Partial Differential Equation

Let σ be an open subset of Rn and let u : σ → R be a given

function.Directional derivative of u at x ∈ σ,along the vector ζ ∈ Rn denoted
u(x +hζ)−u(x )
as ∂u
∂ζ (x ) = limh→0 h ,provided that the limit exists.
The directional derivative u at x ∈ σ , along the xi axis, is called the i-th
partial derivative of u at x and is given by
uxi = ∂x∂u
i
(x ) = limh→0 u(x +hehi )−u(x ) .ei = (0, 0, ........, 1, 0, 0, 0); i = 1, 2, ....., n

1 / 106
 PDE
Schwartz’s multi-index notation for derivative:A multi-index α ∈ Zn is an
n-tuple α = (α1 , α2 , ....., αn ) of non-negative integers.Let
|α| = α1 + ...... + αn .If α and β two multi-indices,then α ≤ β means αi ≤ βi .
x α = x1α1 .....xnαn ,for any x ∈ Rn .A k-degree polynomial in n variables can be
represented as |α|≤k aα x α .
P

The partial differential operator of order α is denoted as,

α1 αn |α|
D α = ∂x∂ 1 α1 · · · ∂x∂ n αn = ∂x1 α1∂.....∂xn αn .we let D k u(x ) = {D α u(x ) : |α| = k}.For
k = 1 we get Du and we regard Du as being arranged in a vector,∇ = Du =
(D (1,0,0,0....,0) , D (0,1,0,0,0....,0) , .....D (0,0,0.....1) ) = ( ∂x∂ 1 , ∂x∂ 2 , ....., ∂x∂ n ),We call this
the gradient vector.Now 4 = ∇ · ∇ is called the Laplacian and
2
4 = i=1(n) ∂x∂ i 2 .
P

2 / 106
 PDE

For k = 2 , we regard D 2 as being arranged in a matrix form (called the

Hessian matrix),
∂2 ∂2
 
∂x 1
2 · · · ∂x1 ∂xn
 ∂2 ∂2 

∂x2 ∂x1 · · · ∂x2 ∂xn 
D2 = 

 .. . .. 

 . .. . 
 
∂2 2
· · · ∂x∂ n 2
∂xn ∂x1
The trace of a Hessian matrix is called the Laplace operator , denoted as
2
4 = i=1(n) ∂x∂ i 2 .
P
k
Note that D k u(x ) can be regarded as a vector in Rn .

3 / 106
 PDE
A partial differential equation (or briefly a PDE) is a mathematical equation that involves
two or more independent variables, an unknown function (dependent on those variables),
and partial derivatives of the unknown function with respect to the independent variables.
Definition 0.0.1 (Order of a PDE)
The order of a PDE is defined to be the order of the highest partial derivative occurring in
the equation.
e.g, ut + aux = 0 (ist-order PDE),uxx + uyy = 0(IInd-order PDE), ut + kuxx = 0(IInd-order
PDE).
Definition 0.0.2 (Degree of PDE)
The degree of a PDE is defined to be the degree of the highest order derivative occurring in
the equation, after the equation has been rationalized. eg. all above PDEs are of first degree.

4 / 106
 PDE

Let Ω be a connected open subset of Rn . A k-th order Partial Differential

k k−1
equation is a given by a map F : Rn × Rn × .. × Rn × R × Ω → R having
the form F (D k u(x ), D k−1 u(x ), ...., Du(x ), u(x ), x ) = 0, 99K (1)
for each x ∈ Ω and u : Ω → R is unknown. For instance, a first order PDE is
represented as F (Du(x ), u(x ), x ) = 0 and second order pde is
F (D 2 u(x ), Du(x ), u(x ), x ) = 0. A fi
rst order PDE with three variable unknown function u(x , y , z) is written as
F (ux , uy , uz , u, x , y , z) = 0 with F depending at least, on one of ux , uy and uz .

5 / 106
 PDE
Definition 0.0.3
we say u : Ω → R is a solution of the k-th ordere PDE(1) if (a) ∂ α u exists for all α. (b) u
satisfies the equation (1).
Solution of a PDE not necessarily continuous, eg, ux (x , y ) = 0 in R2 . By integrating w.r.t x
we have u(x , y ) = f (y ). So for any choice of f : R → R there is a solution u of the PDE. f
may be discontinuous.
Definition 0.0.4
We say PDE F is linear if (1) has the form |α|≤k aα (x )D α u(x ) = f (x ) for given function f
P

and aα (|α| ≤ k).If f = 0,we say F is homogeneous.

Example 0.0.5
• xuy − yux = u is linear and homogeneous pde.
• xux + yuy = x 2 + y 2 is linear pde.
• utt − c 2 uxx = f (x , t) is linear PDE.

6 / 106
 PDE

F is said to be semilinear, if it is linear only in the highest order, i.e., F has

the form |α|=k aα (x )D α u(x ) + a0 (D k−1 u, ....., Du, u, x ) = 0
P

Example 0.0.6
• ux + uy − u 2 = 0 is semilinear PDE.
• ut + uux + uxxx = 0 is semilinear PDE.
2
• utt + uxxx = 0 is semilinear PDE.

7 / 106
 PDE 5
Definition 0.0.7
We
X say k-th order PDE F is quasi-linear if it has the form
aα (D k−1 u(x ), ......, Du(x ), u(x ), x )D α u + a0 (D k−1 u, ....., Du, u, x ) = 0, i.e., the
|α|=k
coefficient of its highest (k-th ) order derivative depends on u and its derivatives only upto
the previous(k − 1)-th order.
we say F is fully-nonlinear if it depends nonlinearly on the highest(k-th) order derivatives.
Example 0.0.8
(1) ux + uuy − u 2 = 0 is quasi-linear
(2)ux u + uy = 0 is quasi linear.
(3) ux uy − u = 0 is non-linear
(4) ux2 + uy2 = 1 is non-linear .

8 / 106
 PDE
Example 0.0.9
1. consider the equation ux (x , y ) = 0. Integrating both sides we get u(x , y ) = f (y ),where
f ∈ C1
2.Consider the PDE uxy = 4x 2 y 3.Consider the PDE ux (x , y ) = uy (x , y )
We try to find the general solutions and particular solution of first order PDE. A general first
PDE has the form F (∇u(x ), u(x ), x ) = 0
Example 0.0.10
Find the first order PDE by eliminating the arbitrary function f , satisfied by u.
(i) u(x , y ) = xy + f (x 2 + y 2 )
(ii) u(x , y ) = f (x /y )

9 / 106
 PDE
Proof.
(i): Differentiating the given equation w.r.t x and y , we get ux = y + 2xf 0 , uy = x + 2yf 0 ,
respectively. Eliminating f 0 by multiplying y and x respectively, we get yux − xuy = y 2 − x 2 .
(ii): Do it in your own.

Example 0.0.11
Find the first order PDE, by eliminating the arbitrary constants a and b, satisfied by u
(i) : u(x , y ) = ax + by
Proof.
Differentiating the given equation w.r.t x and y , we get ux = a, uy = b, respectively.
Eliminating a and b we get xux + yuy = u.

10 / 106
 PDE

Family of curves We already observed that the solutions of PDE occur as

family of curves given by a constant or arbitrary function. In fact, by
eliminating the constant or function, via differentiation may lead to the
differential equation it solves. We now look at some family of curves which
arise as a solution to first order PDE’s.

11 / 106
 PDE

Let A ⊂ R2 be an open subset that represents a parameter set and consider

u : R2 × A → R a two parameter family of smooth surfaces in R3 ,u(x , y , a, b),
where (a, b) ∈ A
Example 0.0.12
Consider the family of circles u(x , y , a, b) = (x − a)2 + (y − b)2 .Thus ,
ux = 2(x − a) and uy = 2(y − b) and eliminating a and b , we get thepde
ux2 + uy2 − 4u = 0.Whose solutions are the given surfaces u.

12 / 106
 PDE

Normal Vector of a surface

Let S(x , y , z) = 0 be the equation of a surface S in R3 .Let us fix a point
p0 = (x0 , y0 , z0 ) ∈ S. We need to find the normal vector at p0 for the surface
S.Let us fix an arbitrary CURVE C lying on the surface passing through the
point p0 . Let the parametrized form of the curve r (t) = (x (t), y (t), z(t))
such that r (t0 ) = p0 . Since the curve C ≡ r (t) lies on the surface for all t,we
have S(r (t)) = 0. Thus, S(x (t), y (t), z(t)) = 0. Differentiation w.r.t. t (
dx (t) ∂S dy (t) ∂S dz(t)
using chain rule), we get ∂S ∂x dt + ∂y dt + ∂z dt = 0,
(Sx , Sy , Sz ) · (x 0 (t), y 0 (t), z 0 (t)) = 0 , ∇S(r (t)) · r 0 (t) = 0.

13 / 106
 Normal Vector of Surface

Since r 0 (t0 ) is the slope of the tangent at t0 to the curve C , we see that the
vector ∇S(p0 ) is perpendicular to the tangent vector at p0 .Since this
argument is true for any curve taht passes through p0 . we have that ∇S(p0 )
is normal vector to the tangent plane at p0 In particular , the equation of the
surface is given as S(x , y , z) = u(x , y ) − z, for some u : R2 → R, then
∇S(p0 ) = (∇u(x0 , y0 ), −1).

14 / 106
 How to solve 1st-order Quasi-linear equation
To understand the derivation of the characteristic equation and the geometry involved,
explained in this section, let us see what is happening two varibale first order
quasi-linear equation:

a(x , y , u)ux + b(x , y , u)uy = c(x , y , u) (1)

Solving for u(x , y ) in the above equation 1 is equivalent to finding the surface
S = {(x , y , u(x , y ))} generated by u in R3 . If u is a solution of equation 1 at each
(x , y ) in the domain of u,
a(x , y , u)ux + b(x , y , u)uy = c(x , y , u)
a(x , y , u)ux + b(x , y , u)uy − c(x , y , u) = 0
(a(x , y , u), b(x , y , u), c(x , y , u)) · (ux , uy , −1) = 0
(a(x , y , u), b(x , y , u), c(x , y , u)) · (∇u(x , y ), −1) = 0

15 / 106
 Continue....
But (∇u(x , y ), −1) is normal to S at the point (x , y ). Hence, the coefficients
(a(x , y , u), b(x , y , u), c(x , y , u)) are perpendicular to the normal. Thus, the coefficients
(a(x , y , u), b(x , y , u), c(x , y , u)) lie on the tangent plane to S at (x , y , u(x , y )). Hence,
finding u is equivalent to finding the integral surface corresponding to the coefficient
vector field V = (a(x , y , u), b(x , y , u), c(x , y , u)). The surface is the union of curves
which satisfy the property of S. Thus, for any curve Γ ⊂ S such that at each point of
Γ, the vector V (x , y ) = (a(x , y , u), b(x , y , u), c(x , y , u) is tangent to the curve.
Parametrizing the curve Γ by the variable s, we see that we are looking for the curve
Γ = {x (s), y (s), z(s)} ⊂ R3 such that
dx dy
ds = a(x (s), y (s), u(x (s), y (s))), ds = b(x (s), y (s), u(x (s), y (s))),
dz
ds = c(x (s), y (s), u(x (s), y (s))). The three ODE’s obtained are called characteristic
equations. The union of these characteristic (integral) curves give us the integral
surface.

16 / 106
 Continue....

Example 0.0.13
Consider the PDE (Linear Transport Equation) in two variables : ut + bux = 0,
x ∈ R, t ∈ (0, ∞), where the constant b in R is given. The characteristic
equations are dx dt dz
ds = b, ds = 1, ds = 0. Solving the three ODE’s, we get
x (s) = bs + c1 , t(s) = s + c2 , and z(s) = c3 . Eliminating the parameter s, we
get the curves (lines) x − bt = a constant and z =a constant. Therefore,
u(x , t) = f (x − bt) is the general solution, for an arbitrary function f .

17 / 106
 Continue....
Example 0.0.14
Let us compute the general solution (in terms of arbitrary functions) of the first order
dy
PDE ux + 2xuy = u 2 . The characteristic equations (ODE’s) are dx ds = 1, ds = 2x (s),
dz 2 2
ds = z (s). Solving we get x (s) = s + c1 , y (s) = s + 2c1 s + c2 and
z(s) = −1/(s + c3 ). Eliminating s between x and y , we get the characteristic curves
to be y − x 2 = a constant and x + 1/z = a constant. Thus, the general solution is
F (y − x 2 , x + 1/z) = 0. Explicitly, u(x , y ) = x −f (y1 −x 2 ) , for some arbitrary function f .
H.W: 1. Compute the general solution (in terms of arbitrary functions) of the first
order PDE yux − xuy = 2xyu.
2. Compute the general solution (in terms of arbitrary functions) of the first order PDE
2yux + uuy = 2yu 2 .

18 / 106
 Intregral surface

Definition 0.0.15
A curve in Rn is said to be an integral curve for a given vector field, if the
curve is tangent to the vector field at each of its point.
Similarly, a surface in Rn is said to be an integral surface for a given vector
field, if the surface is tangent to the vector field at each of its point.
In the spirit of the above definition and arguments, finding a solution to the
quasi-linear PDE is equivalent to finding an integral surface S corresponding
to the coefficient vector field V .

19 / 106
 Solution of 1st-order Non-linear PDE

SOLUTION OF Nonlinear- PDE The arguments in previous section can be

carried over to a general nonlinear first order PDE. Consider the first order
nonlinear PDE, F : Rn × R × Ω → R such that F (∇u(x ), u(x ), x ) = 0.99K (1)
in Ω, where Ω ⊂ Rn ,F is given and u is unknown to be found. We want find a
surface described by the solution u in Rn+1 . In the quasi-linear case,using the
equation, we obtained a unique direction (b, c) which is tangential,at each
point, to our desired surface S. In the non-linear case, however, we have no
such unique direction cropping out of the equation.For any point x ∈ Ω , the
point (x, z), where z = u(x ) is in the solution surface. Further,if p = ∇u(x )
then we have the relation F (p, z, x ) = 0. Thus, for a fixed (x , z) ∈ Ω × R,

20 / 106
 Solution of 1st order Non-linear PDE
we consider the more general equation F (p, z, x ) = 0 and denote the solution
set as V (x , z) = {p ∈ Rn | F (p, z, x ) = 0}.Therefore ,solving (1) is equivalent
to finding a u ∈ C 1 (Ω) such that , for all x ∈ Ω there is apair (x,z) for which
z = u(x ) and p = ∇u(x ).Every choice of p is a possibvle normal vector
candidate (p, −1) at (x0 , z0 ) on S.In general these family of normals envelope
a cone with vertex at (x0 , z0 ) perpendicular to S. As p varies in V (x , z), we
have we have a possible family of (tangent) planes through (x0 , z0 ) given by
the equation (z − z0 ) = p · (x − x0 ) where one of the planes is tangential to
the surface S . The envelope of this family is a cone C (x0 , z0 ),called a Monge
cone,with vertex at (x0 , z0 ).The envelope of the family of planes is that
surface which is tangent at each of its point to some palne from the family.

21 / 106
 Solution of 1st order Non-linear PDE

Definition 0.0.16
A surface S in Rn+1 is said to be an integral surface, if at each point
(x0 , z0 ) ∈ S ⊂ Rn × R it is tangential to the Monge cone with vertex at
(x0 , z0 )

22 / 106
 Method of characteristics, Charpits Methods
Let us consider the first order nonlinear PDE (1) in new independent variables p ∈ Rn ,
z ∈ R and x ∈ Ω consequently F (p, z, x ) = F (p1 , p2 , ..........., pn , z, x1 , x2 , ......., xn ) is a map
of 2n + 1 variable.We now introduce the derivatives (assume it exists) of F corresponding to
each variable,∇p F = (Fp1 , ........, Fpn ) , ∇x = (Fx1 , ., Fxn ).
The method of characteristics reduces a given first order PDE to a system of ODE. We must
choose a curve x(s) in Ω such that we can compute u and ∇u along this curve.
Differentiating the given PDE (1) w.r.t xi ,we get nj=1 Fpj uxj xi + Fz uxi + Fxi = 0, 99K (2)
P

we seek to find x(s) such that

Pn
j=1 Fpj (p(s), z(s), x (s))uxj xi (x (s)) + Fz (p(s), z(s), x (s))pi (s) + Fxi (p(s), z(s), x (s)) = 0
Now diff.pi w.r.t s, dpdsi (s) = nj=1 uxi xj (x (s)) dxds
j (s)
P

23 / 106
 Method of characteristics

Set dxdsj (s)

= Fpj (p(s), z(s), x (s)).Thus, dxds(s) = ∇p F (p(s), z(s), x (s)).99K (3)
Now substituting this in the first order equation, we get
dpi (s)
ds = −Fz (p(s), z(s), x (s))pi (s) − Fxi (p(s), z(s), x (s)).
Thus
dp(s)
ds = −Fz (p(s), z(s), x (s))pi (s) − ∇x F (p(s), z(s), x (s)),99K (4).Similarly,
we differentiate z(s) w.r.t s,
dz(s) Pn dxj (s) Pn
ds = j=1 uxj (x (s)) ds = j=1 uxj (x (s))Fpj (p(s), z(s), x (s))
dz(s)
Thus, ds = p(s) · ∇p F (p(s), z(s), x (s)),99K (5). We have 2n+1 first order
ODE called the characteristic equations of (1).

24 / 106
 Method of characteristics

In first order linear problem, the ODE reduces to, as follows: Let
F (∇u, u, x ) = b(x ) · ∇u(x ) + c(x )u(x ) = 0, x ∈ Ω ⊂ Rn .Then in new
variable F (p, z, x ) = b(x ) · p + c(x )z .Then dxds(s) = ∇p F =
b(x (s)).Also, dz(s)
ds = b(x (s)) · p(s) = b(x (s)) · ∇u(x (s)) = −c(x (s))z(s).

25 / 106
 Nonlinear Pde
In this section, we study the form of general solutions of a first order
PDE,i.e.,F (∇u(x ), u(x ), x ) = 0 inΩ.Let A ⊂ Rn be an open set which is the
parameter set.Let
us introduce the n ×(n + 1) matrix
ua1 ux1 a1 · · · uxn an
2
 . .. ... .. 
(Da u, Dxa u) =  ..

. .  
uan ux1 an · · · uxn an
Definition 0.0.17
A C 2 function u = u(x ; a)is said to be complete integral in Ω × A if u(x ; a)
2
solves the PDE for each a ∈ A and rank of the matrix (Da u, Dxa u) is n.
.

26 / 106
 Complete Integral
Example 0.0.18
Which one of the folowing is a solution of the nonlinear pde ux uy = u
(a)u(x , y , a, b) = xy + ab + ax + by
(b)u(x , y , a, b) = xy + ax + by
Envelopes and General Integrals
Definition 0.0.19
Let Ω ⊂ Rn and A ⊂ Rm be open subsets and let u = u(x ; a) be a C 1 -function of both
x and a .Suppose the equation Da u(x ; a) = 0 is solvable for a , as a C 1 -function of x ,
say a = φ(x ) i.e.,Da u(x ; φ(x )) = 0 then v (x ) = u(x , φ(x )) is the envelope of the
functions {u(x , a)}a∈A .
The idea is that for each x ∈ Ω, the graph of v is tangent to the graph of u(x , a) for
a = φ(x ).

27 / 106
 CompIete Integral

Theorem 0.0.20
Suppose for each a ∈ A, u(x; a) is a solution to the given PDE and the
envelope v of u ,given as v (x ) = u(x , φ(x )),exists then v also solves the given
PDE.
Envelope of a Non-linear PDE can be derived by the following definition.
Definition 0.0.21
The envelope w of the functions u(x ; a0 , h(a0 )) where
a0 = (a1 , a2 , .........., an−1 ) and h : A0 ⊂ Rn−1 → R(provided it exists) such
that Da0 u(x , a0 ) = 0 solvable for a0 .

28 / 106
 CompIete Integral
Example 0.0.22
Let us compute a complete integral of the first order PDE ux uy = u(x , y ). Let z = u.
∂u ∂u
Then the equation is of the form F (p, z, x ) = p1 p2 − z, where p1 = ∂x and p2 = ∂y .
The characteristic equations are ( dxds(s) , dyds(s) ) = (p2 (s), p1 (s)), dp(s)
ds = p(s). Then
dp1 dp2 s
ds = p1 (s) and ds = p2 (s). Thus, on integrating, we get p1 (s) = c1 e and
s dz dx (s) dy (s)
p2 (s) = c2 e . Now ds = p1 (s) ds + p2 (s) ds i.e.,
dz
ds = (p1 (s), p2 (s)) · (p2 (s), p2 (s)) = 2p1 (s)p2 (s). By the given PDE, we have
z(s) = c1 c2 e 2s . Using p, we solve for x to get x (s) = c2 e s + b and y (s) = c1 e s + a.
Therefore, z(x , y , a, b) = (x − b)(y − a). Thus, u(x , y , a, b) = (x − b)(y − a) is the
complete integral(solution ) of the given PDE.

29 / 106
 Complete Integral
Example 0.0.23
The complete integral of the non linear PDE ux uy = u is u(x , y , a, b) = xy + ab + ax + by .
Let h : R → R defined by h(a) = a ,then
u(x , y , a, h(a)) = u(x , y ; a, a) = xy + a2 + a(x + y ). Now Da u = 2a + x + y = 0 which
2
yields a = φ(x ) = −(x2+y ) Therefore the envelope w (x ) = u(x , y , φ(x ), h(φ(x )) = −(x 4−y ) .
Example 0.0.24
Compute a complete integral of the first order nonlinear pde ux uy = u.The equation is of the
form F (p, z, x ) = p1 p2 − z .The characteristics equations are ( dxds(s) , dyds(s) ) = (p2 (s), p1 (s))
, dp(s) s
ds = p(s). Thus, on integrating, we get p1 (s) = c1 e and p2 (s) = c2 e s . Therefore,
p1 c1 √ q
z
p2 = c2 = a say. Using this equation with p1 p2 = z, we get p1 = ± az and p2 = ± a .

30 / 106
 Complete Integral
Now,
dz dx dy
= (p1 (s), p2 (s)) · ( , )
ds ds ds
√ dx (s) q dy (s)
= ± az ± z/a
ds ds
1 dz √ dx (s) q dy (s)
√ = ±( a + 1/a )
z ds ds ds
√ √ y
2 z = ±( ax + √ ) + c
a
√
Thus, u(x , y ) = [b + 1/2( ax + √ya )]2 is a complete solution , if we had taken a > 0.

31 / 106
 Complete Integral
Like nonlinear ODEs, some nonlinear PDEs also have a singular solution (or singular
integral) that is obtained by constructing the envelope of the entire two-parameter
family of surfaces represented by the complete integral.
Definition 0.0.25
A solution which satisfies a partial differential equation but is not a member of the
family of surfaces represented by it is called a singular solution.
To summarize, the complete integral for a nonlinear PDE includes a rich variety of
solutions.
• The envelope of the entire two-parameter family is a solution called the singular
integral of the PDE.
• The complete integral is not unique, but any other complete integral for the PDE
can be obtained from it by the process of envelope formation.

32 / 106
 Complete Integral

Example 0.0.26
Find the complete integral, general solution and singular solution of the fully
nonlinear PDE ux2 + uy2 = 1 + 2u.
The equation of the PDE is F (p, z, x ) = p12 + p22 − 2z − 1 = 0. The
characteristics equations are ( dxds(s) , dyds(s) ) = (2p1 (s), 2p2 (s)),
dp(s) dp1 (s) p1 (s)
ds = (2p1 (s), 2p2 (s)). Thus, dp2 (s) = p2 (s) . On integration we have,
p1 (s)
p2 (s) = a, constant. Using the PDE, we get (1 + a2 )p22 = 1 + 2z. Thus,
q q
1+2z 1+2z
p2 = ± 1+a2 and p1 = ±a 1+a2 .

33 / 106
 Complete Integral
Now,
dz dx dy
= (p1 (s), p2 (s)) · ( , )
ds s ds ds
1 + 2z dx (s) dy (s)
=± (a + )
1 + a2 ds ds
s
1 dz 1 dx (s) dy (s)
=± 2
(a + )
1 + 2z ds 1+a ds ds
√ ax + y
1 + 2z = ± √ ±b
1 + a2
Thus, u(x , y ) = 21 ( √ax1+a
+y 2 1
2 + b) − 2 is a complete integral. Here we see that

no choice of a and b will give the constant solution u = − 21 . Thus, u = − 12 is

the singular solution.
34 / 106
 Cauchy Problem
Example 0.0.27 (Linear Transport equation)
Consider the linear transport equation in two variables, ut + bux = 0, x ∈ R
and t ∈ (0, ∞). Where the constant b is given. Thus the given vector field
V (x , t) = (b, 1, 0). The Characteristic equation are dx dt dz
ds = b, ds = 1, ds = 0.
Solving these three ode’s , we get x (s) = bs + c1 , t(s) = s + c2 , and
z(s) = c3 . Note that solving the system of ODE’s requires some initial
condition.
If u is prescribed on the curve (x , 0) in the xt-plane., Then we can solve the
above problem.
Thus, the problem of finding a function u satisfying the given first order PDE
such that u is known on a curve Γ in the xy -plane is called the Cauchy
problem.
35 / 106
 Cauchy Problem
Definition 0.0.28
A Cauchy problem states that: given a hyper-surface Γ ⊂ Rn , can we find a
solution u of F (x , u(x ), ∇u(x )) = 0, whose graph contains Γ?.
Consider the general first order qausilinear PDE with n independent variables

F (x , u, ∇u(x )) = b(x , u(x )) · ∇u(x ) − c(x , u(x )) = 0 (2)

where b = (b1 , b2 , · · · , bn ) is a cofficient vector with n components.

Definition 0.0.29
We say a hyper-surface Γ ⊂ Ω ⊂ Rn is non-characteristics w.r.t., the
quasilinear PDE [2] if ni=1 bi (x , u(x ))υi (x ) 6= 0, ∀x ∈ Γ, where υ(x ) is the
P

normal vector of Γ at x.
36 / 106
 Cauchy Problem

For instance, in the two dimensional case, Γ = {γ1 (r ), γ2 (r )} ⊂ Ω ⊂ R2 is

non-characteristic for the quasi-linear Cauchy problem

a(x , y , u)ux + b(x , y , u)uy = c(x , y , u), (x , y ) ∈ Ω

u = g on Γ

if Γ is nowhere tangent to (a(γ1 , γ2 , g), b(γ1 , γ2 , g)), i.e.,

(a(γ1 , γ2 , g), b(γ1 , γ2 , g)) · (−γ20 , γ10 ) 6= 0.

37 / 106
 Cauchy problem
Example 0.0.30
Let g be a given function g : R −→ R. Consider the linear transport equation ut + bux = 0,
x ∈ R, t ∈ (0, ∞) such that u(x , 0) = g(x ), x ∈ R. We parametrize the curve Γ with
r-variable, i.e., Γ = {γ1 (r ), γ2 (r )} = (r , 0). the Ch. equations are dx ds (r ,s)
= b, dt(r
ds
,s)
= 1,
dz(r ,s)
and ds = 0.
with initial conditions x (r , 0) = r , t(r , 0) = 0, and z(r , 0) = g(r ). Solving the ODEs , we
get x (r , s) = bs + c1 (r ), t(r , s) = s + c2 (r ), and z(r , s) = c3 (r ), with initial conditions
x (r , 0) = c1 (r ) = r , t(r , 0) = c2 (r ) = 0, and z(r , 0) = c3 (r ) = g(r ). Therefore
x (r , s) = bs + r , t(r , s) = s, and z(r , s) = g(r ). The idea is to solve for r and s in terms of
x and t. Let us set u(x , t) = z(r (x , t), s(x , t)). In this case we can solve for r and s in terms
of x and t, to get r (x , t) = x − bt, s(x , t) = t. Therefore,
u(x , t) = z(r , s) = g(r ) = g(x − bt).

38 / 106
 Cauchy problem
Let us study the above example with a different data curve Γ.
Example 0.0.31
let g be a given function g : R −→ R. Consider the linear transport equation ut + bux = 0,
x ∈ R, t ∈ (0, ∞). such that u(bt, t) = g(t), t ∈ (0, ∞).
Solution 0.0.32
we parametrize the curve Γ with r-variable, i.e., Γ = {γ1 (r ), γ2 (r )} = (br , r ). The Ch.
equations are dx ds (r ,s)
= b, dt(r
ds
,s)
= 1, and dz(r
ds
,s)
= 0 with initial conditions x (r , 0) = br ,
t(r , 0) = r , and z(r , 0) = g(r ). Solving the ODEs, we get
x (r , s) = bs + c1 (r ),t(r , s) = s + c2 (r ), and z(r , s) = c3 (r ) with x (r , 0) = c1 (r ) = br ,
t(r , 0) = c2 (r ) = r , and z(r , 0) = c3 (r ) = g(r ). Therefore x (r , s) = b(s + r ),
t(r , s) = s + r , z(r , s) = g(r ). Note that in this case we can not solve for r and s in terms
of x , t.

39 / 106
 Cauchy problem
For instance, in the two dimension case,Γ = {γ1 (r ), γ2 (r )} ⊂ Ω ⊂ R2 is
non-characteristic for the quasi linear Cauchy problem
a(x , y , u)ux + b(x , y , u)uy = c(x , y , u), (x , y ) ∈ Ω, 99K (1),
u=g on Γ.If Γ is nowhere tangent to (a(γ1 , γ2 , g), b(γ1 , γ2 , g)),i.e.,
(a(γ1 , γ2 , g), b(γ1 , γ2 , g)) · (−γ20 , γ10 ) 6= 0.
Example 0.0.33
Consider the equation 2ux (x , y ) + 3uy (x , y ) = 1 in R2 . Let Γ be a straight
liney = mx + c in R2 .The parametrisation of the line is Γ(r ) = (r , mr + c) for
r ∈ R.
Therefore,(a(γ1 , γ2 , g), b(γ1 , γ2 , g)) · (−γ20 , γ10 ) = (2, 3) · (−m, 1) = −2m + 3.
Thus, the line is not a non-characteristic for m = 32 , i.e., all lines with slope 32
is not a non-characteristic.
40 / 106
 Cauchy problem

Theorem 0.0.34
Let a, b and c, the coefficients of (1) have continuous partial derivatives w.r.t
x, y, u.Let Γ(r ) = (γ1 (r ), γ2 (r )) be the parametraization of an initial curve on
which u(γ1 (r ), γ2 (r )) = φ(r ) such that γ1 , γ2 and φ are continuously
differentiable and the initial curve is non-characteristic
i.e.,(a(γ1 , γ2 , g), b(γ1 , γ2 , g)) · (−γ20 , γ10 ) 6= 0.Then there exists a unique
solution u(x, y) in some neighborhood of Γ which satisfies (1).

41 / 106
 Cauchy problem

Example 0.0.35
Find the general solution (in terms of arbitrary functions) of the first order
PDE 2ux (x , y ) + 3uy (x , y ) + 8u(x , y ) = 0, For the PDE given above, check
for the characteristic property of the following curves
(a) y = x in the xy-plane.
(b) y = 3x2−1 .
Discuss the particular solution of the above PDE corresponding to
(a)u(x , x ) = x 4 on y = x .

42 / 106
 Cauchy problem
Example 0.0.36
Let Ω = {(x , y ) ∈ R2 |x > 0, y > 0}. Let Γ = {(x , 0)|x > 0}.
Consider the linear PDE xuy (x , y ) − yux (x , y ) = u(x , y ) in Ω,
u(x , 0) = φ(x ) on Γ
Solution 0.0.37
Note that in this example the data curve Γ is parametrised using the variable r
and the characteristic curves is parametrised using the variable s.
The parametrisation of the initial curve is Γ(r ) = (r , 0) for
r > 0.Therefore,(a(γ1 (r ), γ2 (r )), b(γ1 (r ), γ2 (r ))) · (-
γ20 (r ), γ10 (r )) = (0, r ) · (−γ20 (r ), γ10 (r )) = r 6= 0.Hence, the given initial curve is
non-characteristic. The characteristic equations are

43 / 106
 Continue
dx (r ,s)
ds = −y ; dy ds(r ,s)
= x ; dz(r
ds
,s)
= z(r , s), With initial conditions
x (r , 0) = r , y (r , 0) = 0, z(r , 0) = φ(r ).
2 2
Then d xds(r ,s) = −x (r , s), d yds(r ,s) = −y (r , s). Then we have,
x (r , s) = c1 (r ) cos s + c2 (r ) sin s and y (r , s) = c3 (r ) cos s + c4 (r ) sin s. Using
initial conditions, We get c1 (r ) = r , c3 (r ) = 0. Also, dx ds (r ,s)
s=0
= −y (r , 0) we
shall get c2 (r ) = 0. Similarly, c4 (r ) = r . Also, z(r , s) = c5 (r )e s , where
c5 (r ) = φ(r ). Thus we have (x (r , s), q y (r , s)) = (r cos s, r sin s), where
r > 0,and 0 ≤ s ≤ 2 . Hence, r = x 2 + y 2 and s = tan−1 (y /x ). Thus for
π

any given (x , y ) ∈ Ω we have q

−1
u(x , y ) = z(r , s) = φ(r )e s = φ( x 2 + y 2 )e tan (y /x ) .

44 / 106
 Continue

Home work: Let Ω = {(x , y ) ∈ R2 : y > 0} and Γ = {(x , 0) : x ∈ R}. Solve

the following Cauchy problem

ux (x , y ) + uy (x , y ) = u 2 (x , y ) in Ω
u(x , 0) = φ(x ) on Γ

45 / 106
 Cauchy problem for non-linear pde
Definition 0.0.38
We sayΓ ⊂ Ω ⊂ Rn is non-characteristic for the nonlinear Cauchy problem
F (∇u(x ), u(x ), x ) = 0, x ∈ Ω,
u = g on Γ. if Γ is nowhere tangent to the Monge cone, i.e., there exists function v such
that F (v (r ), g(r ), γ1 (r ), ..., γn (r )) = 0 and satisfies ni=1 Fpi (v , g, γ1 , ...., γn ) · υ(x ) 6= 0,
P

where υ(x ) is normal to Γ at x.

In particular, in the two dimension case,Γ = {γ1 (r ), γ2 (r )} ⊂ Ω ⊂ R2 is non-characteristic
for the nonlinear Cauchy problem Γ is nowhere tangent to monge cone.We need initially
p1 = α(r ), p2 = β(r ) such that F (α(r ), β(r ), g(r ), γ1 (r ), γ2 (r )) = 0 and we get another
relation g 0 (r ) = α(r )γ10 (r ) + β(r )γ20 (r ).Γ satisfies
(Fp2 (γ1 , γ2 , g, α, β), Fp1 (γ1 , γ2 , g, α, β)) · (−γ20 , γ10 ) 6= 0

46 / 106
 cauchy problem for non linear pde

Example 0.0.39
Let Ω = {(x , y ) ∈ R2 |x > 0}.Let Γ = {(0, y )|y ∈ R} Consider fully nonlinear
pde ux uy = u(x , y ) in Ω.
u(0, y ) = y 2 on Γ. The parametrisation of the initial curve is Γ(r ) = (0, r ) for
all r ∈ R. We first look for the functions α and β such that α(r )β(r ) = r 2 .
and 2r = β(r ).solving we get β(r ) = 2r and α(r ) = 2r .Again we have
Fp2 γ20 (r ) = p1 = α(r ) = 2r 6= 0 for r 6= 0.Hence the given initial curve is
non-characteristic.

47 / 106
 continue..

The characteristic equations are ( dx ds (r ,s) dy (r ,s)

, ds ) = (p2 (r , s), p1 (r , s)).
dp(r ,s)
ds = p(r , s).
dz(r ,s)
ds = 2p1 (r , s)p2 (r , s).With initial conditions
x (r , 0) = 0, y (r , 0) = r , z(r , 0) = r 2 , p1 (r , 0) = α(r ), p2 (r , 0) = β(r ).
p1 (r , s) = c1 (r )e s ,p2 (r , s) = c2 (r )e s , for all s ∈ R.Solve for x we get
x (r , s) = 2r (e s − 1),y (r , s) = 2r (e s + 1) Solving z we get
2
z(r , s) = r2 (e 2s + 1).solving r and s in terms x and y we get .

48 / 106
 Classification of PDE
A general second order PDE is of the form F (D 2 u(x ), Du(x ), u(x ), x ) = 0 for each
x ∈ Ω ⊂ Rn and u : Ω → R is the unknown. A Cauchy problem poses the following:
Given the knowledge of u on a smooth hypersurface Γ ⊂ Ω can one find the solution u
of the PDE? The knowledge of u on Γ is said to be the Cauchy data. Viewing the
Cauchy problem as an initial value problem corresponding to ODE, we know that a
unique solution exists to the second order ODE y 00 (x ) + P(x )y 0 (x ) + Q(x )y (x ) = 0,
x ∈ I.
y (x0 ) = y0 ,
y 0 (x0 ) = y00 , where P and Q are continuous on I and for any point x0 ∈ I. This
motivates us to define the Cauchy problem for second order PDE as:
F (D 2 u(x ), Du(x ), u(x ), x ) = 0, x ∈ Ω,
u(x ) = g(x ),
Du(x ) · υ(x ) = h(x ),x ∈ Γ. where υ is the outward unit normal vector on the
hypersurface Γ and g,h are the known functions on Γ.

49 / 106
 Continue.......
Consider the general second order qausilinear PDE with n independent variable

F (x , u, du, D 2 u) = A(x ) · D 2 u − D(∇u, u, x ) (3)

where A = Aij is an n × n matrix with entries Aij (x , u, ∇u), D 2 u is the hessian

matrix. Since we demand the solution to be in C 2 , the mixed derivatives are
equal and we can assume, without loss generality that, matrix A is symmetric.
t
In fact if A is not symmetric, we can replace A with As = (A+A 2
)
, which is
symmteric since A · D 2 u = As · D 2 u.
Definition 0.0.40
We say a hypersurface Γ ⊂ Ω ⊂ Rn is non characteristic w.r.t the PDE if
Pn Pn
i=1 j=1 Aij (x )υi (x )υj (x ) 6= 0,where υ(x ) is the normal vector of Γ at x.

50 / 106
 continue...
Since any real symmetric matrix can always be diagonalised, there is a
coordinate transformation T (x ) such that the matrix T (x )A(x )T t (x ) is
diagonal with diagonal entries, say λ1 , λ2 , ....., λn . Since A(x ) is real
symmetric all λi ∈ R. Thus, we classify PDE at a point x ∈ Ω based on the
eigen values of the matrix A(x ). Let p denote the number of eigenvalues that
are strictly positive and z denote the number of zero eigenvalues.
Definition 0.0.41
We say a PDE is hyperbolic at a point x ∈ Ω, if z = 0 and either p = 1 or
p = n − 1. We say it is parabolic if z > 0. we say it is elliptic , if z = 0 and
either p = n or p = 0.

51 / 106
 continue..
Example 0.0.42
Classify the PDE: uxx + uyy + uzz + 2uyz = 0.
Consider the Cauchy problem for the second order semi-linear PDE in two variables
(x , y ) ∈ Ω ⊂ R2 ,

A(x , y )uxx + 2B(x , y )uxy + C (x , y )uyy = D

u(x , y ) = g(x , y ) (x , y ) ∈ Ω
(4)
ux = h1 (x , y ), (x , y ) ∈ Γ
uy (x , y ) = h2 (x , y ), (x , y ) ∈ Γ

where D(x , y , u, Ux , uy ) may be nonlinear and Γ is a smooth curve in Ω. Now

Γ(s) = (γ1 (s), γ2 (s)) be the parametrization of Γ. Then we have the compatibility
condition that ġ(s) = h1 γ˙1 (s) + h2 γ˙2 (s). By computing the second derivatives of u on
Γ and considering uxx , uyy , uxy as unknowns, we have the linear system of three
equations in three unknowns on Γ
52 / 106
 continue
Auxx + 2Buxy + Cuyy = D
γ˙1 (s)uxx + γ˙2 (s)uxy = h˙1 (s) (5)
γ˙1 (s)uxx + γ˙2 (s)uxy = h˙2 (s)

This system of equation is solvable if the determinant of the coefficients are non-zero,
i.e.,

A 2B C
γ˙1 γ˙2 0 6= 0
0 γ˙1 γ˙2

Definition 0.0.43
we say a curve Γ ⊂ ω ⊂ R2 is characteristic w.r.t the PDE if
Aγ˙2 2 − 2B γ˙1 γ˙2 + C γ˙1 2 = 0. where (γ1 (s), γ2 (s)) is a parametrization of γ.
53 / 106
 continue..

Note that the geometry hidden in the above definition is very similar to that
we encountered in first order equation. Since υ = (−γ˙2 , γ˙1 ) is the normal to Γ
at each point, the above definition says that the curve is noncharacteristic if
Pn Pn 2 2
i=1 j=1 Aij υi υj = Aγ˙2 − 2B γ˙1 γ˙2 + C γ˙1 6= 0, where
A11 = A, A12 = A21 = B, A22 = C . If y = y (x ) is a representation of the
curve Γ we have γ1 (s) = s, γ2 (s) = y (s). Then the characteristic equation
reduces as A( dy 2 dy
dx ) − 2B dx + C = 0. Therefore, the√characteristic curves of
B± B 2 −AC
are given by the graphs whose equation is dy dx = A .

54 / 106
 continue..
Thus, we have three situations arising depending on the sign of the discriminant,
B 2 − AC . This classifies the given second order PDE based on the sign of its
discriminant d = B 2 − AC . We say a second order PDE
Definition 0.0.44
We say a second order PDE is of
(a) hyperbolic type if d > 0,
(b) parabolic type if d = 0,
(c) elliptic type if d < 0.
Example 0.0.45
Consider the quasilinear PDE uxx − uuyy = 0. The discriminant is d = u. So, it is
hyperbolic for {u > 0}, elliptic when {u < 0} and parabolic when {u = 0}.
Example 0.0.46
Find the characteristics of the pde for hyperbolic type uxx + 2uxy + sin2 (x )uyy + uy = 0
55 / 106
Invariance of Discriminant, Reduction to a simple form

The classification of second order semi-linear PDE is based on the

discriminant B 2 − AC . In this section, we note that the classification is
independent of the choice of coordinate system (to represent a PDE).
Consider the two-variable semilinear PDE:
A(x , y )uxx + 2B(x , y )uxy + C (x , y )uyy = D(x , y , u, ux , uy ), (x , y ) ∈ Ω ⊂ R2 .
99K (3). Also, one of the coefficients A,B or C is identically non-zero
otherwise it is not second order PDE., and D may appear non-linearly. We
shall observe how (3) changes under coordinate transformation.

56 / 106
 continue..
Definition 0.0.47
For any PDE of the form (3) we define its discriminant as B 2 − AC .
Let T : R2 → R2 be the coordinate transformation as
T (x , y ) = (w (x , y ), z(x , y )) ,where w , z : R2 → R. We assume that
w(x,y),z(x,y) are such that w, z are both continuous and twice differentiable
w.r.t (x , y ) , and the Jacobian J of T is non-zero.

wx wy
6 0
=
zx zy

57 / 106
 continue..
We compute the derivatives of u in the new variable,
u x = u w wx + u z z x ,
uy = uw wy + uz zy ,
uxx = uww wx2 + 2uwz wx zx + uzz zx2 + uw wxx + uz zxx ,
uyy = uww wy2 + 2uwz wy zy + uzz zy2 + uw wyy + uz zyy ,
uxy = uww wx wy + uwz (wx zy + wy zx ) + uzz zx zy + uw wxy + uz zxy . Substituting above
equations in (3), we get a(w , z)uww + 2b(w , z)uwz + c(w , z)uzz = d(w , z, u, uw , uz ).
where D transform into d and
a(w , z) = Awx2 + 2Bwx wy + Cwy2
b(w , z) = Awx zx + B(wx zy + wy zx ) + Cwy zy (6)
c(w , z) = Azx2 + 2Bzx zy + Czy2

Note that the coefficients in the new coordinate system satisfy

b 2 − ac = (B 2 − AC )J 2 . Thus both b 2 − ac abd B 2 − AC have the same sign. Thus,
the sign of the discriminant is invariant under coordinate transformation. All the above
arguments can be carried over to quasi-linear and non-linear PDE.
58 / 106
 Simple form
The advantage of above classification helps us in reducing a given PDE into simple
forms. Given a PDE, one can compute the sign of the discriminant and depending on
its clasification we can choose a coordinate transformation (w , z) such that
(i) For hyperbolic, a = c = 0 or b =0 and a = −c.
(ii) For parabolic, c = b = 0 or a = b = 0. We conveniently choose c = b = 0
situation so that a 6= 0 (so that division by zero is avoided in the equation for
characteristic curves).
(iii) For elliptic, b = 0 and a = c.
If B 2 − AC > 0 we have two real characteristics. We are √ looking for the coordinate
wx −B± B 2 −AC
system w,z such that a = c = 0. This gives us wy = A = zzyx .

59 / 106
 Simple form

Therefore, we need to find w and √ z such that along the slopes of the
dy B± B 2 −AC
characteristic curves, dx = A = −w
wy . This means that, using the
x

parametrisation of the characteristic curves, wx γ˙1 (s) + wy γ˙2 (s) = 0 and hence
dw
ds = 0. Similarly, for z. Thus, w and z are chosen such that they are
constant on the √characteristic curves. The characteristic curves are found by
B± B 2 −AC
solving dy
dx = A , and the coordinates are then chosen such that along
the characteristic curve w (x , y ) = a constnat and z(x , y ) = a constnat .

60 / 106
 continue..
Example 0.0.48
Reduce the PDE uxx − c 2 uyy = 0 to its canonical form. Note that A = 1, B = 0, C = −c 2 and
B 2 − AC = c 2 . Therefore the given PDE is hyperbolic. The characteristic curves are given by the
equation √
dy B ± B 2 − AC
= = ±c.
dx A
Solving we get y ± c = a constant . Thus w (x , y ) = y + cx and z(x , y ) = y − cx . Now writing

uxx = uww wx2 + 2uwz wx zx + uzz zx2 + uw wxx + uz zxx

= c 2 (uww − 2uwz + uzz )
uyy = uww wy2 + 2uwz wy zy + uz zzy2 + uw wyy + uz zyy (7)
= uww + 2uwz + uzz
−c uyy = −c 2 (uww + 2uwz + uzz )
2

Substituting into the given PDE, we get 4c 2 uwz = 0 ⇒ uwz = 0

61 / 106
 continue..
2
In the parabolic case, B − AC = 0, we have a single characteristic. We are looking for a
coordinate system such that either b = c = 0.
Example 0.0.49
Reduce the PDE e 2x uxx + 2e x +y uxy + e 2y uyy = xy1 (y 3 ux + x 3 uy ) to its canonical form. Note
that A = e 2x , B = e x +y , C = e 2y and B 2 − AC = 0. The PDE is parabolic. The
ey
characteristic curves are given by the equation dy B
dx = A = e x . Hence, we have
e −y − e −x = constant . Thus, w (x , y ) = e −y − e −x . Now, we shall choose z s.t.

wx wy
6 0
=
zx zy

For example, z(x , y ) = x is one such choice.

62 / 106
 continue..
Then
ux = e −x uw + uz
uy = −e −y uw
uxx = e −2x uww + 2e −x uwz + uzz − e −x uw (8)
uyy = e −2y uww + e −y uw
uxy = −e −y (e −x uww − uwz )

Substituting into the given PDE, we get e x e −y uzz = (e −y − e −x )uw .

w
Replacing x , y in terms of w , z gives uzz = 1+we z uw . In the elliptic case,
2
B − AC < 0, we have no real characteristics. Thus, we choose w , z to be the
real and imaginary part of the solution of the characteristic equation.
63 / 106
 continue..
Example 0.0.50
Reduce the PDE x 2 uxx + y 2 uyy = 0 given in the region {(x , y ) ∈ R2 : x > 0, y > 0} to its canonical
form. Here A = x 2 , B = 0, C = y 2 . So, B 2 − AC = −x 2 y 2 < 0. Therefore the given PDE is elliptic.
The characteristic equation dy iy
dx = ± x . Solving we get ln x ± ln y = c. Let w (x , y ) = ln x and
z(x , y ) = ln y . Then,
uw
ux =
x
uz
uy =
y
uw uww (9)
uxx = − 2 + 2
x x
uz
uyy = − 2 + uzz /y 2
y
Substituting into the PDE, we get uww + uzz = uw + uz .

64 / 106
 HOME TASK
(i) Verify whether the Cauchy initial value problem,

ut + xux = x , 0 ≤ x ≤ 1 and t > 0

(10)
u(x , 0) = 2x

has unique solution, many solution , and a solution which remains bounded as t → ∞.
∂z ∂z
(ii) Find the general solution of the PDE x ∂x + y ∂y =0
(iii) Find the general solution of the surfaces which are perpendicular to the family of
surfaces u 2 = kxy , k ∈ R.
(iv) Consider the PDE z = xzx + yzy + zx zy . Find the complete integral. Also verify whether
the particular solution passing through x = 0, z = y 2 is either ( x4 − y )2 or ( x4 + y )2 .
(v) Classify the given PDEs (i): uxx + 2uxy + (1 − sgn (y ))uyy = 0
(ii): yuxx + xuyy = 0.

65 / 106
 Wave Equation

The Vibrating String: Deriva: Let us consider a homogeneous string of

length L, stretched along the x-axis, with one end fixed at x = 0 and the other
end fixed at x = L. We assume that the string is free to move only in the
vertical direction. Let ρ > 0 denote the density of the string and T > 0
denote the coefficient of tension of the string. Let u(x, t) denote the vertical
displacement of the string at the point x and time t. By derivation we will get
∂2u 2 ∂2u
∂2t = c ∂2x .
The above equation is called the one dimensional wave equation.

66 / 106
 continue..
Consider the wave equation utt (x , t) = c 2 uxx (x , t) in (x , t) ∈ R × (0, ∞), describing the
vibration of an infinite string. We have already solved this equation in previous section . The
general solution is u(x , t) = F (x + ct) + G(x − ct), where F and G are arbitrary functions.
Theorem 0.0.51
Given g ∈ C 2 (R) and h ∈ C 1 (R), there is a unique C 2 solution u of the Cauchy initial value
problem (IVP) of the wave equation,

utt (x , t) − c 2 uxx (x , t) = 0 in R × (0, ∞) ,

u(x , 0) = g(x ), in R (11 )
ut (x , 0) = h(x ), in R
R x +ct
which is given by d’Alembert’s formula u(x , t) = 12 (g(x + ct) + g(x − ct)) + 2c1 x −ct h(y )dy .

67 / 106
 continue..
Proof.
The general solution is u(x , t) = F (x + ct) + G(x − ct) with F , G ∈ C 2 (R). Using the
initial position we get F (x ) + G(x ) = g(x ). Thus, g ∈ C 2 (R). Now,
ut (x , t) = c(F 0 (x + ct) − G 0 (x − ct)) and putting t = 0, we get
F 0 (x ) − G 0 (x ) = c1 h(x ). Therefore, h ∈ C 1 (R). Now solving for F 0 and G 0 , we get
2F 0 (x ) = g 0 (x ) + h(x ) 0 0 h(x )
c . Similarly, 2G (xR) = g (x ) − c . Integrating both these
1 1 x
equations we have F (x ) = 2 (g(x ) + c 0 h(y )dy ) + c1 , and
G(x ) = 12 (g(x ) − c1 0x h(y )dy ) + c2 . As F (x ) + G(x ) = g(x ), we get c1 + c2 = 0.
R

Therefore, u(x , t) = 21 (g(x + ct) + g(x − ct)) + 2c1 xx−ct

R +ct
h(y )dy .

68 / 106
 Domain of Dependence

We observe that the solution u(x , t) depends only on the interval

[x − ct, x + ct] because g takes values only on the end-points of this interval
and h takes values between this interval. The interval [x − ct, x + ct] is called
the domain of dependence. Thus, the region of R × (0, ∞) on which the value
of u(x , t) depends forms a triangle with base [x − ct, x + ct] and vertex at
(x , t). The domain of dependence of (x , t) is marked in x-axis by the
characteristic curves passing through (x , t).

69 / 106
 Domain of Dependence

Figure: Wave equation in one dimension: Domain of dependence for solution at (x , t)

70 / 106
 continue..
Theorem 0.0.52 (Inhomogeneous wave equation)
Given g ∈ C 2 (R) , h ∈ C 1 (R) and f ∈ C 1 (R × [0, ∞)), there is a unique C 2
solution u of the inho- mogeneous Cauchy initial value problem (IVP) of the
wave equation,

utt (x , t) − c 2 uxx (x , t) = f (x , t) in R × (0, ∞)

u(x , 0) = g(x ) in R (12 )
ut (x , 0) = h(x ) in R

given by the formula u(x , t) =

1 1 R x +ct 0 0 R t R x +c(t−t 0 ) 0 0 0 0
2 [g(x + ct) + g(x − ct)] + 2c [ x −ct h(x )dx + 0 x −c(t−t 0 ) f (x , t )dx dt ].

71 / 106
 continue..

Proof.
Fix (x , t) ∈ R × (0, ∞). Consider the open triangle in R × (0, ∞) with
vertices (x , t), (x − ct, 0), (x + ct, 0)[see the diagram 70], and denote it by
T (x , t). Thus T (x , t) = {(x 0 , t 0 ) ∈ R × (0, ∞) : |x 0 − x | < c(t − t 0 )}. The
boundary of the triangle ∂T (x , t) consists of three parts

T0 = {(x 0 , 0) : x − ct < x 0 < x + ct}

T+ = {(x 0 , t 0 ) ∈ R × (0, t) : ct 0 + x 0 = x + ct} (13)
T− = {(x 0 , t 0 ) ∈ R × (0, t) : ct 0 − x 0 = −x + ct}

72 / 106
 proof continued

The unit outward normal at each point of the boundary ∂T (x , t) is given by

υ = (υ1 , υ2 ) defined by



(0, −1), if (x 0 , t 0 ) ∈ T0 ;

υ(x 0 , t 0 ) =  √1+c
1
2 (1, c), if (x 0 , t 0 ) ∈ T+ ;
 √ 1 (−1, c), if (x 0 , t 0 ) ∈ T ;


1+c 2 −

Gauss Divergence Theorem: Let A be an open bounded subset of Rn with C 1

boundary. Given a vector field V = (v1 , v2 , · · · , vn ) on A s.t. vi ∈ C 1 (Ā) for
1 ≤ i ≤ n, then A ∇ · Vdx = ∂A V · υdσ.
R R

73 / 106
 proof continued
Therefore, by Gauss divergence theorem
Z Z
0 0 0 0
f (x , t )dx dt = [utt (x 0 , t 0 ) − c 2 uxx (x 0 , t 0 )]dx 0 dt 0
T (x ,t) T (x ,t)
Z
= [ut υ2 − c 2 ux υ1 ]dσ
∂T (x ,t)
Z Z
= [ut υ2 − c 2 ux υ1 ]dσ + [ut υ2 − c 2 ux υ1 ]dσ+
T0 T+
Z
[ut υ2 − c 2 ux υ1 ]dσ
T−
Z x +ct
0 0 c Z
=− ut (x , 0)dx + √ [ut − cux ]dσ+
x −ct 1 + c 2 T+
c Z
√ [ut + cux ]dσ
1 + c 2 T−
(14)
74 / 106
 proof continued

Note that the second and third integral are just the direction derivatives of u
along the direction (−c, 1) and (c, 1) in the line T+ and T− respectively.
Therefore, we have T (x ,t) f (x 0 , t 0 )dx 0 dt 0 =
R

− xx−ct
R +ct
h(x 0 )dx 0 + cu(x , t) − cu(x + ct, 0) + cu(x , t) − cu(x − ct, 0).
Therefore, u(x , t) =
1 1 R x +ct 0 0 R t R x +c(t−t 0 ) 0 0 0 0
2 [g(x + ct) + g(x − ct)] + [
2c x −ct h(x )dx + 0 x −c(t−t 0 ) f (x , t )dx dt ]

75 / 106
 Separation of variables
Recall the set-up of the vibrating string given by the equation utt = uxx , we
have normalized the constant c. Initially at time t, let us say the string has the
shape of the graph of v, i.e.,u(x , 0) = v (x ).The snapshot of the vibrating
string at each time are called the”standing waves”.The shape of the string at
time t0 can be thought of as some factor (depending on time) of v.This
observation motivates the idea of”separation of variable”,
i.e.,u(x , t) = v (x )w (t), where w (t) is the factor depending on time, which
scales v at time t to fit the shape of u(x,t).The fact that endpoints are fixed is
given by the boundary condition u(0, t) = u(L, t) = 0. We are also given the
initial condition u(x , 0) = g(x ), initial velocity of the string at time
t = 0,ut (x , 0) = h(x ).

76 / 106
 Separation of variables

We solve the following boundary value problem using the method of

separation of variables.

utt (x , t) = c 2 uxx (x , t), (x , t) ∈ (0, L) × (0, ∞)

u(0, t) = 0, ∀t > 0
u(L, t) = 0, ∀t > 0 (15)
u(x , 0) = g(x )
ut (x , 0) = h(x )

77 / 106
 continue..
Let us seek for solutions u(x, t) whose variables can be separated. Let u(x , t) = v (x )w (t).
Differentiating and substituting in the wave equation, we get v (x )w 00 (t) = c 2 v 00 (x )w (t).
00
v 00 (x )
Hence cw2 w(t)
(t) = v (x ) . Since RHS is a function of x and LHS is a function t, they must equal
00 00
a constant, say λ.Thus cw2 w(t) v (x )
(t) = v (x ) = λ. Using the boundary condition
u(0, t) = u(L, t) = 0, we get v (0)w (t) = v (L)w (t) = 0. If w (t) ≡ 0 then u(x , t) ≡ 0 and
this cannot be a solution. Hence w 6= 0. Thus we have to solve the second order diff. equn.
v 00 (x ) = λv (x ), x ∈ (0, L).
v (0) = v (L) = 0,
Note that the λ can be either zero, positive or negative. If λ = 0, then v 00 (x ) = 0. So the
general solution is v (x ) = αx + β, where α and β are any two constants. Since v (0) = 0,
we get β = 0 and v (L) = 0 implies α = 0. Thus, v = 0, hence u = 0. In this case we get a
trivial solution, but our interest lies in getting non-trivial solution.

78 / 106
 continue..

√ √
If λ > 0, then v√(x ) = αe λx + √βe − λx . Equivalently,
v (x ) = c1 cosh( λx ) + c2 sinh( λx ) such that α = c1 +c
2
2
and β = c1 −c
2 .
2

Using the boundary

√ condition v (0) = 0, we get c1 = 0 and hence
v (x ) = c2 sinh( λx ). Using v (L) = 0 we get c2 = 0. Thus we have v (x ) = 0.
Hence again we have trivial solution u = 0.

79 / 106
 continue..
√
Finally, if λ < 0, then set ω = −λ. We need to solve the simple harmonic
equation v 00 (x ) + ω 2 v (x ) = 0, x ∈ (0, L),
v (0) = v (L) = 0. The general solution is v (x ) = α cos(ωx ) + β sin(ωx ).
Using v (0) = 0 we get α = 0. Hence v (x ) = β sin(ωx ). Now using v (L) = 0
we have β sin(ωL) = 0. Thus either β = 0 or sin(ωL) = 0. But β = 0 yields a
trivial solution. For non trivial solution consider β 6= 0. Then we have
ωL = kπ ⇒ ω = kπ/L for all k ∈ Z − {0}. Since ω > 0, we consider k ∈ N.
Hence, for each k ∈ N there is a solution (vk , λk ) for the eigen value problem
with vk (x ) = βk sin( kπx kπ 2
L ).for some constant βk and λk = −( L ) .

80 / 106
 continue..
It now remains to solve w for each of these λk . For each k ∈ N, we solve for wk in the ode
wk00 (t) + (ckπ/L)2 wk (t) = 0. The general solution is
wk (t) = ak cos(ckπt/L) + bk sin(ckπt/L). So for each k ∈ N, we have
uk (x , t) = [ak cos(ckπt/L) + bk sin(ckπt/L)] sin(kπx /L) for some constants ak and bk . By
principle of superposition we get the general solution of the given problem
∞
X
u(x , t) = [ak cos(ckπt/L) + bk sin(ckπt/L)] sin(kπx /L). Using the initial position of the
k=1
∞
ak sin(kπx /L). By multiplying sin( mπx
X
string g(x ) = u(x , 0) = L ) on both sides of the
k=1
expression of g and integrating from 0 to L, where m is any fixed integer belongs to
{1, 2, · · · }, we get 0L g(x ) sin( mπx )dx = 0L [ ∞ mπx
R R P
L k=1 ak sin(kπx /L)] sin( L )dx =
P∞ RL 2 RL
k=1 ak 0 sin(kπx /L) sin(mπx /L)dx . Therefore, ak = L 0 g(x ) sin(kπx /L)dx .

81 / 106
 continue..

Finally, by differentiating u w.r.t t, we get

ut (x , t) = ∞ k=1 ckπ/L[bk cos ckπt/L − ak P sin ckπt/L] sin(kπx /L). As
P

ut (x , 0) = h(x ), we get h(x ) = ut (x , 0) = ∞k=1 bk kcπ/L sin(kπx /L). Hence,

2 RL
bk = kcπ 0 h(x ) sin(kπx /L)dx .

82 / 106
 Heat Equation
Let L be the length of a homogeneous rod insulated along sides and its ends are kept at zero
temperature. Then the temperature u(x , t) at every point of the rod 0 ≤ x ≤ L and time
2 ∂2u
t ≥ 0 is given by the equation ∂u ∂t = c ∂ 2 x , where c is a constant.The temperature zero at
the end points is given by the Dirichlet boundary condition u(0, t) = u(L, t) = 0. Also, given
that the initial temperature of the rod at time t = 0, is u(x , 0) = g(x ), where g is given (or
known) such thatg(0) = g(L) = 0. We look for all the solutions of the Dirichlet problem
ut (x , t) − c 2 uxx (x , t) = 0 in (0, L) × (0, ∞).
u(0, t) = u(L, t) = 0,
u(x , 0) = g(x ) on [0, L]. We look for u(x , t) = v (x )w (t) (variable separated). We need to
solve two ODE to get v and w,
w 0 (t) = λc 2 w (t),and v 00 (x ) = λv (x ).

83 / 106
 Continued

For each k ∈ N, we have the pair (λk , vk ) as solutions to the eigen value
problem involving v , where λk = −(kπ/L)2 and vk (x ) = sin(kπx /L) for some
constant bk . For each k ∈ N, we solve for wk to get ln wk (t) = λk c 2 t + ln α
2
where α is integrating constant. Thus wk (t) = αe −(kcπ/L) t . Hence
2
uk (x , t) = vk (x )wk (t) = βk sin(kπx /L)e −(kcπ/L) t , for some constant βk .

84 / 106
 Continued

By superposition principle, the general solution is

∞ ∞ 2
βk sin(kπx /L)e −(kcπ/L) t . Using u(x , 0) = g(x ),
X X
u(x , t) = uk (x , t) =
k=1 k=1
we get g(x ) = u(x , 0) = ∞k=1 sin(kπx /L). AsR g(0) = g(L) = 0, hence g
P

admits a fourier sine expansion. So, βk = 2/L 0L g(x ) sin(kπx /L)dx .

85 / 106
 Duhamel’s principle
The Duhamel’s principle states that one can obtain a solution of the inhomogeneous
IVP for heat from its homogeneous IVP.
Consider a linear inhomogeneous Heat-equation for a function u : Ω × (0, ∞) → R,
where Ω ⊂ Rn , of the form
ut (x , t) − Lu(x , t) = f (x , t) (x , t) ∈ Ω × (0, ∞)
u|∂Ω = 0 (16)
u(x , 0) = 0 x ∈ Ω

where L is a linear differential operator that involves no time derivatives. Duhamel’s

principle state that solution of this kind of problem is u(x , t) = 0t (w s f )(x , t)ds, where
R

w s f is the solution of the problem

ut − Lu = 0 (x , t) ∈ Ω × (s, ∞)
u|∂Ω = 0 (17)
u(x , s) = f (x , s) x ∈ Ω

86 / 106
 Continue
For a given f, let u(x,t) be the solution of the inhomogeneous heat equa- tion,

ut (x , t) − c 2 uxx (x , t) = f (x , t), (x , t) ∈ Ω × (0, ∞)

u(x , t) = 0, in ∂Ω × (0, ∞) (18)
u(x , 0) = 0, in Ω

for each s ∈ (0, ∞), consider w (x , t; s) as the solution of the homogeneous problem
(auxiliary)

wts (x , t) − wxx
s
(x , t) = 0 (x , t) ∈ Ω × (s, ∞)
s
w (x , t) = 0, in ∂Ω × (0, ∞)
(19)

w s (x , s) = f (x , s) x ∈ Ω

87 / 106
 Continue

Since t ∈ (s, ∞), introducing a change of variable r = t − s, we have

w s (x , t) = w (x , t − s) which solves

wt (x , r ) − c 2 wxx (x , r ) = 0, (x , r ) ∈ Ω × (0, ∞)
w (x , r ) = 0, in ∂Ω × (0, ∞) (20)
w (x , 0) = f (x , s) on Ω

Duhamel’s principle states that

u(x , t) = 0t w s (x , t)ds = 0t w (x , t − s)ds.
R R

88 / 106
 Finite Bar
Solve the initial boundary value problem for a nonhomogeneous heat equation,
with homogeneous boundary conditions and zero initial data:
ut (x , t) − c 2 uxx (x , t) = P(x , t), (x , t) ∈ (0, L) × (0, ∞)
u(0, t) = T0 (t)
(21)
u(L, t) = T1 (t), t > 0
u(x , 0) = f (x ), in [0, L]
Reduce the Boundary Conditions to Homogeneous:
Choose an arbitrary function v (x , t) satisfying the nonhomogeneous boundary
conditions: v (0, t) = T0 (t), v (L, t) = T1 (t), t > 0. Let
v (x , t) = T0 (t) + xL [T1 (t) − T0 (t)], 0 ≤ x ≤ L, t > 0.
Consider w (x , t) = u(x , t) − v (x , t), as the new unknown. The problem for
w (x , t) becomes
89 / 106
 Finite Bar

wt − c 2 wxx = Q(x , t)
w (0, t) = 0
(22)
w (L, t) = 0
w (x , 0) = g(x )

where Q(x , t) and g(x ) are:

Q(x , t) = P(x , t) − vt − c 2 vxx

x (23)
= P(x , t) − T00 (t) − [T10 (t) − T00 (t)]
L
g(x ) = f (x ) − v (x , 0) = f (x ) − T0 (0) − xL [T1 (0) − T0 (0)].
90 / 106
 Finite Bar
Break the above equation 22 into two simpler parts:

wt − c 2 wxx = 0
w (0, t) = 0
(24)
w (L, t) = 0
w (x , 0) = g(x )

wt − c 2 wxx = Q(x , t)
w (0, t) = 0
(25)
w (L, t) = 0
w (x , 0) = 0

91 / 106
 Finite Bar

Let w1 and w2 be the solutions of the problems 24 and 25, respectively. 1st
one can be obtained by seperation of variable technique and for the 2nd one
we can apply Duhamel’s Principle.
The original unknown u(x , t) = v (x , t) + w1 (x , t) + w2 (x , t).

92 / 106
 Continue

Problem 0.0.53
Consider non-homogeneous heat equation

ut (x , t) = 9uxx + sin(3πx )
u(0, t) = 0
(26 )
u(1, t) = 0
u(x , 0) = sin(πx )

Then find u(x , t) . Also, find lim u(1/3, t).

t→∞

93 / 106
 Laplace equation
In cartesian coordiantes, the n-dimensional Laplacian is given as
∂2
∇ = ni=1 ∂x 2.
P
i

Theorem 0.0.54
1 ∂ ∂ 1 ∂2
The two dimensional Laplacian has the representation ∇ = r ∂r (r ∂r ) + r 2 ∂θ in
polar coordinates.
Proof.
∂u ∂u
Let x = r cos θ,y = r sin θ. Then ∂u/∂r = cos(θ) ∂x + sin(θ) ∂y .
∂2u 2 ∂2u 2 ∂2u ∂2u
∂r 2 = cos (θ) ∂x 2 + sin (θ) ∂y 2 + 2 sin(θ) cos(θ) ∂x ∂y . Similarly calculate,
∂u ∂u ∂u 1 ∂2u ∂2u 1 ∂2u
∂θ = −r sin(θ) ∂x + r cos(θ) ∂y . r 2 ∂θ2 = find it . Finally, ∂r 2 + r 2 ∂θ2 gives the
result.

94 / 106
 continue..

Theorem 0.0.55 (2D rectangle .Dirichlet problem)

Let Ω = {(x , y ) ∈ R2 |0 < x < a, 0 < y < b} be a rectangle in R2 .Let
g : ∂Ω → R which vanishes on three sides of the rectangle, i.e.,
g(0, y ) = g(x , 0) = g(a, y ) = 0 and g(x , b) = h(x ), where h is a continuous
function s.t. h(0) = h(a) = 0. Then there is a unique solution to
∇u(x , y ) = 0, (x , y ) ∈ Ω,
u(x , y ) = g(x , y ),(x , y ) ∈ ∂Ω on this rectangle with given boundary value g.

95 / 106
 continue..
Proof.
We begin by looking for solution u(x , y ) whose variables are separated. Let
u(x , y ) = v (x )w (y ). Then v 00 (x )w (y ) + v (x )w 00 (y ) = 0. Hence,
v 00 (x ) w 00 (y )
v (x ) = − w (y ) . Since LHS is function of x and RHS is function of y, they
00 00
must equal a constant, say λ. Thus, vv (x(x)) = − ww (y(y)) = λ. Using boundary
condition u(0, y ) = u(a, y ) = 0 we get v (0)w (y ) = v (a)w (y ) = 0. If
w (y ) = 0 then we get zero solution. For nontriviality, consider w 6= 0. So,
v (0) = v (a) = 0. Thus we need to solve the eigen value problem for 2nd order
ODE
v 00 (x ) = λv (x ), x ∈ (0, a)
v (0) = 0 (27)
v (a) = 0
96 / 106
 continue..

If λ = 0, then v 00 (x ) = 0. The general solution is v (x ) = αx + β. Hence in

this we get trivial solution. √ √
If λ > 0, then v (x ) = c1 cosh( λx ) + c2 sinh( λx ). Using boundary
conditioon, we get v (x )√= 0. In this case we also get trivial solutiuon.
If λ < 0, then set ω = −λ. We need to solve
v 00 (x ) + ω 2 v (x ) = 0, x ∈ (0, a)
v (0) = 0 (28)
v (a) = 0

97 / 106
 continue..

The general solution is v (x ) = α cos(ωx ) + β sin(ωx ). Using boundary

condition v (0) = 0 we get α = 0. Hence, v (x ) = β sin(ωx ). Using v (a) = 0
we get β sin(ωa) = 0. Thus, either β = 0 or sin(ωa) = 0. But β 6= 0
otherwise again we get trivial solutiuon. Hence, sin(ωa) = 0. So, ω = kπ/a,
∀k ∈ Z − {0}. Since ω > 0, so k ∈ N. Hence, for each k ∈ N there is a
solution (vk , λk ) with vk (x ) = βk sin( kπx
a ), for some constant βk and
2
λk = − kπ
a .

98 / 106
 continue..
We now solve w, corresponding to each λk . For each k ∈ N,we solve for wk in
the ODE ,
wk00 (y ) = ( kπ 2
a ) wk (y ), y ∈ (0, b).
w (0) = 0.
Using boundary condition we have wk (y ) = ck sinh(kπy /a). Therefore, for
each k ∈ N, uk (x , y ) = δk sin(kπx /a) sinh(kπy /a) is a solution of the given
problem. Using superposition principle, we get the general solution
∞
X
u(x , y ) = uk (x , y ). The constant δk are obtained by using the boundary
k=1
∞
X
condition u(x , b) = h(x ). h(x ) = δk sin(kπx /a) sinh(kπb/a). Since,
k=1
h(0) = h(a) = 0, h admits a Fourier sine sereis. So, Fourier coefficient
δk = (sinh(kπb/a))−1 a2 0a h(x ) sin(kπx /a)dx .
R

99 / 106
 continue..
Theorem 0.0.56 (Interior Dirichlet problem for disk)
Let Ω = {(x , y ) ∈ R2 |x 2 + y 2 < R 2 }be the disk of radius R in R2 . Let
g : ∂Ω → R is a continuous function. Then there is a unique solution
∇u(x ) = 0, x ∈ Ω,
u(x ) = g(x ),x ∈ ∂Ω on the unit disk with given boundary value g. Then,
solving for u(x , y ) in the Dirichlet problem is to equivalent to finding
∂2
U(r , θ) : Ω → R such that ∇ = 1r ∂∂ (r ∂∂ ) + r12 ∂θ in Ω.
U(r , θ + 2π) = U(r , θ) in Ω,
U(R, θ) = G(θ) on ∂Ω.
where r is the magnitude component and θ is the direction component, then
∂Ω is the circle of radius of one. Here U(r , θ) = u(r cos θ, r sin θ) and
G : [0, 2π) → R is given as G(θ) = g(cos θ, sin θ)
100 / 106
 continue..

Here both U and G are 2π periodic w.r.t θ.

Solution 0.0.57
To find the unknown U(r , θ), we use the separation of variable technique. Let
U(r , θ) = v (r )w (θ). Here both v and w are non-zero. Then proceed. Finally,
2
−r 2 R π G(η)
U(r , θ) = R 2π −π R 2 +r 2 −2rR cos(η−θ) dη.

101 / 106
 continue..
1. Let PQRS be a rectangle in the 1st quadrant whose adjacent sides PQ and QR
have slopes 1 and -1, respectively. Let u(x , t) is a solution of utt − uxx = 0 and
u(P) = 1, u(Q) = −1/2, u(R) = 1/2. Then find u(S).
2. Consider IVP
ut − uxx = 0
e 2x − 1 (29)
u(x , 0) =
e 2x + 1
Find u(x , t). Also, find limt→∞ u(1, t).
3. Let u(x , y ) be a solution of Laplace equation on Ω = {(x , y ) ∈ R2 : x 2 + y 2 ≤ 1}.
If 
sin θ, if 0 ≤ θ ≤ π;
u(cos θ, sin θ) = 
0, for π ≤ θ ≤ 2π
Find u(r , θ). Also find u(0, 0).
102 / 106
 continue..
Theorem 0.0.58 (exterior dirichlet problem for circle)
The problem is described by ∇2 u = 0,
u(R, θ) = G(θ),
u must be bounded as r → ∞.
Example: Solve one dimenssional diffusion equation in the doamin 0 < x < π,t ≥ 0,subject
to the condition (a) u remains finite as t → ∞.
(b) u = 0 ,if x = 0 and x = π for all t,
at t = 0 u = x ,0 ≤ x ≤ π2 ,
π − x , π2 ≤ x ≤ π.
2
u(x , t) = (c1 e λx + c2 e −λx )e αλ t ,
2
u(x , t) = (c1 cos λx + c2 sin λx )e −αλ t ,
u(x , t) = (c1 x + c2 ).

103 / 106
 Maximum Principle

Theorem 0.0.59 (Strong Maximum Principle)

Let Ω be an open ,connected subset of Rn .Let u be harmonic in Ω and
M = maxy ∈Ω̄ u(y ).Then u(x ) < M, for all x ∈ Ω.or u = M is contant in Ω.
Corollary 0.0.60 (Weak maximum Principle)
Let Ω be an open bounded subset of Rn . Let u ∈ C (Ω̄) be harmonic in Ω .
Then maxy ∈Ω̄ u(y ) = maxy ∈∂Ω u(y ).
Read uniqueness of harmonic functions.

104 / 106
 continue..

Theorem 0.0.61
Let Ω be an open bounded connected subset of Rn and g ∈ C (∂Ω). Then the
Dirichlet problem has atmost one solution u ∈ C 2 (Ω) C (Ω̄).Moreover, if u1
T

and u2 are solution to the Dirichlet problem corresponding to g1 and g2 in

C (∂Ω),respectively, then
(a) g1 ≥ g2 on ∂Ω and g1 (x0 ) > g2 (x0 ) for some x ∈ Ω implies that u1 > u2
in Ω.
(b) |u1 (x ) − u2 (x )| ≤ maxy ∈∂Ω |g1 (y ) − g2 (y )| for all x ∈ Ω.

105 / 106
 Internal Assesment
Answer the following questions by making 10 marks.
1. Derive the general integral from a given complete integral u(x , y ; a, b) = xy + ab + (a, b) · (x , y ) of the nonlinear
PDE ux uy = u(x , y ).[4]
2. Classify the PDE: uxx + 2uyy + uzz = 2uxz + 2uyz .[2]
3. Reduce the PDe x 2 uxx + y 2 uyy = 0 in the region{(x , y )|x > 0, y > 0} to its canonical form.[4]
4. Find the complete integral of the pde (ux2 + uy2 )x = ux u and deduce the solution which passes through the
curvex = 0, z = 2y [6]
5. Let Ω = {(x , y )|x > 0} and Γ = {(0, y )|y ∈ R}.Solve the PDE ux uy = u(x , y ) in Ω.
u(0, y ) = y 2 on Γ.[6]

106 / 106