Step (ii): Integrate M with respect to x treating y as a

constant.
(1) Exact differential equation: If M and N are
functions of x and y, the equation 𝑀𝑑𝑥 + 𝑁𝑑𝑦 = 0 is Step (iii): Integrate N with respect to y treating x as
called exact when there exists a function 𝑓(𝑥, 𝑦) of x constant and omit those terms which have been already
and y such that obtained by integrating M.

𝑑[𝑓(𝑥, 𝑦) = 𝑀𝑑𝑥 + 𝑁𝑑𝑦 Step (iv): On adding the terms obtained in steps (ii)
and (iii) and equating to an arbitrary constant, we get
i.e., the required solution.
𝜕𝑓 𝜕𝑓
𝑑𝑥 + 𝑑𝑦 = 𝑀𝑑𝑥 + 𝑁𝑑𝑦 In other words, solution of an exact differential
𝜕𝑥 𝜕𝑦
equation
where
∫ 𝑀𝑑𝑥 + ∫ 𝑁𝑑𝑦 = 𝑐
𝜕𝑓
= Partial derivative of 𝑓(𝑥, 𝑦)
𝜕𝑥
with respect to x (keeping y constant) . Regarding y as constant Only those terms not
𝜕𝑓
= Partial derivative of 𝑓(𝑥, 𝑦) with respect to y
𝜕𝑦 containing x
(treating x as constant)

The necessary and sufficient condition for the

differential condition 𝑀𝑑𝑥 + 𝑁𝑑𝑦 = 0 to be exact is
𝜕𝑀 𝜕𝑁
= 𝜕𝑥 .
𝜕𝑦
Differential Equations

An exact differential equation can always be derived  Definition: An equation involving independent
from its general solution directly by differentiation variable, dependent variable and its derivatives
without any subsequent multiplication, elimination is said to be differential equation.
etc.
e.g. An equation of the form of ∴ 𝑦 = 𝑓(𝑥, 𝑦, 𝑝)
(2) Integrating factor: If an equation of the form
𝑀𝑑𝑥 + 𝑁𝑑𝑦 = 0 is not exact, it can always be made is said to be differential equation, where, 𝑝 =
𝑑𝑦
exact by multiplying by some function of x and y. Such 𝑑𝑥
a multiplier is called an integrating factor. e.g.

𝑑𝑦 3
(3) Working rule for solving an exact differential (1) 𝑦 = (𝑑𝑥 ) + 5
equation:
𝑑2 𝑦 𝑑𝑦
Step (i): Compare the given equation with 𝑀𝑑𝑥 + (2) 𝑥 2 𝑑𝑥 2 + 𝑦 𝑑𝑥 = 5𝑦 + 𝑐 etc.
𝜕𝑀
𝑁𝑑𝑦 = 0 and find out M and N. Then find out 𝜕𝑦 and
𝜕𝑁 𝜕𝑀 𝜕𝑁
. If = , the given equation is exact.
𝜕𝑥 𝜕𝑦 𝜕𝑥
  Order and Degree of differential Equation: 𝜕 2𝑢 𝜕 2𝑢 𝜕 2𝑢
+ + =0
𝜕𝑥 2 𝜕𝑦 2 𝜕𝑧 2
The order of highest order derivative occurred in
differential equation is said to be the order of the
differential equation whereas power/exponent of the
highest order derivative term in the different equation  Methods for solving First Order, First
whereas power/exponent of the highest order Degree Differential equations:
derivative term in the differential equation is said to be
the degree of the differential equation. 𝑑𝑦 𝑀
An equation of the form 𝑑𝑥 = 𝑁 , where M and N be
e.g. the functions of x and y or any constant value, is said
to be the first order, first degree differential equation.
2
𝑑3 𝑦 𝑑2 𝑦 𝑑𝑦 6
(1) 𝑑𝑥 3 + 2 (𝑑𝑥 2 ) + 6 (𝑑𝑥 ) + 7𝑦 = 0 This type of differential equation consists of following
types of function
Here order of differential equation be 3
1. Separable function
Degree of differential equation = 1 2. Reducible separable function
3. Homogeneous function and Non-Homogeneous
3
function
𝑑𝑦 2 2 𝑑3 𝑦
(2) [1 + (𝑑𝑥 ) ] = 𝑑𝑥 3 4. Linear Differential equation
5. Reducible linear differential equation
6. Exact differential equation
Order of this differential equation = 3

Degree of this differential equation = 2

Solved Example
Types of Differential Equation:
1. Find the general solution of differential
1. Ordinary Differential Equation: A
equation
differential equation involving single
independent variable, is said to be an ordinary
𝑑𝑦
differential equation. = (1 + 𝑥 2 )(1 + 𝑦 2 ).
𝑑𝑥
𝑑𝑦
eg. (1) + 6𝑦 = 6𝑥 2 Solution:
𝑑𝑥

𝑑2 𝑦 𝑑𝑦 𝒅𝒚
(2) + 2 𝑑𝑥 + 6𝑦 = 0 = (𝟏 + 𝑥 2 ). (𝟏 + 𝑦 2 )
𝑑𝑥 2
𝒅𝒙
𝒅𝒚
2. Partial Differential Equation: A differential ⇒ = (𝟏 + 𝑥 2 )𝒅𝒙
𝟏 + 𝑦2
equation having two or more than two
independent variables is said to be a partial
Integrating both sides, we get
differential equation.

If 𝑢 = 𝑓(𝑥, 𝑦, 𝑧), then its partial differential equation 𝒙𝟑

tan−1 𝑦 = 𝑥 + +𝑐
(P.D.E) will be 2
 𝟐𝒙 + 𝒙𝟑 + 𝟐𝒄 𝑑𝑦 𝑥𝑑𝑣
⇒ 𝒚 = 𝒕𝒂𝒏 ( ) =𝑣+
𝟐 𝑑𝑥 𝑑𝑥
𝑑𝑦
On replacing the value of 𝑑𝑥 in equ (1), it becomes

 Method for solving the Homogeneous 𝑥𝑑𝑣 𝑥 + 𝑣𝑥

differential equation 𝑣+ = = (1 + 𝑣)
𝑑𝑥 𝑥
𝑑. 𝑣
Step 1: First of all arrange the given equation as ⇒𝑥 = 1+𝑣−𝑣 =1
𝑑𝑥
𝑑𝑥
𝑑𝑦 𝜙(𝑥) ⇒ 𝑑𝑣 =
𝑥
= ,
𝑑𝑥 𝑓(𝑥)
where 𝜙(𝑥) and 𝑓(𝑥) having same degree. 𝑂𝑛 𝑖𝑛𝑡𝑒𝑔𝑟𝑎𝑡𝑖𝑛𝑔 𝑏𝑜𝑡ℎ 𝑠𝑖𝑑𝑒𝑠, we get

Step2: Put 𝑦 = 𝑣𝑥 and differentiate it on both sides to 𝑣 = 𝑙𝑜𝑔𝑥 + 𝑙𝑜𝑔𝑐

𝑦
get = 𝑙𝑜𝑔𝑥𝑐
𝑥
𝑑𝑦 𝑥𝑑𝑣 (where 𝑙𝑜𝑔 𝑐 = integration constant)
=𝑣+ 𝑦 = 𝑥𝑙𝑜𝑔𝑥𝑐 which is the required solution.
𝑑𝑥 𝑑𝑥

Step 3: Then put these value in the given equation and

further hence using variable separable method,
solution will be obtained in the form of v.  Linear Differential Equation: An equation of
𝑑𝑦
the form of + 𝑃. 𝑦 = 𝑄, where P and Q is in
𝑦 𝑑𝑥
Step 4: Then replace v by 𝑥 the form of x or any constant (but not of y), is
said to be a linear differential equation.
& hence find the result.
 Method of Solving Linear Differential
Equation (L.D.E.)

Solved Example Step 1: First of all, find the integrating factor (I.F.)

1. Solve the differential equation 𝐼. 𝐹. = 𝒆∫ 𝑷.𝒅𝒙

𝑥+𝑦 Step 2: Then, the general solution will be given as

𝑦′ = .
𝑥
𝑦(𝐼. 𝐹. ) = ∫ 𝑄. (𝐼. 𝐹. )𝑑𝑥
𝒅𝒚 𝒙+𝒚
Solution: 𝒚′ = 𝒅𝒙 = . (1)
𝒙
Solved Example
It is a homogeneous equation. 𝑑𝑦
1. Solve 𝑑𝑥 + 𝑦𝑥 = 𝑥 3 .
Now, put 𝑦 = 𝑣𝑥
𝑑𝑦 1
Solution: Given 𝑑𝑥 + (𝑥) . 𝑦 = 𝑥 3
(2) On differentiating both sides of equ (2), we get
which is a Linear Differential Equation (L.D.E.)  Equation Reducible to Linear Form
(Bernoulli Equation): There are the equations
1 of the type
Here, 𝑃 = 𝑥 and 𝑄 = 𝑥 3
𝑑𝑦
Now, 𝐼. 𝐹. = 𝒆∫ 𝑷.𝒅𝒙 = 𝒆∫ 𝟏/𝒙.𝒅𝒙 = 𝑒 ln 𝑥 = 𝑥 + 𝑃𝑦 = 𝑄. 𝑦 𝑛 … … … …(1)
𝑑𝑥

∴ Hence the general solution will be Where P and Q are the functions of x, which do not
𝑦. (𝐼. 𝐹) = ∫ 𝑄. (𝐼. 𝐹). 𝑑𝑥 appear directly to be of linear form, but can be easily
𝑦. 𝑥 = ∫ 𝑥 3 . 𝑥. 𝑑𝑥 = ∫ 𝑥 4 . 𝑑𝑥 reduced into the linear form by a suitable
𝑥5 transformation.
⇒ 𝑦𝑥 = +𝑐
5
⇒ 5𝑥𝑦 = 𝑥 5 + 𝐾
 Method for solving Bernoulli Equations
which is the required solution.
Step 1: Divide both sides of equation (1) by 𝑦 𝑛 , we get

1 𝑑𝑦 1
 Linear Differential Equation in the form of + 𝑃. 𝑦 𝑛−1 = 𝑄 (2)
𝑦 𝑛 𝑑𝑥
𝒅𝒙
: Sometimes, we see that the given linear
𝒅𝒚 1
𝑑𝑦 Step 2: Then put 𝑦 𝑛−1 = 𝑦 −𝑛+1 = 𝑧
differential equation is not in the form of (𝑑𝑥 ).
𝑑𝑦
But it is linear in (𝑑𝑥 ). After the certain
𝑑𝑧 𝑑𝑦
manipulations in such case the linear equation ⇒ = (−𝑛 + 1). 𝑦 −𝑛 .
𝑑𝑥 𝑑𝑥 𝑑𝑥
can be derived in the form of 𝑑𝑦 as shown 𝑛
𝑑𝑦 𝑑𝑧
⇒ (1 − 𝑛)𝑦 . =
below: 𝑑𝑥 𝑑𝑥

𝑑𝑥 Then equation (2) becomes

+ 𝑃𝑥 = 𝑄
𝑑𝑦
1 𝑑𝑧
. + 𝑃𝑧 = 𝑄
where P and Q is in the form of y or any constant (but 1 − 𝑛 𝑑𝑥
𝑑𝑧
not in x) ⇒ + (1 − 𝑛). 𝑃𝑧 = (1 − 𝑛). 𝑄
𝑑𝑥
For solving this type of LDE, we adopt the same
This equation is linear in z. And hence it can be solved
procedures as in the previous case.
by the methods discussed in previous section.
Step 1: First of all, find I.F. (Integrating factor)

𝐼. 𝐹. = 𝒆∫ 𝑷.𝒅𝒙 Solved Example

Step 2: Further hence, the general solution will be 𝑑𝑦 𝑢 𝑦2
given by 𝑥. 𝐼. 𝐹 = ∫ 𝑄. (𝐼. 𝐹. )𝑑𝑦 1. Solve the equation: 𝑑𝑥 + 𝑥 = 𝑥 2

Sol. Dividing both sides by 𝑦 2 , we get

 1 𝑑𝑦 1 1 Here, 𝑃 = 2𝑥 and 𝑄 = 𝑥𝑙𝑜𝑔𝑥
. 𝑑𝑥 + 𝑥𝑦 = 𝑥 2 ……… (1)
𝑦2

1 𝑑𝑧 1 𝑑𝑦 ∴ Integrating factor, 𝐼. 𝐹. = 𝑒∫ 𝑃. 𝑑𝑥 = 𝑒∫ 2𝑥. 𝑑𝑥 =

Now put 𝑦 = 𝑧, it gives = − 𝑦 2 . 𝑑𝑥 𝑒𝑙𝑜𝑔𝑥2 = 𝑥2
𝑑𝑥

2
Now the equation (1) reduces to 𝐼. 𝐹. = 𝒆∫ 𝑷.𝒅𝒙 = 𝒆∫ 𝟐/𝒙.𝒅𝒙 = 𝑒 ln 𝑥 = 𝑥 2

𝑑𝑧 𝑧 1 Hence the general solution will be

− + = 2
𝑑𝑥 𝑥 𝑥
𝑑𝑧 𝑧 1 𝑦(𝐼. 𝐹. ) = ∫ 𝑄. (𝐼. 𝐹. )𝑑𝑥
⇒ − =− 2
𝑑𝑥 𝑥 𝑥 ⇒ 𝑦𝑥 2 = ∫ (𝑥𝑙𝑜𝑔𝑥)𝑥 2 . 𝑑𝑥
which is the linear equation in the form of z.
⇒ 𝑦𝑥 2 = ∫ 𝑥 3 . 𝑙𝑜𝑔𝑥. 𝑑𝑥
1 1 ⇒ 𝑦𝑥 2 = 𝑙𝑜𝑔𝑥∫ 𝑥 3 𝑑𝑥 − ∫ (𝑑(𝑙𝑜𝑔𝑥)𝑑𝑥∫ 𝑥3𝑑𝑥)𝑑𝑥
Here, 𝑃 = − 𝑥 and 𝑄 = 𝑥 2
(Integrating by parts)
1
𝐼. 𝐹. = 𝒆∫ 𝑷.𝒅𝒙 = 𝒆∫ −𝟏/𝒙.𝒅𝒙 = 𝑒 ln 1/𝑥 =
𝑥 𝑥4 1 𝑥4
⇒ 𝑦𝑥 2 = 𝑙𝑜𝑔𝑥 − ∫ ( . ) 𝑑𝑥
4 𝑥 4
Hence the general solution will be 4
2
𝑥 ∫ 𝑥3
⇒ 𝑦𝑥 = 𝑙𝑜𝑔𝑥 − 𝑑𝑥
1 1 1 4 4
𝑧.
= ∫ − 2 . . 𝑑𝑥 4 4
𝑥 𝑥
𝑥 𝑥 𝑥 ⇒ 𝑦𝑥 2 = 𝑙𝑜𝑔𝑥 − +𝑐
𝑧 1 1 1 4 16
⇒ = ∫ − 3 . 𝑑𝑥 = − . 2 + 𝑐
𝑥 𝑥 2 𝑥
𝑧 1 𝑥 2 𝑙𝑜𝑔𝑥 𝑥 2
⇒ = 2+𝑐 ⇒𝑦= − +𝑐
𝑥 2𝑥 4 16
1 1 𝑥2
⇒ = 2+𝑐 ⇒𝑦= (4𝑙𝑜𝑔𝑥 − 1) + 𝑐
𝑦𝑥 2𝑥 16
⇒ 2𝑥 = 𝑦 + 2𝑥 2 𝑦𝑐
2𝑥
⇒𝑦=
1 + 𝑘𝑥 2

which is the required solution.

Laplace transform

A function only has a Laplace transform if the

𝑑𝑦 2 Laplace transform integral converges. The Laplace
2. Solve: the equation: 𝑥 𝑑𝑥 + 2𝑦 = 𝑥 𝑙𝑜𝑔𝑥
transform is defined as
Sol: ∞

𝑑𝑦 𝐿{𝑓(𝑡)} = ∫ 𝑒 −𝑠𝑡 𝑓(𝑡) 𝑑𝑡

Given Differential Equation is 𝑥 𝑑𝑥 + 2𝑦 = 𝑥 2 𝑙𝑜𝑔𝑥
0

𝑑𝑦
Dividing by x on both sides, we get 𝑑𝑥 + 2𝑦𝑥 = 𝑥𝑙𝑜𝑔𝑥 .
 3 √(t) √π/2s(3/2)
4 𝑠𝑖𝑛(𝑎𝑡) 𝑎
Standard notation: Where the notation is clear, we 𝑠2 + 𝑎2
will use an uppercase letter to indicate the Laplace 5 𝑡 𝑠𝑖𝑛(𝑎𝑡) 2𝑎𝑠
transform, e.g, L(f; s) = F(s). (𝑠 + 𝑎2 )2
2

6 𝑠𝑖𝑛(𝑎𝑡 + 𝑏) (𝑠 𝑠𝑖𝑛(𝑏) + 𝑎𝑐𝑜𝑠(𝑏)

The Laplace transform we defined is sometimes 𝑠 2 + 𝑎2
called the one-sided Laplace transform. There is a 7 𝑠𝑖𝑛ℎ(𝑎𝑡) 𝑎
two-sided version where the integral goes from −∞ to 𝑠 − 𝑎2
2

∞. 8 𝑒 (𝑎𝑡) 𝑠𝑖𝑛(𝑏𝑡) 𝑏
(𝑠 − 𝑎)2 + 𝑏 2
Properties of Laplace Transform 9 e𝑒 (𝑐𝑡) 𝑓(𝑡) 𝐹(𝑠 − 𝑐)
10 𝑓′(𝑡) 𝑠𝐹(𝑠) – 𝑓(0)
Some of the Laplace transformation properties are: 11 𝑒 (𝑎𝑡) 1
𝑠 − 𝑎
If f1 (t) ⟷ F1 (s) and [note: ⟷ implies Laplace 12 𝑡 𝑝 , 𝑎𝑡 𝑝 > −1 Γ(p+1)/s(p+1)
Transform] 13 𝑡 (𝑛−1/2) at 𝑛 = (1.3.5…(2n-1)√π)/(2n
1,2, .. s(n+1/2)
f2 (t) ⟷ F2 (s), then 14 𝑐𝑜𝑠(𝑎𝑡) 𝑠
𝑠 + 𝑎2
2
Linearity Property 𝐴 𝑓1 (𝑡) + 𝐵 𝑓2 (𝑡) 15 𝑡 𝑐𝑜𝑠(𝑎𝑡) 𝑠 2 − 𝑎2
⟷ 𝐴 𝐹1 (𝑠) (𝑠 2 + 𝑎2 )2
+ 𝐵 𝐹2 (𝑠) 16 𝑐𝑜𝑠(𝑎𝑡 + 𝑏) (𝑠𝐶𝑜𝑠(𝑏) − 𝑎𝑆𝑖𝑛(𝑏)
Frequency Shifting e 0 f(t)) ⟷ F(s – s0)
s t
𝑠 2 + 𝑎2
Property 17 𝑐𝑜𝑠ℎ(𝑎𝑡) 𝑠
Integration 𝑡
𝑠 2 − 𝑎2
∫ 𝑓(𝜆) 𝑑𝜆 ⟷ 1⁄𝑠 𝐹(𝑠) 18 𝑒 (𝑎𝑡) 𝑐𝑜𝑠(𝑏𝑡) 𝑠−𝑎
0 (𝑠 − 𝑎)2 + 𝑏 2
Multiplication by 𝑇 𝑓(𝑡) ⟷ (−𝑑 𝐹(𝑠) ⁄ 𝑑𝑠) 19 𝑡 𝑛 𝑓(𝑡) at 𝑛 = (−1)𝑛 𝐹 𝑛 𝑠
Time 1,2,3..
Complex Shift 𝑓(𝑡) 𝑒 −𝑎𝑡 ⟷ 𝐹(𝑠 + 𝑎) 20 𝑓”(𝑡) 𝑠 2 𝐹(𝑠) − 𝑠𝑓(0) − 𝑓′(0)
Property
Time Reversal 𝑓 (−𝑡) ⟷ 𝐹(−𝑠) Laplace Transform of Differential Equation
Property
Time Scaling 𝑓 (𝑡 ⁄ 𝑎) ⟷ 𝑎 𝐹(𝑎𝑠) The Laplace transform is a well-established
Property mathematical technique for solving a differential
equation. Many mathematical problems are solved
Laplace Transform Table using transformations. The idea is to transform the
problem into another problem that is easier to solve.
The following Laplace transform table helps to solve On the other side, the inverse transform is helpful to
the differential equations for different functions: calculate the solution to the given problem.

Sl f(t) L(f(t)) = F(s) For better understanding, let us solve a first-order

No. differential equation with the help of Laplace
1 1 1/s transformation,
2 tn at t = 1,2,3,… n!/s(n+1)
Consider 𝑦’ − 2𝑦 = 𝑒 3𝑥 and 𝑦(0) = −5. Find the School of lndustrial/Applied Sciences
value of 𝐿(𝑦).
Department of Mathematics/Statistics
First step of the equation can be solved with the help Second Semester Examination August 2017
of the linearity equation:
MTH 312: Advanced Calculus (Morning
3𝑥
𝐿(𝑦’ – 2𝑦] = 𝐿(𝑒 ) Programme)

1 Instruction: Answer any four questions Time

𝐿(𝑦’)– 𝐿(2𝑦) = Allowed: 2/ hours.
𝑠−3
1
(because 𝐿(𝑒 𝑎𝑥 ) = 𝑠−𝑎
1a) Using the method of separation of variables, solve
1 the following equation
𝐿(𝑦’)– 2𝑠(𝑦) =
𝑠−3 𝑑𝑦 𝑥+1
= 2
𝑑𝑥 𝑥 𝑦
1
𝐿(𝑦)– 𝑦(0)– 2𝐿(𝑦) =
𝑠−3 1b) A body at 1200c is placed in a room which has a
constant temperature of 500c.The rate of cooling at
(Using Linearity property of the Laplace transform) any time is proportional to the temperature
1
differential between the body and room temperatures.
𝐿(𝑦)(𝑠 − 2) + 5 = 𝑠−3 (Use value of y(0) ie -5 if at the end of 20 minutes, the temperature of the
(given)) body is 800c, find how long it will take the body to
reduce to 650c.
1
𝐿(𝑦)(𝑠 − 2) = –5 2a) Find the value of k for which the equation is exact
𝑠−3
and then solve
−5𝑠+16
𝐿(𝑦) = (𝑠−2)(𝑠−3) …..(1) (𝑘𝑥𝑦 + 𝑥 2 )𝑑𝑥 + (𝑥 2 + 𝑦)𝑑𝑦 = 0

−5𝑠+16 6 1 2b) solve the P.D.E. by Direct Integration

here (𝑠−2)(𝑠−3) can be written as − 𝑠−2 + 𝑠−3 using
partial fraction method 𝜕 2𝑧
= 5𝑥 2 𝑦 2
𝜕𝑥𝜕𝑦
6 1
(1) Implies 𝐿(𝑦) = − 𝑠−2 + 𝑠−3 Given that 𝑧(𝑥, 0) = 𝑥 2 and 𝑧(0, 𝑦) = 𝑦 2

𝐿(𝑦) = −6𝑒 2𝑥 + 𝑒 3𝑥 3a) Solve the following P.D.E. by the method of

separation of variables
𝜕𝑢 𝜕𝑢
4 + = 3𝑢
𝜕𝑥 𝜕𝑡
3b) Find the relative extremum of the function
𝑓(𝑥, 𝑦) = 𝑥 2 − 2𝑦 3 + 6𝑦
4a) Using the method of undetermined coefficients, 𝑦2
find the solution of the equation = ln 𝑥 + ln 𝑥 2 + 𝐶
2
𝑦 ′′ + 𝑦 ′ = 4𝑥 2 𝑦 2 = 2 ln 𝑥 + 2 ln 𝑥 2 + 2𝐶
4b) Find the inverse transform of the following 𝑦 = √2 ln 𝑥 + 2 ln 𝑥 2 + 2𝐶
𝑆+5
(𝑠 + 1)(𝑠 + 3) 1(b) From Newton’s law of cooling, 𝑞𝑓 = 𝑞𝑖 𝑒 −𝑘𝑡

5a) Solve the following using the D-operator method Now, for the interval in which temperature falls from
120 to 80oC.
2𝑦 ′′ + 5𝑦 ′ − 12𝑦 = 2𝑥𝑒 2𝑥
5b) Solve the following homogeneous equation (80– 50) = (120 – 50) 𝑒 −𝑘.20

𝑑𝑦 4𝑦 − 3𝑥 30 3
= 𝑒 −20𝑘 = =
𝑑𝑥 2𝑥 − 𝑦 70 7
6a) Solve the following initial value problem by the −𝑙𝑛( )
3
7
Laplace transform method 𝑘 = . . . . (a)
20

𝑦 ′′ − 3𝑦 ′ + 2𝑦 = 2𝑒 −4𝑥
Now, for the next interval;
𝑦(0) = 0, 𝑦′(0) = 1
(65 – 50) = (80 – 50)𝑒 −𝑘𝑡
6b) Using the method of variation of parameters,
obtain the solution of the following equation: 𝑦 ′′ − 15 1
3𝑦′ + 2𝑦 = 𝑥 2 𝑒 −𝑘𝑡 = =
30 2
1
−𝑙𝑛( )
2
𝑘= . . . . (b)
Solutions 𝑡

𝑑𝑦 𝑥+1 From equation (a) and (b);

1(a) =
𝑑𝑥 𝑥2𝑦

Solve using separable variables 3 1

−𝑙𝑛 (7) −𝑙𝑛 (2)
=
𝑥+1 20 𝑡
𝑦𝑑𝑦 = 𝑑𝑥
𝑥2
1
𝑥 1 𝑙𝑛 (2)
𝑦𝑑𝑦 = ( 2 + 2 ) 𝑑𝑥 𝑡 = 20 × [ ] = 16.36 𝑚𝑖𝑛.
𝑥 𝑥 3
𝑙𝑛 (7)
1 1
𝑦𝑑𝑦 = ( + 2 ) 𝑑𝑥
𝑥 𝑥 2(a) (𝑘𝑥𝑦 + 𝑥 2 )𝑑𝑥 + (𝑥 2 + 𝑦)𝑑𝑦 = 0
By integrating both sides
Compare the given equation with 𝑀𝑑𝑥 + 𝑁𝑑𝑦 = 0
1 1 𝜕𝑀 𝜕𝑁
∫ 𝑦 𝑑𝑦 = ∫ ( + 2 ) 𝑑𝑥 and find out M and N. Then find out 𝜕𝑦 and 𝜕𝑥 .
𝑥 𝑥
 𝑀 = 𝑘𝑥𝑦 + 𝑥 2 Integrating, 𝑋 = 𝑐1 𝑒 (𝑘+3)𝑥/4
𝜕𝑀
= 𝑘𝑥 and 𝑌 = 𝑐2 𝑒 −𝑘𝑦
𝜕𝑦
Also, 𝑁 = 𝑥2 + 𝑦 So, 𝑢 = 𝑐1 𝑐2 𝑒 (𝑘+3)𝑥/4−𝑘𝑦

𝜕𝑁 Let,
(𝑘+3)
= 𝜆; 𝐾 = 4𝜆 − 3
= 2𝑥 4
𝜕𝑥
𝜕𝑀 𝜕𝑁 The characteristics solution
If = , the given equation is exact.
𝜕𝑦 𝜕𝑥
𝑦𝑝 = 𝐴1 𝑒 𝜆𝑥−(4𝑦−3)𝑦
∴ 𝑘𝑥 = 2𝑥
𝑥
𝑘=2 𝑦𝑝 = 𝑒 3𝜆 𝐴1 𝑒 4𝜆[4−𝑦]
𝜕2 𝑧 𝑥
2(b) = 5𝑥 2 𝑦 2 𝑦𝑝 = 𝑒 3𝜆 𝐴1 𝑒 𝑢1 [4−𝑦]
𝜕𝑥𝜕𝑦

𝜕2 𝑧
Consider, The general solution;
𝜕𝑥𝜕𝑦

𝑥 𝑥
𝜕 2𝑧 𝜕 𝜕𝑧
= ( ) 𝑦 = 𝑒 3𝜆 𝐴1 𝑒 𝑢1 [4−𝑦] + 𝑒 3𝜆 𝐴2 𝑒 𝑢2 [4−𝑦]
𝜕𝑥𝜕𝑦 𝜕𝑥 𝜕𝑦
𝜕𝑧 𝑦(0) = 𝑒 3𝜆 𝐴1 𝑒 −𝑢1 𝑦 + 𝑒 3𝜆 𝐴2 𝑒 −𝑢2 𝑦
= 10𝑥 2 𝑦
𝜕𝑦
Now I check this against the initial value given as
𝜕 𝜕𝑧 𝜕 𝑢 = 3𝑒 −𝑦 − 𝑒 −5𝑦 when 𝑥 = 0
( )= (10𝑥 2 𝑦) = 20𝑥𝑦
𝜕𝑥 𝜕𝑦 𝜕𝑥
The final answer also (obviously) has two
𝜕 2𝑧 components 𝑢 = 3𝑒 𝑥−𝑦 − 𝑒 2𝑥−5𝑦
𝑇ℎ𝑢𝑠, = 20𝑥𝑦
𝜕𝑥𝜕𝑦
3(b) 𝑓(𝑥, 𝑦) = 𝑥 2 − 2𝑦 3 + 6𝑦
3(a) Solve the following P.D.E. by the method of
separation of variables
Our main task is to seek and solve the relative
𝜕𝑢 𝜕𝑢 extrema if it exist.
4 + = 3𝑢
𝜕𝑥 𝜕𝑡
We know that relative extrema occurs in conjunction
Let 𝑢 = 𝑋(𝑥)𝑌(𝑦). So, 4𝑋′𝑌 + 𝑋𝑌′ = 3𝑋𝑌 with the following two condition,

4𝑋′−3𝑋 −𝑌 ′ i. ▽ 𝑓 = 0 which implies two of the following partial

Separating the variables, = =𝑘
𝑋 𝑌 𝜕𝑓 𝜕𝑓
derivatives are true 𝜕𝑥 = 0 and 𝜕𝑦 = 0
3+𝑘
(say) 𝑋′ − 𝑋=0
4
ii. ▽ 𝑓 does not exist.
𝑌′ + 𝑘𝑌 = 0
The same idea is taken from single variable calculus!
So, let's proceed. 𝐷 > 0 it must be a relative extrema (𝑓𝑥𝑥 > 0 relative
minimum 𝑓𝑥𝑥 < 0 relative maximum)
Differentiate partially with respect to x
𝐷 = 0 it can be either a saddle point or a relative
-variable keeping y extrema.

-constant. For second order partial derivative,

If, 𝑓(𝑥, 𝑦) = 𝑥 2 − 2𝑦 3 + 6𝑦 𝜕 2𝑓
=2
𝜕𝑓 𝜕𝑥 2
,then 𝜕𝑥 = 2𝑥
𝜕 2𝑓
= −12𝑦
Onto partial derivative with respect to y 𝜕𝑦 2

-variable and x Mixed partial derivative,

-constant, 𝜕 2𝑓
=0
𝜕𝑓 𝜕𝑥𝜕𝑦
If, 𝑓(𝑥, 𝑦) = 𝑥 2 − 2𝑦 3 + 6𝑦, then 𝜕𝑦 = −6𝑦 2 + 6
𝜕 2𝑓
𝜕𝑓 =0
Now let's set = 0; 2𝑥 = 0 𝜕𝑦𝜕𝑥
𝜕𝑥

With fundamental theorem of algebra: an equation of Test with the first set of order pairs,
x unknown with power n−th has exactly nth solutions
and some may repeat. I) (0,1)

So 𝑥 = 0 is our solutions. 𝐷 = (2)(−12(1)) − 0

−6𝑦 2 + 6 = 0 𝐷 = −24 < 0

𝑦 = ±1 it is our solution. So it must be a saddle point

We have three set of order pairs now. (0, ±1) II) (0, −1)

Let me introduce you with this equation, 𝐷 = (2)(−12(−1)) − 0

2 𝐷 = 24 > 0
𝜕 2𝑓 𝜕 2𝑓 𝜕 2𝑓
𝐷 = 2. 2−( )
𝜕𝑥 𝜕𝑦 𝜕𝑥𝜕𝑦 So, it is relative minimum

This equation says that if 4(a) 𝑦 ′′ + 𝑦 ′ = 4𝑥 2

Solution:
𝐷 < 0 it must be a saddle point
𝑦 ′′ + 𝑦 ′ = 4𝑥 2
First obtain complementary solution for the
Corresponding homogeneous equation 4(b) Find the inverse Laplace transform of
𝑦 ′′ + 𝑦 ′ = 0 s+4
We set 𝑦 = 𝑒 𝑚𝑥 s(s-1)(62+4)
We obtain an auxiliary equation Solution:

𝑚2 + 𝑚 = 0 Let us first resolve into partial fraction

𝑚(𝑚 + 1) = 0 s+4 A B Cs+D ()

Factorizing we have; s(s-1)(s2 +4) S S s2 +4 )

(𝑚 − 0)(𝑚 + 1) = 0 s+4 A(s - 1)(s+4) + Bs(s2 +4) + Cs(s-1) (**)

𝑚 = 0, −1 set s = 0, we get 4 -4A A = -1

𝑦𝑐 = 𝐶1 𝑒 0𝑥 + 𝐶2 𝑒 −𝑥 set s= 1, we get 5 B 1 (1+4) > B = 1

𝑦𝑐 = 𝐶1 + 𝐶2 𝑒 −𝑥 Equating coefficients of s* on both sides of (*), we

get
(Observe that 𝑒 −𝑥 appeared in the non-
0 A+B+C * 0 = -1 +1+C C= 0
homogeneous part which is 4𝑥 2 in the case)
Equating coefficients ofs on both sides of (*), we get
For the y see the UC function is x and the UC set is
1 4A+4B- D 1=-4+4- D D =-1
{𝑥 2 , 𝑥, 1}
On substituting the values of A, B, C and D in (*), we
Set 𝑦𝑝 = 𝐴𝑥 2 + 𝐵𝑥 + 𝐶, for undermined coefficient get
A, B and C that are to be determined
St4 1 1 1
𝑦𝑝 ′ = 2𝐴𝑥 + 𝐵, 𝑦𝑝 ′′ = 2𝐴
--t-
′ ′′
Substitute for 𝑦𝑝 , 𝑦𝑝 , 𝑦𝑝 in eqn(i), to get
s(s-1)(s2 +4)sS-1 s2+4
2
2𝐴 + 2𝐴𝑥 + 𝐵 = 4𝑥 s+4 1 1 1
2 2
0𝑥 + 2𝐴𝑥 + 2𝐴 + 𝐵 = 4𝑥 + 0𝑥 Ls(s - 1)(s2+4) S s-1 s2+4
Equating the corresponding coefficient we have 11
2𝐴 = 0; 𝐴 = 0 -L1(-|+L1{ - -1 s2+4 1
2𝐴 + 𝐵 = 0; 2(0) + 𝐵 = 0 Ls- 1/
𝐵=0 1
Hence, 𝑦𝑝 = 0 =-1+e-sin 2t
The general solution 𝑦 = 𝑦𝑐 + 𝑦𝑝 2
𝑦 = 𝐶1 + 𝐶2 𝑒 −𝑥
6(a) First, we apply the Laplace transform on both
sides:

𝐿{𝑦 ′′ } + 3𝐿{𝑦 ′ } + 2𝐿{𝑦} = 𝐿{2𝑒 −4𝑥 }.

Expanding this yields

(𝑠 2 𝑌(𝑠) − 𝑦(0) ⋅ 𝑌(𝑠) − 𝑦 ′ (0)) + 3(𝑠𝑌(𝑠) − 𝑦(0))

1
+ 2𝑌(𝑠) = 2 ( )
𝑠−4

By the Product Rule for derivatives, we have

2
(𝑠 2 𝑌(𝑠) − 1) − (𝑠𝑌 ′ (𝑠) + 𝑌(𝑠)) + 2𝑌(𝑠) = ,
𝑠−4

and thus

−𝑠𝑌′(𝑠) + (𝑠2 − 3)𝑌(𝑠) = 𝑠2 + 6𝑠 − 1𝑠2.