MA 201: Partial Differential Equations Lecture - 4

MA 201: Partial Differential Equations
Lecture - 4
IIT Guwahati MA201 (2022)

Solution technique for quasi-linear equations: Method of
Characteristics
• Recall a first-order quasi-linear PDE:
∂u ∂u
a(x, y , u) + b(x, y , u) = c(x, y , u). (1)
∂x ∂y

Characteristics
∂u ∂u
a(x, y , u) + b(x, y , u) = c(x, y , u). (1)
∂x ∂y
• Let us assume that an integral surface u = u(x, y ) of (1) can be

found. Write this integral surface in implicit form as
F (x, y , u) = u(x, y ) − u = 0.

Characteristics
∂u ∂u
a(x, y , u) + b(x, y , u) = c(x, y , u). (1)
∂x ∂y

F (x, y , u) = u(x, y ) − u = 0.
• Equation (1) may be written as
aux + buy − c = ha, b, ci · hux , uy , −1i = 0. (2)

Characteristics
∂u ∂u
a(x, y , u) + b(x, y , u) = c(x, y , u). (1)
∂x ∂y

F (x, y , u) = u(x, y ) − u = 0.
• Equation (1) may be written as
aux + buy − c = ha, b, ci · hux , uy , −1i = 0. (2)
• This shows that the vector ha, b, ci and the gradient vector ∇F are
orthogonal at the point (x, y , u).
• In other words, at each point (x, y , u) of the integral surface, the
tangent vector to the integral surface is given by the vector ha, b, ci.

• Recall that a tangent vector to the integral surface at
(x, y , u) = (x(t), y (t), u(t)) is given by
r′ (t) = hx ′ (t), y ′ (t), u ′ (t)i, (3)
for some t belonging to an interval I .

r′ (t) = hx ′ (t), y ′ (t), u ′ (t)i, (3)
Then, we must have

dx dy du
= a(t), = b(t), = c(t). (4)
dt dt dt
The solutions of (4) are called the

r′ (t) = hx ′ (t), y ′ (t), u ′ (t)i, (3)
Then, we must have

dx dy du
= a(t), = b(t), = c(t). (4)
dt dt dt
The solutions of (4) are called the characteristic curves for the
quasi-linear PDE.

r′ (t) = hx ′ (t), y ′ (t), u ′ (t)i, (3)
Then, we must have

dx dy du
= a(t), = b(t), = c(t). (4)
dt dt dt
The solutions of (4) are called the characteristic curves for the
quasi-linear PDE.

Example
Find an integral surface of the quasi-linear PDE
∂u ∂u
x(y 2 + u)p − y (x 2 + u)q = (x 2 − y 2 )u, p = & q=
∂x ∂y
which contains the straight line x + y = 0, u = 1.

Example
∂u ∂u
x(y 2 + u)p − y (x 2 + u)q = (x 2 − y 2 )u, p = & q=
∂x ∂y
Remarks:
• Characteristic equations are
dx
= x(y 2 + u) = f1 (x, y , u)
dt
dy
= −y (x 2 + u) = f2 (x, y , u)
dt
du
= (x 2 − y 2 )u = f3 (x, y , u).
dt

Example
∂u ∂u
x(y 2 + u)p − y (x 2 + u)q = (x 2 − y 2 )u, p = & q=
∂x ∂y
Remarks:
• Characteristic equations are
dx
= x(y 2 + u) = f1 (x, y , u)
dt
dy
= −y (x 2 + u) = f2 (x, y , u)
dt
du
= (x 2 − y 2 )u = f3 (x, y , u).
dt
• These are nonlinear equations in x and y , and we do not go ahead

to solve them at this moment. Instead, we next focus on a familiar
method due to Lagrange.
Solution Technique: Method of Lagrange
Background:
Theorem
If φ = φ(x, y , u) and ψ = ψ(x, y , u) are two given functions of x, y and
u, and if G (φ, ψ) = 0, where G is an arbitrary function of φ and ψ, then
u = u(x, y ) satisfies a first-order PDE
∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

+ = , (5)
∂x ∂(y , u) ∂y ∂(u, x) ∂(x, y )
where
∂(φ, ψ) φx φy

= .

∂(x, y ) ψx ψy

Solution Technique: Method of Lagrange
Background:
Theorem
If φ = φ(x, y , u) and ψ = ψ(x, y , u) are two given functions of x, y and
u, and if G (φ, ψ) = 0, where G is an arbitrary function of φ and ψ, then
u = u(x, y ) satisfies a first-order PDE
∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

+ = , (5)
∂x ∂(y , u) ∂y ∂(u, x) ∂(x, y )
where
∂(φ, ψ) φx φy

= .

∂(x, y ) ψx ψy
How to establish it?

What is given?

What is given? A surface G (φ, ψ) = 0.

What is given? A surface G (φ, ψ) = 0. How to get a PDE out of it?

• Differentiate G (φ, ψ) = 0 with respect to x and y , respectively:
Gφ φx + ux φu ) + Gψ ψx + ux ψu ) = 0, (6)
Gφ φy + uy φu ) + Gψ ψy + uy ψu ) = 0. (7)

• Nontrivial solutions for ∂G ∂G

∂φ (= Gφ ) and ∂ψ (= Gψ ) can be found when

φx + ux φu ψx + ux ψu

= 0.
ψy + uy ψu ψy + uy ψu


• Nontrivial solutions for ∂G ∂G

∂φ (= Gφ ) and ∂ψ (= Gψ ) can be found when

φx + ux φu ψx + ux ψu

= 0.
ψy + uy ψu ψy + uy ψu

• Expanding this determinant obviously gives a first-order quasi-linear

PDE.

• Since surface G (φ, ψ) = 0 leads to equation
∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

+ = , (8)
∂x ∂(y , u) ∂y ∂(u, x) ∂(x, y )
therefore, G (φ, ψ) = 0 is a solution for the equation.

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

+ = , (8)
∂x ∂(y , u) ∂y ∂(u, x) ∂(x, y )
therefore, G (φ, ψ) = 0 is a solution for the equation. Since G is

arbitrary, this solution is called the general solution.

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

+ = , (8)
∂x ∂(y , u) ∂y ∂(u, x) ∂(x, y )

• This may give an idea to find the general solution for the
quasi-linear equation
∂u ∂u
a(x, y , u) + b(x, y , u) = c(x, y , u). (9)
∂x ∂y

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

+ = , (8)
∂x ∂(y , u) ∂y ∂(u, x) ∂(x, y )

∂u ∂u
a(x, y , u) + b(x, y , u) = c(x, y , u). (9)
∂x ∂y
• How?

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

+ = , (8)
∂x ∂(y , u) ∂y ∂(u, x) ∂(x, y )

∂u ∂u
a(x, y , u) + b(x, y , u) = c(x, y , u). (9)
∂x ∂y
• How? Both equations (8) and (9) will be identical if we select

φ = φ(x, y , u) and ψ = ψ(x, y , u) in such a way that

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

+ = , (8)
∂x ∂(y , u) ∂y ∂(u, x) ∂(x, y )

∂u ∂u
a(x, y , u) + b(x, y , u) = c(x, y , u). (9)
∂x ∂y
• How? Both equations (8) and (9) will be identical if we select

φ = φ(x, y , u) and ψ = ψ(x, y , u) in such a way that
∂(φ, ψ) ∂(φ, ψ) ∂(φ, ψ)

a=λ , b=λ , c=λ ,
∂(y , u) ∂(u, x) ∂(x, y )
for some λ.

Theorem (The method of Lagrange)
The general solution of the quasi-linear PDE
∂u ∂u
a(x, y , u) + b(x, y , u) = c(x, y , u) (10)
∂x ∂y
is given by
G (φ, ψ) = 0, (11)
where G is an arbitrary function of φ and ψ with φ(x, y , u) = c1 and
ψ(x, y , u) = c2 being the solutions of the equations
dx dy du
= = . (12)
a b c

Theorem (The method of Lagrange)
The general solution of the quasi-linear PDE
∂u ∂u
a(x, y , u) + b(x, y , u) = c(x, y , u) (10)
∂x ∂y
is given by
G (φ, ψ) = 0, (11)
where G is an arbitrary function of φ and ψ with φ(x, y , u) = c1 and
ψ(x, y , u) = c2 being the solutions of the equations
dx dy du
= = . (12)
a b c
Proof. Observe that
dφ = φx dx + φy dy + φu du = 0, (13)
dψ = ψx dx + ψy dy + ψu du = 0. (14)

On the integral surface, we have (through characteristic equations)
dx dy du
= = . (15)
a b c

dx dy du
= = . (15)
a b c
Then, we must have (comparing with (13) and (14):
aφx + bφy + cφu = 0, (16)

aψx + bψy + cψu = 0. (17)

dx dy du
= = . (15)
a b c
aφx + bφy + cφu = 0, (16)

aψx + bψy + cψu = 0. (17)
Solving these equations for a, b and c, we obtain

a b c
∂(φ,ψ)
= ∂(φ,ψ)
= ∂(φ,ψ)
. (18)
∂(y,u) ∂(u,x) ∂(x,y)
This completes the rest of the proof.

dx dy du
= = . (15)
a b c
aφx + bφy + cφu = 0, (16)

aψx + bψy + cψu = 0. (17)
Solving these equations for a, b and c, we obtain

a b c
∂(φ,ψ)
= ∂(φ,ψ)
= ∂(φ,ψ)
. (18)
∂(y,u) ∂(u,x) ∂(x,y)
This completes the rest of the proof.

Remark
Any expression involving such φ and ψ is an integral surface for (1).

Example
Find the general integral/general solution of xux + yuy = u.

Example
Solution. The associated system of equations are
dx dy du
= = .
x y u

Example
dx dy du
= = .
x y u
From the first two relations, we have
dx dy x
= =⇒ ln x = ln y + ln c1 =⇒ = c1 .
x y y

Example
dx dy du
= = .
x y u
dx dy x
= =⇒ ln x = ln y + ln c1 =⇒ = c1 .
x y y
Similarly,
du dy u
= =⇒ = c2 .
u y y

Example
dx dy du
= = .
x y u
dx dy x
= =⇒ ln x = ln y + ln c1 =⇒ = c1 .
x y y
Similarly,
du dy u
= =⇒ = c2 .
u y y
x
Take φ = y and ψ = uy . The general integral is given by

Example
dx dy du
= = .
x y u
dx dy x
= =⇒ ln x = ln y + ln c1 =⇒ = c1 .
x y y
Similarly,
du dy u
= =⇒ = c2 .
u y y
x
Take φ = y and ψ = uy . The general integral is given by

x u
G , = 0,
y y
where G is an arbitrary function of φ and ψ.
Example
Find the general integral for u(x + y )ux + u(x − y )uy = x 2 + y 2 .
Solution. The characteristic equations are

Example
dx dy du
= = 2 . (19)
u(x + y ) u(x − y ) x + y2

Example
dx dy du
= = 2 . (19)
u(x + y ) u(x − y ) x + y2
Each of these ratios is equivalent to
dx dy du dx dy du
y +x −u =0=x −y −u .
dt dt dt dt dt dt

Example
dx dy du
= = 2 . (19)
u(x + y ) u(x − y ) x + y2
dx dy du dx dy du
y +x −u =0=x −y −u .
dt dt dt dt dt dt
Consequently, we have
u2

1 2 2 2
d xy − = 0 and d (x − y − u ) = 0.
2 2

Example
dx dy du
= = 2 . (19)
u(x + y ) u(x − y ) x + y2
dx dy du dx dy du
y +x −u =0=x −y −u .
dt dt dt dt dt dt
u2

1 2 2 2
d xy − = 0 and d (x − y − u ) = 0.
2 2
Integrating the above, we obtain
2xy − u 2 = c1 = φ and x 2 − y 2 − u 2 = c2 = ψ.

Example
dx dy du
= = 2 . (19)
u(x + y ) u(x − y ) x + y2
dx dy du dx dy du
y +x −u =0=x −y −u .
dt dt dt dt dt dt
u2

1 2 2 2
d xy − = 0 and d (x − y − u ) = 0.
2 2
Integrating the above, we obtain
2xy − u 2 = c1 = φ and x 2 − y 2 − u 2 = c2 = ψ.
Thus, the general solution is G (2xy − u 2 , x 2 − y 2 − u 2 ) = 0, where G is

an arbitrary function.
Alternative method of solution. From (19), we can choose
ydx + xdy du
= 2 ,
uy (x + y ) + ux(x − y ) x + y2
which gives us

ydx + xdy du
= 2 ,
uy (x + y ) + ux(x − y ) x + y2
which gives us
ydx + xdy = udu
leading to
u2
xy − = c1 .
2

ydx + xdy du
= 2 ,
uy (x + y ) + ux(x − y ) x + y2
which gives us
ydx + xdy = udu
leading to
u2
xy − = c1 .
2
In a similar manner, from (19), we can choose
xdx − ydy du
= 2 ,
ux(x + y ) − uy (x − y ) x + y2
which gives us

ydx + xdy du
= 2 ,
uy (x + y ) + ux(x − y ) x + y2
which gives us
ydx + xdy = udu
leading to
u2
xy − = c1 .
2
xdx − ydy du
= 2 ,
ux(x + y ) − uy (x − y ) x + y2
which gives us
xdx − ydy = udu
leading to
x 2 − y 2 − u 2 = c2 .

ydx + xdy du
= 2 ,
uy (x + y ) + ux(x − y ) x + y2
which gives us
ydx + xdy = udu
leading to
u2
xy − = c1 .
2
xdx − ydy du
= 2 ,
ux(x + y ) − uy (x − y ) x + y2
which gives us
xdx − ydy = udu
leading to
x 2 − y 2 − u 2 = c2 .
Hence, we obtain
2xy − u 2 = c1 = φ and x 2 − y 2 − u 2 = c2 = ψ.

ydx + xdy du
= 2 ,
uy (x + y ) + ux(x − y ) x + y2
which gives us
ydx + xdy = udu
leading to
u2
xy − = c1 .
2
xdx − ydy du
= 2 ,
ux(x + y ) − uy (x − y ) x + y2
which gives us
xdx − ydy = udu
leading to
x 2 − y 2 − u 2 = c2 .
Hence, we obtain
2xy − u 2 = c1 = φ and x 2 − y 2 − u 2 = c2 = ψ.
Thus, the general solution is G (2xy − u 2 , x 2 − y 2 − u 2 ) = 0, where G is
an arbitrary function.

MA 201: Partial Differential Equations Lecture - 4

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MA 201: Partial Differential Equations Lecture - 4

Uploaded by

Copyright:

Available Formats

MA 201: Partial Differential Equations

IIT Guwahati MA201 (2022)

IIT Guwahati MA201 (2022)

• Let us assume that an integral surface u = u(x, y ) of (1) can be

IIT Guwahati MA201 (2022)

• Let us assume that an integral surface u = u(x, y ) of (1) can be

• Equation (1) may be written as

aux + buy − c = ha, b, ci · hux , uy , −1i = 0. (2)

IIT Guwahati MA201 (2022)

• Let us assume that an integral surface u = u(x, y ) of (1) can be

• Equation (1) may be written as

aux + buy − c = ha, b, ci · hux , uy , −1i = 0. (2)

IIT Guwahati MA201 (2022)

r′ (t) = hx ′ (t), y ′ (t), u ′ (t)i, (3)

for some t belonging to an interval I .

IIT Guwahati MA201 (2022)

r′ (t) = hx ′ (t), y ′ (t), u ′ (t)i, (3)

for some t belonging to an interval I .

Then, we must have

The solutions of (4) are called the

IIT Guwahati MA201 (2022)

r′ (t) = hx ′ (t), y ′ (t), u ′ (t)i, (3)

for some t belonging to an interval I .

Then, we must have

IIT Guwahati MA201 (2022)

r′ (t) = hx ′ (t), y ′ (t), u ′ (t)i, (3)

for some t belonging to an interval I .

Then, we must have

IIT Guwahati MA201 (2022)

IIT Guwahati MA201 (2022)

IIT Guwahati MA201 (2022)

• These are nonlinear equations in x and y , and we do not go ahead

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

IIT Guwahati MA201 (2022)

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

How to establish it?

IIT Guwahati MA201 (2022)

IIT Guwahati MA201 (2022)

IIT Guwahati MA201 (2022)

IIT Guwahati MA201 (2022)

IIT Guwahati MA201 (2022)

• Nontrivial solutions for ∂G ∂G

IIT Guwahati MA201 (2022)

• Nontrivial solutions for ∂G ∂G

• Expanding this determinant obviously gives a first-order quasi-linear

IIT Guwahati MA201 (2022)

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

therefore, G (φ, ψ) = 0 is a solution for the equation.

IIT Guwahati MA201 (2022)

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

therefore, G (φ, ψ) = 0 is a solution for the equation. Since G is

IIT Guwahati MA201 (2022)

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

therefore, G (φ, ψ) = 0 is a solution for the equation. Since G is

IIT Guwahati MA201 (2022)

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

therefore, G (φ, ψ) = 0 is a solution for the equation. Since G is

IIT Guwahati MA201 (2022)

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)

therefore, G (φ, ψ) = 0 is a solution for the equation. Since G is

• How? Both equations (8) and (9) will be identical if we select

IIT Guwahati MA201 (2022)

∂u ∂(φ, ψ) ∂u ∂(φ, ψ) ∂(φ, ψ)