x … (5)
If we differentiate (5) again, we get
y" = – 4 A cos 2x – 4 B sin 2x – 2 A' sin 2x + 2 B' cos 2x … (6)
[Briefly A (x) = A, A' = A' (x), B (x) = B, B' = B' (x)]
Putting values of y, y' and y" in the differential equation
d2 y
+ 4y = sec 2x,
dx2
We have
(– 4A cos 2x – 4B sin 2x – 2A' sin 2x + 2B' cos 2x) + (4A cos 2x + 4B sin 2x) = sec 2x
⇒– 2A' sin 2x + 2B' cos 2x = sec 2x … (7)
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.36) LINEAR DIFFERENTIAL EQUATIONS…
…

Solving (4) and (7) simultaneously, we have

A' cos 2x + B' sin 2x = 0
1
– A' sin 2x + B' cos 2x = sec 2x
2
1 1
B' = ⇒ B = x and
2 2
–1 1
A' = tan 2x ⇒ A = log (cos 2x)
2 4
Hence P.I. = A cos 2x + B sin 2x
1 x
= cos 2x log (cos 2x) + sin 2x
4 2
Hence the complete solution is
x 1
y = c1 cos 2x + c2 sin 2x + sin 2x + cos 2x log cos 2x
2 4
Alternative Method :
(D2 + 4) y = sec 2x
C.F. = A cos 2x + B sin 2x = Ay1 + By2
Here y1 = cos 2x and y2 = sin 2x
Let P. I. = u(x) y1 + v(x) y2
y1 y2 cos 2x sin 2x
W = = = 2 (cos2 2x + sin2 2x) = 2
y' y' – 2 sin 2x 2 cos 2x
1 2

– y2X – sin 2x sec 2x 1

⌠
⌡ ⌠
⌡ = – ⌠
2⌡
u = dx = tan 2x dx
W 2
1
= log (cos 2x)
4
y1X cos 2x sec 2x 1 1
v = ⌠
⌡ dx = ⌠
⌡ dx = ⌠
⌡ dx = x
W 2 2 2
1  1 
∴ P.I. =  log (cos 2x) cos 2x +  x sin 2x
4  2 
Hence the general solution is
1 1
y = A cos 2x + B sin 2x + cos 2x log (cos 2x) + x sin 2x
4 2
d2y
Ex. 2 : Solve by method of variation of parameters + y = cosec x . (Dec. 2004)
dx2
Sol. : A.E. is D2 + 1 = 0 ∴ D=±i
C.F. = A cos x + B sin x
= Ay1 + By2
Here y1 = cos x and y2 = sin x
Let P.I. = uy1 + vy2

y1 y2
cos x sin x
W = = = 1
y' y' – sin x cos x
1 2

⌠ – y2 X ⌠ – sin x cosec x dx = ⌠ – dx
u = ⌡ W
dx = ⌡
1 ⌡

= –x
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.37) LINEAR DIFFERENTIAL EQUATIONS…
…

⌠ y1 X
⌠
dx = ⌡
cos x cosec x
⌠
and v = ⌡ D 1
dx = ⌡ cot x dx

= log (sin x)
∴ P. I. = (–x) cos x + {log (sin x)} sin x
Hence the general solution is
y = A cos x + B sin x – x cos x + sin x log (sin x)
d2y 2
Ex. 3 : Solve by method of variation of parameters –y = . (Dec. 2005, 2006, May 2017)
dx2 1 + ex
Sol. : A.E. is D2 – 1 = 0 ∴ D = ± 1
C.F. = c1 ex + c2 e–x
= c1y1 + c2y2
x –x
Here y1 = e and y2 = e , then
y1 y2 ex e–x
W = = = –2
y' y' ex –e–x
1 2

⌠ ⌠ y2 X = – ⌠ e
–x
– y2 X  2 x
u = ⌡ W
dx = – ⌡
–2 ⌡ –2 1 + e 
dx

⌠ e–x ⌠ dx ⌠  1x – 1 x dx
= ⌡ 1 + ex
dx = ⌡ x
e (1 + ex)
=⌡
e 1 + e 
⌠ e–x dx – ⌡
⌠ e–x dx
= – e–x + log (1 + e–x)
u = ⌡ e–x + 1

⌠ ⌠ e
x
y1 X  2 x
v = ⌡ W
dx = ⌡
–2 1 + e 
dx

= –⌡
⌠ ex dx
= – log (1 + ex)
1 + ex
∴ P.I. = u y1 + v y2 = [– e–x + log (1 + e–x)] ex – {log (1 + ex)} e–x
= – 1 + ex log (e–x + 1) – e–x log (ex + 1)
Hence the general solution is
y = c1 ex + c2 e–x – 1 + ex log (e–x + 1) – e–x log (ex + 1)
e3x
Ex. 4 : Solve by method of variation of parameters (D2 – 6D + 9) y = . (May 2009, 2014, Dec. 2010, Nov. 2019)
x2
Sol. : A.E. is D2 – 6D + 9 = 0 ∴ D = 3, 3
C.F. = (c1x + c2) e3x
= c1y1 + c2y2
Here y1 = x e3x and y2 = e3x
Let P.I. = uy1 + vy2

y1 y2 x e3x e3x
W = = = – e6x
y' y' (3x + 1) e3x 3e3x
1 2

–y2X – e3x (e3x/x2) 1

u = ⌠
⌡ dx = ⌠
⌡ dx = ⌠
⌡ dx
W – e6x x2
1
= –
x
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.38) LINEAR DIFFERENTIAL EQUATIONS…
…

y1X xe3x (e3x/x2) 1

and v = ⌠
⌡ dx = ⌠
⌡ dx = ⌠
⌡ – dx
W – e6x x
= – log x
1
∴ P.I. = – (x e3x) – log x (e3x) = – e3x (1 + log x)
x
Hence the general solution is
y = (c1x + c2) e3x – e3x (1 + log x)
Ex. 5 : Use method of variation of parameters to solve (D2 – 2D + 2) y = ex tan x (May 2007, 2008, Nov. 2014, 2015)
Sol. : A.E. is D2 – 2D + 2 = 0 ∴ D = 1 ± i.
C.F. = ex (c1 cos x + c2 sin x)
= c1y1 + c2y2
Here, y1 = ex cos x and y2 = ex sin x
Let P.I. = uy1 + vy2
y1 y2 ex cos x ex sin x
W = = = e2x
y' y' ex (cos x – sin x) ex (sin x + cos x)
1 2

– y2X – ex sin x ex tan x

u = ⌠
⌡ ⌠
dx = ⌡ dx
W e2x
2 2
– sin x (1 – cos x)
= ⌠
⌡ dx = – ⌠
⌡ cos x dx
cos x
= – log (sec x + tan x) + sin x
yX ex cos x ex tan x
⌠ 1 dx = ⌡
v = ⌡ ⌠ dx
W e2x

= ⌠
⌡ sin x dx = – cos x

∴ P.I. = (– log sec x + tan x + sin x) ex cos x + (– cos x) ex sin x

Hence the general solution is
y = ex (c1 cos x + c2 sin x) – ex cos x log (sec x + tan x)
1
Ex. 6 : Solve by method of variation of parameters (D2 + 9) y = . (Dec. 2008)
1 + sin 3x
2
Sol. : A.E. is D + 9 = 0 ∴ D = ± i3.
C.F. = c1 cos 3x + c2 sin 3x
= c1y1 + c2y2
Here, y1 = cos 3x and y2 = sin 3x
Let P.I. = uy1 + vy2
y1 y2 cos 3x sin 3x
W = = = 3
y' y' – 3 sin 3x 3 cos 3x
1 2

– y2X – sin 3x (1/1 + sin 3x)

u = ⌠
⌡ dx = ⌠
⌡ dx
W 3
1 sin 3x (1 – sin 3x)
= – ⌠
⌡ dx
3 (1 + sin 3x) (1 – sin 3x)
1 sin 3x – sin2 3x 1
⌠
⌡ dx = – ⌠
3⌡
= – (sec 3x tan 3x – tan2 3x) dx
3 cos2 3x
1
⌠ (sec 3x tan 3x – sec2 3x + 1) dx
3⌡
= –

1 1 1 
= – sec 3x + tan 3x – x
3 3 3 
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.39) LINEAR DIFFERENTIAL EQUATIONS…
…

y1X cos 3x (1/1 + sin 3x) 1 cos 3x

v = ⌠
⌡ dx = ⌠
⌡ dx = ⌡⌠ dx
W 3 3 1 + sin 3x
1
= log (1 + sin 3x)
9
1  1 
∴ P.I. =  (– sec 3x + tan 3x – 3x) cos 3x +  log (1 + sin 3x) sin 3x
 9   9 
Hence the general solution is
1 1
y = (c1 cos 3x + c2 sin 3x) + (– 1 + sin 3x – 3x cos 3x) + sin 3x log (1 + sin 3x)
9 9

d2y
Ex. 7 : Solve by method of variation of parameters – y = e–x sin (e–x) + cos (e–x) … (1)
dx2

Sol. : C.F. = A ex + B e–x

Let P. I. = y = A (x) ex + B (x) e–x … (2)
A (x) and B (x) are functions to be determined.
Differentiating (2), we have
y' = A ex – B e–x + A' ex + B' e–x … (3)
x –x
Put A' e + B' e = 0, … (4)
x –x
then (3) will become y' = A e – B e
Differentiating again
y" = A ex + B e–x + A' ex – B' e–x … (5)
Putting values of y" and y in (1), we have
A ex + B e–x + A' ex – B' e–x – Aex – Be–x = e–x sin (e–x) + cos (e–x)
∴ A' ex – B' e–x = e–x sin (e–x) + cos (e–x) … (6)
Solving (4) and (6) simultaneously for A', B',
A' ex + B' e–x = 0
A' ex – B' e–x = e–x sin (e–x) + cos (e–x) … (7)
Adding the equations in (7), we have
2 A' ex = e–x sin (e–x) + cos (e–x)
1 –x –x
∴ A' = e [e sin (e–x) + cos (e–x)] … (8)
2
1 x –x
and similarly, B' = – e [e sin (e–x) + cos (e–x)] … (9)
2
Integrating (8)

1 ⌠ e–x [e–x sin (e–x) + cos (e–x)] dx [put e–x = t, –e–x dx = dt]
A =
2 ⌡

1 ⌠ 1
A = –
2 ⌡ [t sin t + cos t] dt = –
2
[– t cos t + sin t + sin t]

1
= t cos t – sin t
2
1 –x
Hence A (x) = e cos (e–x) – sin (e–x) … (10)
2
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.40) LINEAR DIFFERENTIAL EQUATIONS…
…

If we integrate (9),

1 ⌠ ex (e–x sin e–x + cos e–x) dx = –

1 x
e · cos e–x
B = –
2 ⌡ 2
1 x
B (x) = – e cos e–x, hence P.I. will be given by
2

y = ex
1 e–x cos e–x – sin e–x
 –
1 x
e cos (e–x) · e–x
2  2
1 1
= cos (e–x) – ex sin (e–x) – cos e–x = – ex sin (e–x)
2 2
Hence the complete solution is
y = A ex + B e–x – ex sin (e–x)
Ex. 8 : By the method of variation of parameters, solve
(D3 + D) y = cosec x … (I)
Sol. : C.F. = A + B cos x + C sin x
Let the P.I. = yp = A (x) + B (x) cos x + C (x) sin x … (II)
where A (x), B (x) and C (x) are the parameters to be determined. For brevity, take A (x) = A, B (x) = B, C (x) = C.
Hence P.I. = y = A + B cos x + C sin x
y' = A' + (B' cos x – B sin x) + (C' sin x + C cos x)
Put A' + B' cos x + C' sin x = 0 … (III)
So that the new value of y' becomes
y' = – B sin x + C cos x
∴ y" = – B' sin x – B cos x + C' cos x – C sin x
Choose B' and C' such that
– B' sin x + C' cos x = 0 … (IV)
hence y" = – B cos x – C sin x
and y"' = – B' cos x + B sin x – C' sin x – C cos x
Substituting in (I) values of y, y' and y"', we get
– B' cos x – C' sin x = cosec x …(V)
Solving simultaneously (III), (IV) and (V), we get
A' = cosec x, B' = – cot x and C' = – 1
and integration yields A = log [cosec x – cot x]
B = – log sin x
C = –x
∴ P.I. = log (cosec x – cot x) – cos x log sin x – x sin x
Hence the complete solution is
y = A + B cos x + C sin x + log [cosec x – cot x] – cos x log (sin x) – x sin x

EXERCISE 1.3

Solve the following differential equations by the method of variation of parameters.

1. y" + y = sec x (May 2007, 2019) Ans. y = A cos x + B sin x + x sin x + cos x log cos x
d2 y x x2
2. + y = x sin x. (Dec, 2010) Ans. y = A cos x + B sin x + sin x – cos x
dx2 2 4
3. (D2 + 3D + 2) y = sin ex Ans. y = c1e–x + c2e–2x – e–2x sin ex
ENGINEERING MATHEMATICS – III (Comp. Engg. and IT Group) (S-II) (1.41) LINEAR DIFFERENTIAL EQUATIONS…
…

d2 y dy ex
4. –2 = ex sin x Ans. y = A + B e2x – sin x
dx2 dx 2

5. (D2 + 4) y = 4 sec2 2x. (May 2005) Ans. y = A cos 2x + B sin 2x – 1 + sin 2x log (sec 2x + tan 2x)

6. (D2 + D) y = (1 + ex)–1 (May 2006, 2008) Ans. y = c1 + c2 e–x + x – log (1 + ex) – e–x log (1 + ex)

7. (D2 – 1) y = (1 + e– x)–2 Ans. y = A ex + B e–x – 1 + e–x log (1 + ex)

d2 y dy x x
8. 2 + 3 + 2y = ee (May 2005, 2006, May 2018) Ans. y = Ae–x + B e–2x + e–2x ee
dx dx

9. (D2 – 4D + 4) y = e2x sec2 x (May 2010) Ans. y = [c1 + c2 x + log (sec x)] e2x
d2 y
10. + y = tan x. (Dec. 2004) Ans. y = A cos x + B sin x – cos x log (sec x + tan x)
dx2

d2 y 1
11. + 4y = tan 2x Ans. y = A cos 2x + B sin 2x – cos 2x log (sec 2x + tan 2x)
dx2 4

d2 y
12. + y = sec x tan x Ans. y = A cos x + B sin x + x cos x – sin x + sin x log (sec x)
dx2

13. y" + y = sec x (Dec. 2004) Ans. y = A cos x + B sin x + x sin x + cos x log cos x

14. (D2 + 1) y = 3x – 8 cot x. Ans. y = c1 cos x + c2 sin x + 3x – 8 sin x log (cosec x – cot x)
1 1 1 1 
15. (D2 + 4) y = Ans. y = c1 cos 2x + c2 sin 2x + (cos 2x) log (1 + cos 2x) + x – tan x sin 2x
1 + cos 2x 4 2 2 
1.12 EQUATIONS REDUCIBLE TO LINEAR WITH CONSTANT COEFFICIENTS
We shall now study two types of linear differential equations with variable coefficients which can be reduced to the case of linear
differential equation with constant coefficients by suitable transformations of variables.
1.13 CAUCHY'S OR EULER'S HOMOGENEOUS LINEAR DIFFERENTIAL EQUATION
An equation of the type

(a0 xn Dn + a1 xn–1 Dn–1 + … + an–1 x D + an) y = F (x)

Where, a0, a1, a2 …… an are constants is called Cauchy's Homogeneous Equation. It is sometimes attributed to Euler also. It may
also be written as
dn y n–1 d
n–1
y dy
a0 xn n + a1 x + … + an–1 x + an y = F (x) … (1)
dx dxn–1 dx

It can be reduced to linear differential equation with constant coefficients by putting

x = ez or z = log x … (2)
dy dy dz 1 dy
Now = =
dx dz dx x dz

dy dy d
or x = = Dy, here we took D ≡
dx dz dz

d2 y d 1 dy 1 dy 1 d dy dz
Also, = = – +
dx2 dx  x dz x2 dz x dz dz dx

d y2
2
1 dy 1 1 1 dy 1 d2 y
= – + = – 2 + 2
x2 dz x dz  x x dz x dz2