Course Handout

MAT 201 : Partial Differential Equations and

Complex Analysis
Module 1 : Partial Differential Equations
Binu R, Eva Cherian
Department of Mathematics

Rajagiri School of Engineering and Technology

Pre requisites
(i) Calculating partial derivatives

(ii) Evaluating partial derivatives at a point

(iii) Chain rule for partial derivatives

(iv) Partial integration (integrating with respect to only one variable)

(v) Basic integration techniques and formulae

1 Partial differential equations

Definition 1.1. An equation involving an unknown function z of two or more variables

and its partial derivatives is called a partial differential equation (PDE).

1.1 Notations used

Throughout this module z will be used to denote the unknown function z(x, y) of the
2 independent variables x, y.
We will also use p, q to denote the partial derivatives with respect to x, and y
respectively, i.e,
∂z ∂z
p= ,q =
∂x ∂y

1.2 Some well known PDEs

The following are some well known partial differential equations from physics and
engineering:

(i) Laplace’s equation

∆z = zxx + zyy = 0

(ii) Heat (or diffusion ) equation

zt − ∆z = 0

(iii) Wave equation

ztt − ∆z = 0

(iv) Schrödinger’s equation

izt + ∆z = 0

(v) Telegraph equation

ztt + dzt − zxx = 0

(vi) Airy’s equation

zt + zxxx = 0

(vii) Beam equation

ztt + zxxxx = 0
1.3 Order and degree of a PDE

Definition 1.2. The order of the highest partial derivative in the equation is called the
order of the PDE.
∂z ∂z ∂2z ∂z 3


For example, z = x ∂x + y ∂y has order 1, and ∂x∂y
= ∂x
has order 2.

Definition 1.3. The degree of the highest order partial derivative is called the degree
of the PDE.
∂z 3


Both equations in the above example have degree 1. But z = ∂x
has degree 3.

2 Formation of partial differential equations

First we look at how to form a PDE from a given relation.
A relation involving 3 variables might include arbitrary constants or arbitrary func-
tions. To form a PDE from such a relation, we have to eliminate the arbitrary quantities
in the relation by taking the partial derivatives.
Partial differential equations can be formed using the following two methods from
a relation f(x,y,z,a,b)=0 involving three or more variables

(i) Eliminating arbitrary constants

(ii) Eliminating arbitrary function

When we get a relation involving arbitrary constants / functions, we don’t know

what the order of the differential equation will be. This means we don’t know what
order partial derivatives to take to eliminate the arbitrary constants / functions.
Luckily, we have a handy rule of thumb for how many times to differentiate:

(i) If the number of arbitrary constants equals the number of independent variables,
then the PDE is of first order.

(ii) If the number of arbitrary constants is greater than the number of independent
variables, then the PDE is of second or higher order.

(iii) If the PDE is formed by eliminating arbitrary functions, then the order of the
PDE is in general equal to the number of arbitrary functions eliminated.
2.1 Elimination of arbitrary constants
For such problems, we find the partial derivatives of z to eliminate the arbitrary con-
stants.
x2 y2
Example 1. Derive a PDE from the relation 2z = a2
+ b2
.

Solution: The given relation has 2 arbitrary constants a and b. There are 2 independent
variables x, y. So by our rule of thumb, the resulting PDE should be of first order.
Find the first order partial derivatives of z.

∂z 2x ∂z 2y
2 = 2,2 = 2
∂x a ∂y b
Simplifying the above expressions, we find
∂z x ∂z y
p= = 2,q = = 2
∂x a ∂y b
.
Now we can write both arbitrary constants in terms of p, q, x, andy as

1 p 1 q
2
= , 2 =
a x b y
Finally, substitute for the constants in the original relation to find

2z = px + qy

Example 2. Find the differential equation of all planes which are at a constant distance
a from the origin.

Solution: A plane at distance a from the origin will have an equation of the form
lx + my + nz = a, where l, m, n are the direction cosines of the normal from the origin
to the plane.
This relation has 3 arbitrary constants l, m, n. But these constants are the direction
√
cosines, and so satisfy the property : l2 + m2 + n2 = 1, or n = 1 − l2 − m2 . So the
equation of the plane can be written as
√
lx + my + 1 − l2 − m2 z = a (1)

Observe that the relation now has just 2 arbitrary constants, and 2 independent
variables. So the PDE satisfied by this relation should be of first order.
 Take the partial derivatives of (1) with respect to x and y. We get the following
equations:

√
l+ 1 − l2 − m2 p = 0 (2)
√
m + 1 − l2 − m2 q = 0 (3)
Next we eliminate the arbitrary constants l, m from equations (1), (2), (3).
Squaring and adding (2) and (3),

l2 + m2 = (1 − l2 − m2 )(p2 + q 2 )

Expand the right hand side and simplify to find

l2 + m2 = (p2 + q 2 ) − (l2 + m2 )(p2 + q 2 )

p2 + q 2
l2 + m2 =
1 + p2 + q 2
Hence,
1
1 − l2 − m2 =
1 + p2 + q 2
We can substitute this expression in (2), (3) to find l = − √ p 2 2 , m = − √ q 2 2 .
1+p +q 1+p +q
√
2 2
Finally, substitute for l, m, 1 − l − m in (1)

px qy z
−p −p +p =a
1 + p2 + q 2 1 + p2 + q 2 1 + p2 + q 2
Rearranging this equation gives us the required PDE

p
z = px + qy + a 1 + p2 + q 2
Exercise
Form PDEs from the following equations by eliminating arbitrary constants.

(i) z = ax + by + ab

(ii) z = ax + a2 y 2 + b

(iii) z = (x2 + a)(x2 + b)

x2 y2 z2
(iv) a2
+ b2
+ c2
=1

(v) (x − a)2 + (y − b)2 = z 2 cot2 α, where α is a fixed constant.

2.2 Elimination of arbitrary functions
While eliminating arbitrary functions also, we will directly take the partial derivatives
of z. In problems where this is not possible, we will use the chain rule for partial
derivatives.

Example 3. Form the PDE by eliminating arbitrary functions from z = f (x + at) +

g(x − at), where a is a fixed constant.

Solution: In this relation, the independent variables are x, t and there are 2 arbitrary
functions f, g.
Again, by the rule of thumb, we have to differentiate as many times as the number
of arbitrary functions - in this case 2.
∂2z ∂2z
Evaluate ,
∂x2 ∂y 2
.

∂z ∂z
= f 0 (x + at) + g 0 (x − at), = af 0 (x + at) − ag 0 (x − at)
∂x ∂t
(By applying the chain rule)

∂ 2z 00 00 ∂ 2z
2
= f (x + at) + g (x − at), 2
= a2 f 00 (x + at) + a2 g 00 (x − at)
∂x ∂t
Now by comparing the second order derivatives we can write

∂ 2z 2
2∂ z
= a
∂t2 ∂x2
Note that this is just the wave equation from the examples.

Example 4. Form the PDE by eliminating arbitrary function from z = (x+y)φ(x2 −y 2 )

Solution: The given relation has only 1 arbitrary function, so the required PDE should
be of first order.
The partial derivatives of z are:
∂z
p= = (x + y)φ0 (x2 − y 2 ).2x + φ(x2 − y 2 )
∂x
∂z
q= = (x + y)φ0 (x2 − y 2 ).(−2y) + φ(x2 − y 2 )
∂y
(Using product rule and the chain rule)
 z
From the original relation, we have φ(x2 − y 2 ) = x+y
. Substituting for φ(x2 − y 2 )
in the partial derivatives and rearranging gives us the following equations:

z
p− = 2x(x + y)φ0 (x2 − y 2 )
x+y
z
q− = −2y(x + y)φ0 (x2 − y 2 )
x+y
Remember that to get the required PDE, we need to eliminate the arbitrary function
φ. It is easy to see that dividing the above 2 equations will eliminate the φ term.
The division gives
p − z/(x + y) x
=−
q − z/(x + y) y
Taking the LCM and simplifying gives

[p(x + y) − z]y + [q(x + y) − z]x = 0

(x + y)(py + qx) − z(x + y) = 0

Dividing throughout by x + y, the required PDE is

py + qx = z

Example 5. Form the PDE by eliminating arbitrary functions from the given relation:
f (x2 + y 2 , z − xy) = 0

Solution: For questions like these, we can’t evaluate the partial derivatives of z directly.
To solve, let u = x2 + y 2 , v = z − xy, so that the relation is f (u, v) = 0.
Now apply the chain rule to find the partial derivative of the above function with
respect to x.
   
∂f ∂u ∂x ∂u ∂z ∂f ∂v ∂x ∂v ∂z
+ + + =0
∂u ∂x ∂x ∂z ∂x ∂v ∂x ∂x ∂z ∂x
∂z
If we now substitute for the appropriate partial derivatives of u, v and let p = ∂x
,
then

∂f ∂f
(2x) + (−y + p) = 0 (4)
∂u ∂v
 Similarly, we can evaluate the partial derivative with respect to y and get the
equation

∂f ∂f
(2y) + (−x + q) = 0 (5)
∂u ∂v
Now that we have 2 equations involving ∂f , ∂f , we can eliminate these partial
∂u ∂v
derivatives from the equations:
Multiply the first equation by y and the second equation by x. If we then subtract
these 2 new equations, we find

∂f
(−y 2 + py + x2 − qx) = 0
∂v
∂f
Then either ∂v
= 0, or (−y 2 + py + x2 − qx) = 0. But the partial derivative cannot
identically be equal to zero (as f is arbitrary, we cannot say anything specific about
its partial derivatives).
Therefore,
(−y 2 + py + x2 − qx) = 0

So the required PDE is

qx − py = x2 − y 2

Exercise
Form PDEs from the following equations by eliminating arbitrary functions.

(i) z = f1 (x)f2 (y)

(ii) z = emy φ(x − y)

1


(iii) z = y 2 = 2f x
+ log y

(iv) z = xf1 (x + t) + f2 (x + t)
y


(v) z = f x

3 Solution of Partial Differential Equations

Definition 3.1. The solution f(x,y,z,a,b)=0 of a first order differential equation
which contains two arbitrary constants is called a complete Integral.
Definition 3.2. A solution obtained from the complete integral which contains by as-
signing particular values to the arbitrary constants is called a particular Integral.

Metods for Solution of Partial Differential Equations

• Method 1 :- Equations Solvable by Direct Integration

• Method 2 :- Linear Equations of the First Order (Lagrange’s Linear Equation)

• Method 3 :- Non Linear Equations of the First Order (Charpit’s Method)

• Method 4 :- Method of Separation of Variables

3.1 Method 1

Equations Solvable by Direct Integration

In this method PDE can be solved by direct integration and in place
of the usual constant of integration we use arbitrary functions of the
variable held fixed.
∂ 3z
Example 6. Solve + 18xy 2 + sin(2x − y) = 0
∂x2 ∂y
Solution: Integrating the given PDE with respect to x (keeping y fixed)

∂ 2z 1
+ 9x2 y 2 − cos(2x − y) = f (y) (6)
∂x∂y 2
[For the constant of integration, we use arbitrary functions of variable
held fixed]
Integrating the equation (1) with respect to x (keeping y fixed)
∂z 1
+ 3x3 y 2 − sin(2x − y) = xf (y) + g(y) (7)
∂y 4
Integrate equation (2) with respect to y (keeping x fixed)
Z Z
1
z + x3 y 3 − cos(2x − y) = x f (y)dy + g(y)dy + w(x) (8)
4
 R R
Then we substitute f (y)dy = u(y) and g(y)dy = v(y)
1
z + x3 y 3 − cos(2x − y) = xu(y) + v(y) + w(x) (9)
4
1
⇒ z= cos(2x − y) − x3 y 3 + xu(y) + v(y) + w(x)
4
where u, v and w are arbitrary function.
∂ 2z
Example 7. Solve = xy
∂x2
Solution: Integrate the given PDE with respect to x (keeping y fixed)
∂z x2
= y + f (y) (10)
∂x 2
Integrate equation (5) with respect to x (keeping y fixed)
x3
Z
z = y + f (y) dx + g(y) (11)
6
1
⇒ z = x3 y + xf (y) + g(y)
6
where f and g are arbitrary function.
∂ 2z
Example 8. Solve 2
+ z = 0, given that when x = 0, z = ey and
∂x
∂z
=1
∂x
Soution: Here z is a function of x only. So we can treat the given PDE
as ordinary differential equation (ODE) but constants can depend on y.
⇒ (D2 + 1)z = 0
Auxiliary equation is m2 + 1 = 0 ⇒ m = i, −i

⇒ z = f (y) sin x + g(y) cos y (12)

P ut x = 0 in the equation(7) ( Applying the given conditions )

⇒ ey = f (y) sin 0 + g(y) cos 0
⇒ ey = g(y) and substitute in equation (7)

⇒ z = f (y)sin x + ey sin y (13)

∂z
Put = 1 when x = 0 in equation (8)
∂x
∂z
= f (y) cos x + 0 (14)
∂x
W hen x = 0 ⇒ 1 = f (y)

Hence the desired solution is

z = f (y) sin x + g(y) cos y (15)

z = sin x + ey cos y (16)

∂ 3z
Example 9. Solve = cos (2x + 3y)
∂x2 ∂y
Solution: Integrating the given PDE with respect to y (keeping x fixed)

∂ 2z 1
= sin (2x + 3y) + f (x) (17)
∂x2 3
Integrating the equation (12) with respect to x
Z
∂z 1
= − cos (2x + 3y) + f (x)dx + g(y) (18)
∂x 6
Integrating the equation (13) with respect to x
Z Z Z
1
z = − sin (2x + 3y) + f (x)dxdx + g(y) dx + h(y) (19)
12
1
z = − sin (2x + 3y) + u(x) + xg(y) + h(y) (20)
12
RR
where u(x) = f (x)dxdx is an arbitrary function of x
Exercise
Solve the following equations by direct integration:
 ∂ 2z x
(i) = +a
∂x∂y y
∂ 2z 2 ∂z ∂z
(ii) = a z given that when x = 0, = a siny and =0
∂x2 ∂x ∂y
∂ 2z
(iii) = e−t cosx
∂x ∂t
∂ 2z ∂z
(iv) 2
= z, given that when y = 0, z = e x
and = e−x
∂y ∂y
∂ 2z
(v) = x2 y
∂x∂y

3.2 Method 2

Linear Equations of the First Order

Definition 3.3. If the unknown function and its partial derivatives ap-
pear to the power of 1 in a PDE then it is called First Order Linear
Partial Differential Equation. Otherwise it is called non-linear PDE.

Remark 1. If z = f (x, y) then we shall employ the following notation

for the partial derivatives.

∂z ∂z ∂ 2z ∂ 2z ∂ 2z
= p, = q, = r, = s, =t
∂x ∂y ∂x2 ∂x∂y ∂y 2
Example 10. (i) px + yq = 3z- First order linear PDE

(ii) xp − yq = y 2 − x2 -First order linear PDE

(iii) p2 + q 2 = 1- Non linear

(iv) pq + p + q = 0-Non linear

Definition 3.4. A linear PDE of the form P p + Qq = R commonly

known as Lagrange’s linear equation where P, Q and R are functions
of x, y and z

Procedure for the solution of P p + Qq = R

To solve the equation P p + Qq = R

Step 1 Identify the terms P, Q and R from P p + Qq = R

dx dy
Step 2 Form the subsidiary equations (Auxilary equation) = =
P Q
dz
R
Step 3 Solve the auxilary equation by the following method to get two
independent solutions u = a and v = b where a and b are arbitrary
constants

(i) Grouping or Rearranging

(ii) Method of Lagrange multipliers
(iii) Both (i) and (ii)

Step 4 Write the complete solution as φ(u, v) = 0 or u = f (v)

Case 1. Grouping

We may be able to group two fractions from auxiliary equation and cancel
the common terms to get an equation in only two variables. On integrat-
ing the differential equation in only two variables by any methods, we
shall obtain one of the relations u in the general solution. This method
may be repeated to give another relation v with the help od another two
fractions in auxiliary equation.
y2z
Example 11. Solve p + xzq = y 2
x
Solution:-
y2z
p + xzq = y 2 (21)
x
Equation (16) × x ⇒ y zp + x2 zq = xy 2
2

Step 1 P = y 2 z, Q = x2 z, R = xy 2
dx dy dz dx dy dz
Step 2 Auxiliary equation- = = ⇒ 2 = 2 = 2
P Q R y z xz xy
Step 3 Solve the auxiliary equation by grouping
dx dy
Taking first two fractions 2 = 2
y z xz
⇒ x2 dx = y 2 dy
x3 y3 a
On integrating = + ⇒ x3 − y 3 = a
3 3 3
3 3
Set u = x − y = a
dx dz
Taking first and third fractions 2 = 2
y z xy
⇒ xdx = zdz
x2 z2 b
On integrating = + ⇒ x2 − z 2 = b
2 2 2
Set v = x2 − z 2 = b

Step 4 General equation is φ(u, v) = 0 or u = f (v)

i.e. φ(x3 − y 3 , x2 − z 2 ) = 0 or x3 − y 3 = f (x2 − z 2 )

Example 12. Solve p tanx + q tany = tanz

Solution:-

Step 1 P = tanx, Q = tany, R = tanz

 dx dy dz dx dy dz
Step 2 Auxiliary equation- = = ⇒ = =
P Q R tanx tany tanz
Step 3 Solve the auxiliary equation by grouping
dx dy
Taking first two fractions =
tanx tany
⇒ cotx dx = coty dy

Result
R R cos x
cotx dx = dx = log (sinx) + c
sin x

On integrating ⇒ log sin x = log sin y + a

sinx
⇒ log sin x − log sin y = a ⇒ log =a
siny
sin x
Set u = log =a
sin y
dy dz
Taking second and third fractions =
tany tanz
⇒ coty dy = cotz dz
On integrating ⇒ log sin y = log sin z + b
sin y
⇒ log sin y − log sin z = b ⇒ log =b
sin z
sin y
Set v = log =b
sin z
Step 4 General
 equation is φ(u,v) = 0 or u = f (v)  
sin x sin y sin x sin y
i.e. φ log , log = 0 or log = f log
sin y sin z sin y sin z
Example 13. Solve (p − q)z = z 2 + (x + y)2

Solution:- Given PDE can be rearranged as pz − qz = z 2 + (x + y)2

Step 1 P = z, Q = z, R = z 2 + (x + y)2
dx dy dz dx dy dz
Step 2 Auxiliary equation- = = ⇒ = = 2
P Q R z −z z + (x + y)2
Step 3 Solve the auxiliary equation by grouping
dx dy
Taking first two fractions =
z −z
⇒ dx = −dy
On integrating ⇒ x = −y + a = x + y = a
set u = x + y = a
dx dz
Taking first and third fractions = 2
z z + (x + y)2
put x + y = a in the above equation
dx dz
⇒ = 2
z z + a2
z dz 1 2z dz
⇒ dx = 2 ⇒ dx =
z + a2 2 z 2 + a2

Result Z 0
f (x)
dx = log f (x) + c
f (x)

1 0
On integrating ⇒ x = log (z 2 + a2 ) + b
2
⇒ 2x = log (z 2 + a2 ) + b
⇒ 2x − log (z 2 + (x + y)2 ) = b [since x + y = a]
set u = 2x − log (z 2 + (x + y)2 ) = b

Step 4 General equation is φ(u, v) = 0 or u = f (v)

i.e. φ(x + y, 2x − log (z 2 + (x + y)2 ) = 0 or
x + y = f (2x − log (z 2 + (x + y)2 )
 Note
Properties of ratio
a c e a+c+e a c e
(i) If = = then = = =
b d f b+d+f b d f
a c e la + mc + ne a c e
(ii) = = then = = = where
b d f lb + md + nf b d f
l, m and n are constants or variables

Example 14. Solve (x2 − y 2 − z 2 )p + 2xyq = 2xz

Solution:-

Step 1 P = x2 − y 2 − z 2 , Q = 2xy, R = 2xz

dx dy dz dx dy dz
Step 2 = = ⇒ 2 = =
P Q R x − y2 − z2 2xy 2xz
Step 3 Solve the equation in step 2 by grouping
dy dz
Taking second and third fractions =
2xy 2xz
dy dz
⇒ = , on integrating ⇒ log y = log z + log a
y z
y
⇒ log y − log z = log a ⇒ log = log a
y y z
⇒ = a. Set u = = a
z z
Apply the property of ratio
Multiply each ratio in auxiliary equation by x, y and z respectively

xdx + ydy + zdz xdx + ydy + zdz xdx + ydy + zdz

i.e. = =
x3 − xy 2 − xz 2 + 2xy 2 + 2xz 2 x3 + xy 2 + xz 2 x(x2 + y 2 + z 2 )
Equating the new ratio in to third one

xdx + ydy + zdz dz 2(xdx + ydy + zdz) dz

= ⇒ =
x(x2 + y 2 + z 2 ) 2xz x2 + y 2 + z 2 z
 Result Z 0
f (x)
dx = log f (x) + c
f (x)

On integrating ⇒ log(x2 + y 2 + z 2 ) = log z + log b

⇒ log(x2 + y 2 + z 2 ) − log z = log b
x2 + y 2 + z 2
⇒ log = log b
z
x2 + y 2 + z 2 x2 + y 2 + z 2
⇒ = b ⇒ set v = =b
z z
Step 4 General equation isφ(u, v) = 0 or u
 =2 f (v)2
2 2 2
x + y + z2
 
y x +y +z y
i.e. φ , = 0 or = f
z z z z
Example 15. Solve (x2 − yz)p + (y 2 − zx)q = z 2 − xy

Solution:-

Step 1 P = x2 − yz, Q = y 2 − zx, R = z 2 − xy

dx dy dz
Step 2 Auxiliary equation-
= =
P Q R
dx dy dz
⇒ 2 = 2 = 2
x − yz y − zx z − xy
Step 3 Solve the auxiliary equation using the properties of ratios by
grouping
Taking the difference between first two and last two frac-
tions and equating them
dx − dy dy − dz
=
x2 − yz − (y 2 − zx) y 2 − zx − (z 2 − xy)

dx − dy dy − dz
⇒ =
x2 − y 2 − yz + zx y 2 − z 2 − zx + xy
dx − dy dy − dz
⇒ =
(x + y)(x − y) + z(x − y) (y + z)(y − z) + x(y − z)
 dx − dy dy − dz
⇒ =
(x + y)(x − y) + z(x − y) (y + z)(y − z) + x(y − z)
dx − dy dy − dz
⇒ =
(x − y)(x + y + z) (y − z)(x + y + z)
dx − dy dy − dz
⇒ =
(x − y) (y − z)
On integrating, log (x − y) = log (y − z) + log a
⇒ log (x − y) − log (y − z) = log a
x−y x−y
⇒ log = log a ⇒ =a
y−z y−z
x−y
Set u = =a
y−z
Taking the difference between last two and last and first
fractions and equating them
dy − dz dz − dx
=
y 2 − zx − (z 2 − xy) z 2 − xy − (x2 − yz)
dy − dz dz − dx
⇒ 2 =
y − z 2 − zx + xy z 2 − x2 − xy + yz
dy − dz dz − dx
⇒ =
(y + z)(y − z) + x(y − z) (z + x)(z − x) + y(z − x)
dy − dz dz − dx
⇒ =
(y − z)(x + y + z) (z − x)(x + y + z)
dy − dz dz − dx
⇒ =
(y − z) (z − x)
On integrating, log (y − z) = log(z − x) + log b
⇒ log (y − z) − log(z − x) = log b
y−z y−z
⇒ log = log b ⇒ =b
z−x z−x
y−z
Set v = =b
z−x
Step 4 General
 equationis φ(u, v) = 0 or u 
= f (v) 
x−y y−z x−y y−z
i.e. φ , = 0 or =f
y−z z−x y−z z−x
Example 16. Solve y 2 p − xy q = x(z − 2y)

Solution:- Given PDE can be rearranged as y 2 p − xyq = xz − 2xy

Step 1 P = y 2 , Q = −xy, R = xz − 2xy

dx dy dz dx dy dz
Step 2 Auxiliary equation- = = ⇒ 2 = =
P Q R y −xy xz − 2xy
Step 3 Solve the auxiliary equation by grouping
dx dy
Taking first two fractions 2 =
y −xy
dx dy
⇒ = ⇒ xdx = −ydy
y −x
x2 y2 a
On integrating ⇒ =− + ⇒ x2 + y 2 = a
2 2 2
Set u = x2 + y 2 = a
dy dz
Taking second and third fractions =
−xy x(z − 2y)
dy dz
⇒ = ⇒ (z − 2y)dy = −ydz
−y (z − 2y)
⇒ zdy − 2ydy = −ydz ⇒ zdy + ydz = 2ydy

Result: Product rule for differentiation- d(uv) = u dv + v du

⇒ d(zy) = 2ydy
y2
On integrating zy = 2 + a ⇒ zy − y 2 = a
2
Step 4 General equation is φ(u, v) = 0 or u = f (v)
i.e. φ(x2 + y 2 , zy − y 2 ) = 0 or x2 + y 2 = f (zy − y 2 )

Exercise

Solve the following PDE

(i) xp + yq = 3z
 √ √
(ii) p x + q y = z

(iii) y 2 zp + x2 zq = y 2 x

(iv) pyz + qzx = xy

(v) xp − yq = y 2 − x2

(vi) p − q = log (x + y)

(vii) p + 3q = 5z + tan(y − 3x)

(viii) (y 2 + z 2 − x2 )p − 2xyq + 2xz = 0

Case 2. Method of Lagrange’s Multipliers

Let P1 , Q1 and R1 be the function of x, y and z. Then by the property

dx dy dz
of ratios of each fraction in auxiliary equation = = will be
P Q R
equal to
P1 dx + Q1 dy + R1 dz
P P1 + QQ1 + RR1
If P P1 + QQ1 + RR1 = 0, then P1 dx + Q1 dy + R1 dz = 0. This equation
P1 dx + Q1 dy + R1 dz = 0 can be integrated to give one of the solution
u = a. This method may be repeated to get another solution v = b.

Remark 2. P1 , Q1 and R1 are called Lagrange’s multipliers. As a spe-

cial case, Lagrange’s multipliers can be constants also.

Working Rule for Lagrange’s Multipliers

 (i) Choose P1 , Q1 and R1 as Lagrange’s multipliers such that
P P1 + QQ1 + RR1 = 0

(ii) Solve the differential equation P1 dx + Q1 dy + R1 dz = 0 (by

integration) for the solution u = a

(iii) The aboe procedure may be repeated for the second solution
of auxiliary equation v = b

Example 17. Solve x(y − z)p + y(z − x)q = z(x − y)

Solution:

Step 1 P = x(y − z), Q = y(z − x), R = z(x − y)

dx dy dz
Step 2 Auxiliary equation- = =
P Q R

dx dy dz
⇒ = =
x(y − z) y(z − x) z(x − y)
Step 3 Solve the auxiliary equation by using Lagrange’s multipliers
1 1 1
Choose P1 = , Q1 = and R1 = such that
x y z
1 1 1
P P1 + QQ1 + RR1 = x(y − z) + y(z − x) + z(x − y)
x y z
= y−z+z−x+x−y
= 0

So the solution u = a is the solution of the differential equation

1 1 1
P1 dx + Q1 dy + R1 dz = 0 ⇒ dx + dy + dz = 0
x y z
On integrating, log x + log y + log z = log a
⇒ log (xyz) = log a ⇒ xyz = a
 Set u = xyz = a
Choose P2 = 1, Q2 = 1 and R2 = 1 such that

P P2 + QQ2 + RR2 = x(y − z) + y(z − x) + z(x − y)

= xy − xz + yz − yx + zx − zy
= 0

So the solution v = b is the solution of the differential equation

P2 dx + Q2 dy + R2 dz = 0 ⇒ dx + dy + dz = 0

On integrating, x + y + z = b
⇒ set v = x + y + z = b

Step 4 General equation is φ(u, v) = 0 or u = f (v)

i.e. φ(xyz, x + y + z) = 0 or xyz = f (x + y + z)
Example 18. Solve x(y 2 − z 2 )p + y(z 2 − x2 )q − z(x2 − y 2 ) = 0
Solution:- Given PDE can be rearranged in standard form P p + Qq = R
as x(y 2 − z 2 )p + y(z 2 − x2 )q = z(x2 − y 2 )
Step 1 P = x(y 2 − z 2 ), Q = y(z 2 − x2 ), R = z(x2 − y 2 )
dx dy dz
Step 2 Auxiliary equation- = =
P Q R

dx dy dz
⇒ = =
x(y 2 − z 2 ) y(z 2 − x2 ) z(x2 − y 2 )
Step 3 Solve the auxiliary equation by using Lagrange’s multipliers
1 1 1
Choose P1 = , Q1 = and R1 = such that
x y z
1 1 1
P P1 + QQ1 + RR1 = x(y 2 − z 2 ) + y(z 2 − x2 ) + z(x2 − y 2 )
x y z
= y 2 − z 2 + z 2 − x2 + x2 − y 2
= 0
 So the solution u = a is the solution of the differential equation
1 1 1
P1 dx + Q1 dy + R1 dz = 0 ⇒ dx + dy + dz = 0
x y z
On integrating, log x + log y + log z = log a
⇒ log (xyz) = log a ⇒ xyz = a
Set u = xyz = a
Choose P2 = x, Q2 = y and R2 = z such that

P P2 + QQ2 + RR2 = x2 (y 2 − z 2 ) + y 2 (z 2 − x2 ) + z 2 (x2 − y 2 )

= x2 y 2 − x2 z 2 + y 2 z 2 − y 2 x2 + z 2 x2 − z 2 y 2
= 0

So the solution v = b is the solution of the differential equation

P2 dx + Q2 dy + R2 dz = 0 ⇒ x dx + y dy + z dz = 0
x2 y 2 z 2 a
On integrating, + + =
2 2 2 2
⇒ x + y + z = b ⇒ set v = x2 + y 2 + z 2 = b
2 2 2

Step 4 General equation is φ(u, v) = 0 or u = f (v)

i.e. φ(xyz, x2 + y 2 + z 2 ) = 0 or xyz = f (x2 + y 2 + z 2 )

Example 19. Solve x2 (y − z)p + y 2 (z − x)q = z 2 (x − y)

Solution:

Step 1 P = x2 (y − z), Q = y 2 (z − x), R = z 2 (x − y)

dx dy dz
Step 2 Auxiliary equation- = =
P Q R

dx dy dz
⇒ = =
x2 (y − z) y 2 (z − x) z 2 (x − y)
Step 3 Solve the auxiliary equation by using Lagrange’s multipliers
1 1 1
Choose P1 = , Q1 = and R1 = such that
x y z
1 1 1
P P1 + QQ1 + RR1 = x2 (y − z) + y 2 (z − x) + z 2 (x − y)
x y z
= x(y − z) + y(z − x) + z(x − y)
= xy − xz + yz − yx + zx − zy
= 0

So the solution u = a is the solution of the differential equation

1 1 1
P1 dx + Q1 dy + R1 dz = 0 ⇒ dx + dy + dz = 0
x y z
On integrating, log x + log y + log z = log a
⇒ log (xyz) = log a ⇒ xyz = a
Set u = xyz = a
1 1 1
Choose P2 = 2 , Q2 = 2 and R2 = 2 such that
x y z
1 1 1
P P2 + QQ2 + RR2 = x2 (y − z)2
+ y 2 (z − x) 2 + z 2 (x − y) 2
x y z
= y−z+z−x+x−y
= 0

So the solution v = b is the solution of the differential equation

1 1 1
P2 dx + Q2 dy + R2 dz = 0 ⇒ dx + dy + dz = 0
x2 y2 z2

Result  
−1
Z
d 1 1 1
= 2 then 2
dx = − + c
dx x x x x
 1 1 1
On integrating, −
− − = −b
x y z
1 1 1 1 1 1
⇒ + + = b ⇒ set v = + + = b
x y z x y z

Step 4 General
 equation isφ(u, v) = 0 or u =
 f (v) 
1 1 1 1 1 1
i.e. φ xyz, + + = 0 or xyz = f + +
x y z x y z
∂z ∂z
Example 20. Solve (mz − ny) + (nx − lz) = ly − mx
∂x ∂y
Solution:-

Step 1 P = (mz − ny), Q = (nx − lz), R = (ly − mx)

dx dy dz
Step 2 Auxiliary equation- = =
P Q R
dx dy dz
⇒ = =
mz − ny nx − lz ly − mx
Step 3 Solve the auxiliary equation by using Lagrange’s multipliers
Choose P1 = x, Q1 = y and R1 = z such that

P P1 + QQ1 + RR1 = x(mz − ny) + y(nx − lz) + z(ly − mx)

= mxz − nxy + nyx − lyz + lzy − mxz
= 0

So the solution u = a is the solution of the differential equation

P1 dx + Q1 dy + R1 dz = 0 ⇒ x dx + y dy + z dz = 0
x2 y 2 z 2 a
On integrating, + + =
2 2 2 2
⇒ x + y + z = a ⇒ set u = x2 + y 2 + z 2 = a
2 2 2
 Choose P2 = l, Q2 = m and R2 = n such that

P P2 + QQ2 + RR2 = (mz − ny)l + (nx − lz)m + (ly − mx)n

= mlz − nly + nmx − lmz + lny − mnx
= 0

So the solution v = b is the solution of the differential equation

P2 dx + Q2 dy + R2 dz = 0 ⇒ ldx + mdy + ndz = 0

On integrating, lx + my + nz = b ⇒ set v = lx + my + nz = b

Step 4 General equation is φ(u, v) = 0 or u = f (v)

i.e. φ x2 + y 2 + z 2 , lx + my + nz = 0 or x2 +y 2 +z 2 = f (lx + my + nz)

Exercise

Solve the following PDE

(i) x(z 2 − x2 )p + y(x2 − z 2 )q = z(y 2 − x2 )

(ii) z(x + y)p + z(x − y)q = x2 + y 2 (Hint: (x, −y, z) and (y, x, −z) are
the multipliers)

(iii) (z − y)p + (x − z)q = y − x

Case 3. Combination of Case 1 and Case 2

We can solve auxiliary equation of a linear PDE by using case 1 and case
2 for u = a and v = b.
Example 21. Solve (z 2 − 2yz − y 2 )p + (xy + zx)q = xy − zx

Solution:-

Step 1 P = z 2 − 2yz − y 2 , Q = xy + yz, R = xy − zx

dx dy dz
Step 2 Auxiliary equation- = =
P Q R

dx dy dz
⇒ = =
z 2 − 2yz − y 2 xy + zx xy − zx
Step 3 Solve the auxiliary equation
Calculation of u = a by using Lagrange’s multipliers
Choose P1 = x, Q1 = y and R1 = z such that

P P1 + QQ1 + RR1 = x(z 2 − 2yz − y 2 ) + y(xy + zx) + z(xy − zx)

= xz 2 − 2xyz − xy 2 + xy 2 + zxy + zxy − z 2 x
= 0

So the solution u = a is the solution of the differential equation

P1 dx + Q1 dy + R1 dz = 0 ⇒ x dx + y dy + z dz = 0
x2 y 2 z 2 a
On integrating, + + =
2 2 2 2
⇒ x2 + y 2 + z 2 = a ⇒ set u = x2 + y 2 + z 2 = a
Calculation of v = b by grouping[Taking last two fractions]
dy dz
=
xy + zx xy − zx
dy dz dy dz
⇒ = ⇒ =
x(y + z) x(y − z) (y + z) (y − z)

⇒ (y − z)dy = (y + z)dz ⇒ ydy − zdy = ydz + zdz

⇒ ydy − zdz = ydz + zdy
 ⇒ ydy − zdz = d(yz)
y2 z2 y2 z2
On integrating, − = yz + b ⇒ − − yz = b
2 2 2 2
2 2
y z
Set v = − − yz = b
2 2
Step 4 General equation is φ(u, v) = 
0 or u = f (v)
2 2
 2
z2
 
2 2 2 y z 2 2 2 y
i.e. φ x + y + z , − − yz = 0 or x +y +z = f − − yz
2 2 2 2

Exercise

Solve the following PDE

   
y−z z−x x−y
(i) p+ q=
yz zx xy
(ii) z(xp − yq) = y 2 − x2

(iii) x2 p + y 2 q = (x + y)z

3.3 Method 3

Non Linear Equations of the First Order (CHARPIT’S METHOD)

Definition 3.5. A PDE which involves first order partial derivatives

p and q with degree higher than one and the product of p and q is called
a non linear partial differential equation of the first order.
Example: p2 + q 2 = z, yp + xq + pq = 0
 Note: Non linear partial differential equation of the first order with
two independent variable can be represented as f (x, y, z, p, q) = 0

CHARPIT’S METHOD
This is a general method for solving non linear first order partial differ-
ential equation f (x, y, z, p, q) = 0 with two independent variables. The
complete solution of such an equation contains only two arbitrary con-
stant. (equal to the number of independent variables x and y)
Working Rule for the Solution of f (x, y, z, p, q) = 0 in Charpit’s Method

Step 1 Rearrange the PDE in the form f (x, y, z, p, q) = 0

Step 2 Find out the following Partial derivatives

∂f ∂f ∂f ∂f ∂f
fx = , fy = , fz = , fp = and fq =
∂x ∂y ∂z ∂p ∂q
Step 3 Write down the Charpits’s auxiliary equation
dx dy dz dp dq
= = = =
fp fq (pfp + qfq ) −(fx + pfz ) −(fy + qfz )
Step 4 Solve the above auxiliary equation by grouping or rearrangement
of any two fractions and derive simplest relation involving at least
one of p and q.

Step 5 The simplest relation along with the given PDE to determine
p and q

Step 6 Substitute the value of p and q in dz = pdx + qdy [since z =

f (x, y) ⇒ complete derivtive is dz = pdx + qdy]

step 7 On integration of the above equation gives the complete integral

of the given equation
Example 22. Solve (p2 + q 2 )y = qz

Solution:-

Step 1 Rearrange the given PDE as f (x, y, z, p, q) = 0

⇒ f = p2 y + q 2 y − qz = 0

Step 2 Find the first order partial derivatives

fx = 0, fy = p2 + q 2 , fz = −q, fp = 2py and fq = 2qy − z

Step 3 Write down the Charpits’s auxiliary equation

dx dy dz dp dq
= = = = (22)
fp fq (pfp + qfq ) −(fx + pfz ) −(fy + qfz )

pfp + qfq = p × 2py + q × (2qy − z) = 2p2 y + 2q 2 y − qz

= 2(p2 + q 2 )y − qz
= 2qz − qz [F rom given P DE (p2 + q 2 )y = qz]
= qz

fx + pfz = 0 + p(−q) = −pq

fy + qfz = p2 + q 2 + q(−q) = p2 + q 2 − q 2 = p2
Substitute these values in equation (17)
dx dy dz dp dq
= = = =
2py 2qy − z (qz) −(−pq) −(p2 )
dx dy dz dp dq
= = = = 2
2py 2qy − z qz −pq p
Step 4 Taking last two fractions of above auxiliary
dp dq
= 2
−pq p
⇒ pdp = −qdq ⇒ ⇒ pdp + qdq = 0
p2 q 2 c2
On integrating, + = ⇒ p2 + q 2 = c 2
2 2 2
Step 5 Substitute the value of the equation p2 + q 2 = c2 , from above
step, in the given PDE (p2 + q 2 )y = qz
c2 y
⇒ c2 y = qz ⇒ q =
z
Substitute the value of q in the equation p2 + q 2 = c2
2 c4 y 2 2 p2 z 2 + c4 y 2
⇒, p + 2 = c ⇒ 2
= c 2 ⇒ p2 z 2 + c 4 y 2 = c 2 z 2
z sz
 2 2 p
4 2

c z − c y c z 2 − c2 y 2
p2 z 2 = c2 z 2 − c4 y 2 ⇒ p = ⇒ p =
z2 z
Step 6 Substitute
p the value of p and q in the equation dz = pdx + qdy
c z 2 − c2 y 2 c2 y
⇒ dz = dx + dy
p z z
c z 2 − c2 y 2 dx + c2 ydy
⇒ dz =
z2
p
⇒ zdz − c2 ydy = c z 2 − c2 y 2 dx
zdz − c2 ydy
⇒ p = c dx
z 2 − c2 y 2
[put t = z 2 − c2 y 2 ⇒ dt = 2zdz − c2 2y dy ⇒ dt = 2(zdz − c2 ydy)
dt
= zdz − c2 ydy]
2
d √ √
 
dt 1 R 1
⇒ √ = c dx ( x) = √ ⇒ √ = x+c
2 t dx 2 x 2 x
√ p
Step 7 On integrating, t = cx + a ⇒ z 2 − c2 y 2 = cx + a
⇒ z 2 − c2 y 2 = (cx + a)2
⇒ z 2 = c2 y 2 + (cx + a)2 which is the required complete integral.

Example 23. Solve 2zx − px2 − 2qxy + pq = 0

Solution:-

Step 1 f (x, y, z, p, q) = 2zx − px2 − 2qxy + pq = 0

step 2 fx = 2z − 2xp − 2qy, fy = −2qx, fz = 2x, fp = −x2 + q,
fq = −2xy + p

Step 3 Write down the Charpits’s auxiliary equation

dx dy dz dp dq
= = = =
fp fq (pfp + qfq ) −(fx + pfz ) −(fy + qfz )

pfp + qfq = p(−x2 + q) + q(−2xy + p)

= −px2 + pq − 2xyq + pq
= 2pq − px2 − 2xyq

fx + pfz = 2z − 2xp − 2qy + p.2x = 2z − 2xp − 2qy + 2xp = 2z − 2qy

fy + qfz = −2qx + q.2x = −2qx + 2qx = 0
Substitute these values in the above equation
dx dy dz dp dq
= = = =
−x2 + q −2xy + p 2pq − px2 − 2xyq −(2z − 2qy) 0
dq
Step 4 From the last fraction, ⇒ dq = 0 ⇒ q = a
0
Step 5 Put q = a in the given PDE ⇒ 2zx − px2 − 2axy + pa = 0
⇒ pa − px2 = 2axy − 2zx
⇒ p(a − x2 ) = 2x(ay − z)
2x(ay − z)
⇒ p=
a − x2
Step 6 Substitute the value of p and q in dz = pdx + qdy
2x(ay − z)
⇒ dz = dx + ady
a − x2
2x(ay − z)
⇒ dz − ady = dx
a − x2
dz − ady 2x
⇒ = dx
ay − z a − x2
dz − ady 2x
⇒ = 2 dx
z − ay x −a
{Take negative sign common from denominator}
 Z 
dz − ady
Z
2x
Step 7 on integrating = dx
ay − z a − x2

Result Z 0
f (x)
dx = logf (x) + x
f (x)

⇒ log (z − ay) = log (x2 − a) + log b

⇒ log (z − ay) − log (x2 − a) = log b
z − ay z − ay
⇒ log 2 = log b ⇒ 2 =b
x −a x −a
⇒ z − ay = b(x2 − a)
⇒ z = ay + b(x2 − a)

3.4 Method 4

Method of Separation of Variables

The differential equation together with some specified conditions ( bound-
ary conditions) constitute boundary value problem. Most of the
boundary value problems involving linear PDE can be solved by the
method of separation of variables. This method involves a solution
which break up into a product of functions each of which contains only
one of the variables

Example 24. Solve by the method of separation of variables

∂ 2z ∂z ∂z
− 2 + =0
∂x2 ∂x ∂y
Solution:- Assume the solution of given Partial differential equation is
z = XY
where X is a function of x alone and Y that of y alone. [ From the given
PDE, the dependent variable z is partially differentiate w.r.to x and y.
So in this method, solution is a product of functions of x and y.]

∂z 0 ∂z 0 ∂ 2z 00
If z = XY , then = X Y, = XY , 2
=X Y
∂x ∂y ∂x

Substitute these values in given PDE

00 0 0
⇒ X Y − 2X Y + XY = 0
00 0 0
⇒ X Y − 2X Y = XY
00 0 0
⇒ (X − 2X )Y = −XY
00 0 0
X − 2X Y
⇒ =
X Y
Since x and y are independent variables, it can only be true if each side
is equal to the same constant, a (say)
00 0 0
X − 2X Y
⇒ = a and =a
X Y
00 0 0
⇒ X − 2X = aX and Y = aY
00 0 0
⇒ X − 2X − aX = 0 and Y − aY = 0
00 0
Case 1 Consider the ordinary differential equation X − 2X − aX = 0
and the auxiliary equation is m2 − 2m − a = 0 where
√ !
−b ± b2 − 4ac √
m= =1± 1+a
2a
√ √
(1+ 1+a)x (1− 1+a)x
⇒ X = c1 e + c2 e
Note :-If the roots of an auxiliary equations are real and distinct
(say m1 and m2 ), the solution is y = c1 em1 x + c2 em2 x

0
Case 2 Consider the ordinary differential equation Y + aY = 0 and
the auxiliary equation is m + a = 0 and m = −a.
⇒ y = c3 e−ay
The required solution is z = XY
 √ √ 
⇒ z = c1 e (1+ 1+a)x
+ c2 e (1− 1+a)x
c3 e−ay
 √ √ 
⇒ z = c1 c3 e (1+ 1+a)x
+ c2 c3 e (1− 1+a)x
e−ay
 √ √ 
⇒ z = k1 e (1+ 1+a)x
+ k2 e (1− 1+a)x
e−ay
where c1 c2 = k1 and c2 c3 = k2
∂u
Example 25. Using the method of separation of variables, solve =
∂x
∂u
2 + u where u(x, 0) = 6e−3x .
∂t
∂u ∂u
Solution:- =2 +u
∂x ∂t
Assume the solution of given Partial differential equation is u = XT
where X is a function of x alone and T that of t alone.
∂u 0 ∂u 0
⇒ = X T, = XT
∂x ∂t
Substitute these values in given PDE
0 0
⇒ X T = 2XT + XT
0 0
⇒ X T − XT = 2XT
0 0
⇒ (X − X)T = 2XT
0 0
X −X T
⇒ =
2X T
Since x and y are independent variables, it can only be true if each side
is equal to the same constant, a (say)
0 0
X −X T
⇒ = a and =a
0
2X T 0
⇒ X − X = 2aX and T − aT = 0
0 0
⇒ X − X − 2aX = 0 and T − aT = 0
0 0
⇒ X − (2a + 1)X = 0 and T − aT = 0
0
Case 1 Consider the ordinary differential equation X − (2a + 1)X = 0
and the auxiliary equation is m − (2a + 1) = 0 ⇒ m = 2a + 1
⇒ X = c1 e(2a+1)x
0
Case 2 Consider the ordinary differential equation T − aT = 0 and the
auxiliary equation is m − a = 0 ⇒ m = a
⇒ T = c2 eat
The required solution is u = XT ⇒ u = c1 e(2a+1)x c2 eat
⇒ u = c1 c2 e(2a+1)x eat ⇒ u = ke(2a+1)x eat where c1 c2 = k
Given that u(x, 0) = 6e−3x [i.e. at t = 0, u = 6e−3x ]
⇒ 6e−3x = ke(2a+1)x ⇒ k = 6 and 2a + 1 = −3 ⇒ a = −2
⇒ T he required solution is u = 6e−3x e−2t ⇒ u = 6e−(3x+2t)

References

B S Grewal; Higher Engineering Mathematics; Khanna Publishers, 44th

edition; 2018
Peter V. O’Neil, Advanced Engineering Mathematics, Cengage, 7th Edi-
tion, 2012