Partial Differential Equations Study Notes
Definition and Notation of Partial Differential Equations
Formation of Partial Differential Equations
By Elimination of Arbitrary Constants: Differentiating an equation f(x,y,z,a,b)=0 partially with respect to x and y and eliminating the constants a and b yields a first-order PDE of the form f(x,y,z,p,q)=0
By Elimination of Arbitrary Functions: Eliminating arbitrary functions f from an expression f(u,v)=0 (where u and v are functions of x,y,z) yields a first-order linear PDE Pp+Qq=R
Types of Integrals / Solutions of PDEs
Complete Integral: A solution containing as many arbitrary constants as independent variables (ϕ(x,y,z,a,b)=0)
Particular Integral: A solution obtained by assigning specific numerical values to arbitrary constants in the complete integral
Singular Integral: Represents the envelope of the family of surfaces given by the complete integral, obtained by eliminating a and b from ϕ=0, ∂a∂ϕ=0, and ∂b∂ϕ=0
General Integral: A solution of the form f(u,v)=0 containing arbitrary functions
Lagrange's Solution of First-Order Linear PDEs
Form: Pp+Qq=R, where P,Q,R are functions of x,y,z
Lagrange's Auxiliary Equations: Pdx=Qdy=Rdz
Method: Solve the auxiliary equations to find two independent integrals u(x,y,z)=c1 and v(x,y,z)=c2. The general solution is expressed as f(u,v)=0
Standard Forms for Non-Linear First-Order PDEs
Standard Form I: f(p,q)=0
Complete Integral: z=ax+by+c, where b=ϕ(a) from f(a,b)=0
Standard Form II: f(p,q,z)=0
Method: Substitute X=x+ay, which converts the PDE into an ordinary differential equation (ODE) f(dXdz,adXdz,z)=0
Standard Form III: f(p,x)=g(q,y)
Method: Equate both sides to an arbitrary constant a (f(p,x)=a and g(q,y)=a), solve for p and q, and integrate dz=pdx+qdy
Standard Form IV (Clairaut's Form): z=px+qy+f(p,q)
Charpit's Method for General First-Order Non-Linear PDEs
Charpit's Auxiliary Equations: ∂x∂f+p∂z∂fdp=∂y∂f+q∂z∂fdq=−p∂p∂f−q∂q∂fdz=−∂p∂fdx=−∂q∂fdy
Method: Solve for p and q using the auxiliary equations, then integrate dz=pdx+qdy
Homogeneous Linear PDEs with Constant Coefficients
General Form: F(D,D′)z=f(x,y), where D=∂x∂ and D′=∂y∂
Auxiliary Equation (A.E.): Formed by substituting D=m and D′=1 into F(D,D′)=0
Complementary Function (C.F.):
Distinct real roots α1,α2,…,αn: C.F.=f1(y+α1x)+f2(y+α2x)+⋯+fn(y+αnx)
Repeated root α (twice): C.F.=f1(y+αx)+xf2(y+αx)
Classification of Second-Order Linear PDEs
General Equation: Rr+Ss+Tt+f(x,y,z,p,q)=0
Classification Criterion (based on discriminant S2−4RT):
Hyperbolic: S2−4RT>0
Parabolic: S2−4RT=0
Elliptic: S2−4RT<0
Classical Second-Order Partial Differential Equations
Wave Equation (1D): ∂t2∂2u=c2∂x2∂2u
D'Alembert's Solution: u(x,t)=21[ϕ(x−ct)+ϕ(x+ct)]+2c1∫x−ctx+ctψ(s)ds
Heat Equation (1D): ∂t∂u=c2∂x2∂2u
Laplace Equation (2D): ∇2u=∂x2∂2u+∂y2∂2u=0
Sturm-Liouville Problems and Green's Function
Sturm-Liouville Equation: dxd(p(x)dxdy)+q(x)y=λy
Properties: Eigenvalues λn are real, ordered, and countable; corresponding eigenfunctions yn(x) are orthogonal.
Green's Function: Used to construct integral solutions for non-homogeneous boundary value problems Ly=f(x) in the form y(x)=∫abG(x,t)f(t)dt
Definition and Notation of Partial Differential Equations
Definition: A Partial Differential Equation (PDE) is a differential equation involving partial derivatives of one or more dependent variables with respect to two or more independent variables.
Standard Notations (Monge's Notation):
p=∂x∂z
q=∂y∂z
r=∂x2∂2z
s=∂x∂y∂2z
t=∂y2∂2z
Formation of Partial Differential Equations
Types of Integrals / Solutions of PDEs
Complete Integral: A solution ϕ(x,y,z,a,b)=0 containing as many arbitrary constants (a,b) as independent variables (x,y).
Particular Integral: A solution obtained by assigning specific numerical values to the arbitrary constants in the complete integral.
Singular Integral: Represents the envelope of the family of surfaces given by the complete integral. It is obtained by eliminating a and b from:
ϕ(x,y,z,a,b)=0
∂a∂ϕ=0
∂b∂ϕ=0
General Integral: Obtained by setting b=ϕ(a) in the complete integral to get ϕ(x,y,z,a,ϕ(a))=0, differentiating with respect to a, and eliminating a. It corresponds to the form f(u,v)=0.
Lagrange's Solution of First-Order Linear PDEs
Form: Pp+Qq=R, where P,Q,R are functions of x,y,z.
Lagrange's Auxiliary Equations:
Pdx=Qdy=Rdz
Method of Solution:
Grouping Method: Pair any two ratios to integrate directly and find u(x,y,z)=c1.
Method of Multipliers: Find multipliers (l,m,n) such that lP+mQ+nR=0, implying ldx+mdy+ndz=0, which integrates to v(x,y,z)=c2.
General Solution: Expressed as f(u,v)=0 or u=ϕ(v).
Standard Forms for Non-Linear First-Order PDEs
Standard Form I: f(p,q)=0
Complete Integral: z=ax+by+c, where b is determined from f(a,b)=0.
Standard Form II: f(p,q,z)=0
Method: Substitute X=x+ay, transforming the PDE into an ordinary differential equation (ODE):
f(dXdz,adXdz,z)=0
Standard Form III: f(p,x)=g(q,y)
Method: Equate both sides to an arbitrary constant a (f(p,x)=a⇒p=h1(x,a) and g(q,y)=a⇒q=h2(y,a)), then integrate dz=pdx+qdy to get:
z=∫h1(x,a)dx+∫h2(y,a)dy+b
Standard Form IV (Clairaut's Form): z=px+qy+f(p,q)
Charpit's Method for General First-Order Non-Linear PDEs
Applicable to general first-order equations f(x,y,z,p,q)=0.
Charpit's Auxiliary Equations:
∂x∂f+p∂z∂fdp=∂y∂f+q∂z∂fdq=−p∂p∂f−q∂q∂fdz=−∂p∂fdx=−∂q∂fdy
Method: Solve for p and q using auxiliary equations and the original PDE, then integrate dz=pdx+qdy
Homogeneous and Non-Homogeneous Linear PDEs with Constant Coefficients
General Form: F(D,D′)z=f(x,y), where D=∂x∂ and D′=∂y∂.
Auxiliary Equation (A.E.): Formed by substituting D=m and D′=1 into F(D,D′)=0
Complementary Function (C.F.):
Distinct real roots α1,α2,…,αn:
C.F.=f1(y+α1x)+f2(y+α2x)+⋯+fn(y+αnx)
Repeated root α (m times):
C.F.=f1(y+αx)+xf2(y+αx)+⋯+xm−1fm(y+αx)
Particular Integral (P.I.):
For f(x,y)=exp(ax+by): P.I.=F(a,b)1exp(ax+by) provided F(a,b)=0.
For f(x,y)=sin(ax+by) or cos(ax+by): Substitute D2→−a2, DD′→−ab, and D′2→−b2.
Cauchy's Problem and Existence / Uniqueness
Initial Value Problem: Consists of finding a solution surface passing through a given initial space curve x=x0(t), y=y0(t), z=z0(t).
Uniqueness Condition: For Pp+Qq=R, a unique solution exists if:
P(x0,y0,z0)dtdy0−Q(x0,y0,z0)dtdx0=0
No Solution / Infinite Solutions: If Pdtdy0−Qdtdx0=0, then:
Classification and Canonical Forms of Second-Order Linear PDEs
General Form: Rr+Ss+Tt+f(x,y,z,p,q)=0
Classification Criterion (discriminant Δ=S2−4RT):
Hyperbolic: S2−4RT>0 (Two real characteristic curves). Canonical form: ∂u∂v∂2z=ϕ(u,v,z,zu,zv).
Parabolic: S2−4RT=0 (One real characteristic curve). Canonical form: ∂v2∂2z=ϕ(u,v,z,zu,zv).
Elliptic: S2−4RT<0 (Complex conjugate characteristic curves). Canonical form: ∂ξ2∂2z+∂η2∂2z=ϕ(ξ,η,z,zξ,zη).
Classical Second-Order Partial Differential Equations
Wave Equation (1D): ∂t2∂2u=c2∂x2∂2u
D'Alembert's Solution for infinite domain (−∞<x<∞):
u(x,t)=21[ϕ(x−ct)+ϕ(x+ct)]+2c1∫x−ctx+ctψ(s)ds
Fourier Series Solution for finite domain (0≤x≤l) with u(0,t)=u(l,t)=0:
u(x,t)=∑n=1∞sin(lnπx)[ancos(lnπct)+bnsin(lnπct)]
Heat Equation (1D): ∂t∂u=c2∂x2∂2u
Solved via separation of variables: u(x,t)=∑n=1∞cnsin(lnπx)exp(−l2c2n2π2t).
Laplace Equation (2D): ∇2u=∂x2∂2u+∂y2∂2u=0
Functions satisfying Laplace's equation are called harmonic functions.
In polar coordinates: ∂r2∂2u+r1∂r∂u+r21∂θ2∂2u=0.
Sturm-Liouville Problems and Green's Function
Sturm-Liouville Equation: dxd(p(x)dxdy)+q(x)y+λr(x)y=0
Key Properties:
Eigenvalues λn are real, countable, and ordered (λ1<λ2<…).
Eigenfunctions yn(x) corresponding to distinct eigenvalues are orthogonal with weight function r(x).
Green's Function: Used for non-homogeneous BVP L[y]=f(x):
y(x)=∫abG(x,t)f(t)dt
where G(x,t) is continuous at x=t, satisfies boundary conditions, and possesses a jump discontinuity in derivative at x=t:
$$\left.\frac{\partial G}{\