2
Most read
12
Most read
16
Most read
Partial Differential Equations,
Part I
2015.11.19 Enrique Valderrama, Ph.D.
Table of contents
1 Introduction
2 Laplace’s Equation
Steady-State temperature in a rectangular plate
Math. Parenthesis: The Fourier Series
3 The Diffusion or Heat Flow Equation
Flow of Heat through a slab of thickness
4 The Wave Equation
The Vibrating String
Introduction
Many of the problems of mathematical physics involve the solution
of PDE’s. The same PDE may apply to a variety of physical
problems; then many of the methods will apply to a bigger set of
problems.
Laplace’s Equation: 2
u = 0
Poisson’s Equation: 2
u = f (x, y, z)
The difussion or heat flow equation: 2
u =
1
α2
∂u
∂t
Wave equation: 2
u =
1
v2
∂2u
∂t2
Helmoltz equation: 2
F + k2
F = 0
Schröedinger equation: −
2
2m
2
Ψ + V Ψ = i
∂
∂t
Ψ
Laplace’s Equation:
Example (Steady-State temperature in a rectangular plate)
We have a rectangular plate on the region:
R : {0 < x < 10 , 0 < y} with border conditions:
T(x = 0) = T(x = 10) = T(y = ∞) = 0◦ and T(y = 0) = 100◦
Solution
2
T = 0
∂2T
∂x2
+
∂2T
∂y2
= 0
T(x, y) = X(x)Y (y)
Y
∂2X
∂x2
+ X
∂2Y
∂y2
= 0
Solution
Y
∂2X
∂x2
+ X
∂2Y
∂y2
= 0
1
X
∂2X
∂x2
+
1
Y
∂2Y
∂y2
= 0
⇒
1
X
∂2X
∂x2
= −
1
Y
∂2Y
∂y2
= const. = −k2
X = −k2
X and Y = +k2
Y
(soln’s:)
X(x) = A cos kx + B sin kx and Y (y) = Ceky
+ De−ky
Solution
T(x, y) = X(x)Y (y) = (A cos kx + B sin kx)(Ceky
+ De−ky
)
B.C. : T(x, ∞) = 0 ⇒ C = 0
T(0, y) = 0 ⇒ A = 0 and
T(10, y) = 0 ⇒ sin 10k = 0 ⇒ k = n
π
10
)
⇒
T(x, y) = BD sin n
π
10
x e−n π
10
y
andT(x, 0) = 100 ⇒ BD sin n
π
10
x · 1 = 100
Solution
T(x, y) =
∞
n=1
bne−n π
10
y
sin n
π
10
x
For y = 0, we must have T = 100, then :
T(x, 0) =
∞
n=1
bn sin n
π
10
x
which is just the Fourier sine series · · ·
Fourier Series:
Similar to Taylor series, which we approx. a function using the
summation of the derivatives of the function and polynomials, a
Fourier series will approx. a function with the summation of cosine
and sine functions:
f (x) = a0 +
∞
k=1
an cos
nπx
+ bn sin
nπx
where the coefficients are given by the Euler formulas
a0 =
1
2 −
f (x) dx,
an =
1
−
f (x) cos
nπx
dx, n = 1, 2, · · ·
bn =
1
−
f (x) sin
nπx
dx, n = 1, 2, · · ·
Approximation to a square wave
Considere the square wave, which first wave is:
f = 4 if 0 ≤ x ≤ π
f = 0 if π < x < 2π
and so on
Example
Use FS to approximate the square wave function.
Solution
The Euler coefficients:
a0 =
1
2 −
f (x) dx =
1
2π
0
−π
0 dx +
π
0
4 dx = 2
Solution
an =
1
−
f (x) cos
nπx
dx =
1
π
0
−π
0 dx +
π
0
4 cos
nπx
π
dx
=
1
π
π
0
4 cos nx dx =
1
π
4 sin nx
n
π
0
= 0
bn =
1
−
f (x) sin
nπx
dx =
1
π
0
−π
0 dx +
π
0
4 sin
nπx
π
dx
=
1
π
π
0
4 sin nx dx = −
1
π
4 cos nx
n
π
0
=
4
πn
(1 − cos nπ)
⇒
f (x) = 2 +
8
π
sin x +
8
3π
sin 3x +
8
5π
sin 5x + · · ·
Let’s come back to Laplace PDE · · ·
Solution
T(x, 0) = 100 =
∞
n=1
bn sin n
π
10
x
We recognize f (x) = 100, = 10 therefore bn are given by:
bn =
1
10
0
−10
0 · sin
nπx
10
dx +
10
0
100 sin
nπx
10
dx
= 10
10
πn
−
10 cos πn
πn
=
200
nπ
, n = 1, 3, 5 · · ·
Finally:
T(x, y) =
∞
n=1
bne−n π
10
y
sin n
π
10
x
=
200
π
e− π
10
y
sin
π
10
x +
1
3
e−3π
10
y
sin
3π
10
x + · · ·
The Diffusion or Heat Flow Equation:
Example (Flow of Heat through a slab of thickness )
We have a slab place on the region:
R : {0 < x < , −∞ < y < ∞} with border conditions:
u(x = 0) = 0 and u(x = ) = 100◦ for t = 0 (beginning, a
steady-state temperature distribution)
Solution
2
u =
1
α2
∂u
∂t
, (α2
is a const. char. mate. and u the temp.)
if u = F(x, y, z)T(t)
⇒ T 2
F =
F
α2
∂T
∂t
(dividing by u = FT)
1
F
2
F =
1
α2T
∂T
∂t
⇒
1
F
2
F = −k2
and
1
α2T
∂T
∂t
= −k2
Solution
The solution of the DFQ dependent of the time is trivial:
T(t) = e−k2α2t
and in our particular problem, where we have a very long slab, the
diffusion will be only on the x-direction, then the DFQ dependent
on the space coordinates, also become trivial, because we
recognize as the SHO.
∂2F(x)
∂x2
+ k2
F(x) = 0
soln :
F(x) =A cos kx + B sin kx
Solution
B.C.: u(x = 0) = 0 ⇒ A = 0, and we allow u(x = ) = 0 at a later
time(diffusion). Then if u(x = ) = 0 ⇒ sin k = 0 ⇒ k = nπ
⇒ k =
nπ
(eigenvalues)
Then our base functions (eigenfunctions) are then
u = e−k2α2t
sin
nπx
and the general solution to our problem will be the series:
u =
∞
n=1
bne−k2α2t
sin
nπx
Solution
The problem say at the beginning, a steady-state temperature
distribution is on the slab, which implies u0 satisfy Laplace’s
equation, i.e 2u0 = 0, i.e.
∂u2
0
∂x2 = 0, and the solution for this
equation is u0 = ax + b, then applying border condition we get
u0 =
100
x
Then for t = 0, u(x, t = 0) = u0, i.e.
u(x, t = 0) =
∞
n=1
bn sin
nπx
=
100
x
Which can be solved using F.S., for f (x) = 100
x and the
half-period equal to
Solution
The Euler coefficients:
an = 0 (we dont want solutions with cosine functions (B.C.))
bn =
1
−
f (x) sin
nπx
dx =
1 0
−
0 dx +
0
100
x sin
nπx
dx
=
100
2
π
0
x sin
nπx
dx ⇒ (Integration by parts:)
bn =
100(sin πn − πn cos πn)
π2n2
=
100
π
(−1)n−1
n
⇒
u =
100
π
∞
n=1
(−1)n−1
n
e−k2α2t
sin
nπx
The Wave Equation:
Example (The Vibrating String)
Let a string be stretched tightly and its ends fastened to supports
at x = 0 and x = . When the string is vibrating, its vertical
displacement y from its equilibrium position along the x-axis
depends on x and t. We assume the displacement y is very small
and that the slope ∂y/∂t is small at any point at any time.
Solution
∂2y
∂x2
=
1
v2
∂2y
∂t2
, (v = T/µ)
if y = X(x)T(t)
⇒
1
X
∂2X
∂x2
=
1
v2
1
T
∂2T
∂t2
= −k2
(soln’s:)
X(x) = A cos kx + B sin kx and T(t) = C cos kvt + D sin kvt
Solution
B.C.: y(x = 0) = 0 ⇒ A = 0,
u(x = ) = 0 ⇒ sin k = 0 ⇒ k = nπ ⇒ k = nπ
(eigenvalues)
Then: y1(x, t) = B sin
nπ
x C cos
nπ
vt + D sin
nπ
vt but at
t = 0 every piece of string is not varying with time, the ∂y/∂t = 0
Then D = 0 as well.
Then our base functions (eigenfunctions) are then
y1 = BC sin
nπ
x cos
nπ
vt
and the general solution to our problem will be the series:
y(x, t) =
∞
n=1
bn sin
nπ
x cos
nπ
vt
Solution
Then at t = 0, y(x, t) = y0 = f (x), then
y(x, t = 0) =
∞
n=1
bn sin
nπ
x = f (x)
· · ·

More Related Content

PDF
Chapter 16 1
PPTX
Differential equations of first order
PPT
Newton-Raphson Method
PPTX
classification of second order partial differential equation
PPT
1532 fourier series
PPTX
Partial differential equations
PPTX
Fourier integral
PPT
The wave equation
Chapter 16 1
Differential equations of first order
Newton-Raphson Method
classification of second order partial differential equation
1532 fourier series
Partial differential equations
Fourier integral
The wave equation

What's hot (20)

PPTX
Cauchy integral theorem &amp; formula (complex variable & numerical method )
PPT
Integration Ppt
PPTX
Finite difference method
PDF
Inverse Laplace Transform
PDF
Function of several variables
PPTX
Power method
PPTX
Ordinary differential equation
PPTX
Ordinary differential equation
PPTX
Complex analysis
PPTX
Coordinate systems (and transformations) and vector calculus
PPT
Laplace transforms
PPTX
Jacobi method
PPTX
Complex integration
PDF
Finite Difference Method
PPTX
Calculus ppt on "Partial Differentiation"#2
PDF
Lecture-4 Reduction of Quadratic Form.pdf
PPTX
Runge kutta
PPTX
APPLICATIONS OF DIFFERENTIAL EQUATIONS-ZBJ
PPTX
First order linear differential equation
PPTX
A brief introduction to finite difference method
Cauchy integral theorem &amp; formula (complex variable & numerical method )
Integration Ppt
Finite difference method
Inverse Laplace Transform
Function of several variables
Power method
Ordinary differential equation
Ordinary differential equation
Complex analysis
Coordinate systems (and transformations) and vector calculus
Laplace transforms
Jacobi method
Complex integration
Finite Difference Method
Calculus ppt on "Partial Differentiation"#2
Lecture-4 Reduction of Quadratic Form.pdf
Runge kutta
APPLICATIONS OF DIFFERENTIAL EQUATIONS-ZBJ
First order linear differential equation
A brief introduction to finite difference method
Ad

Viewers also liked (20)

PPTX
Propteties of Standard Deviation
PPTX
ARITHMETIC MEAN AND SERIES
PPTX
Spearman Rank
PPTX
Calculation of arithmetic mean
PPT
Malimu variance and standard deviation
DOCX
PPT
First order non-linear partial differential equation & its applications
PPTX
2.3 Histogram/Frequency Polygon/Ogives
PPT
Spearman Rank Correlation Presentation
PPT
Correlation analysis ppt
PPTX
Standard deviation
PPTX
Standard Deviation and Variance
PPT
Standard deviation (3)
PDF
Correlation and Simple Regression
PPTX
Pearson Correlation
PPTX
Mean, median, mode, & range ppt
PPS
Correlation and regression
PPTX
Correlation ppt...
Propteties of Standard Deviation
ARITHMETIC MEAN AND SERIES
Spearman Rank
Calculation of arithmetic mean
Malimu variance and standard deviation
First order non-linear partial differential equation & its applications
2.3 Histogram/Frequency Polygon/Ogives
Spearman Rank Correlation Presentation
Correlation analysis ppt
Standard deviation
Standard Deviation and Variance
Standard deviation (3)
Correlation and Simple Regression
Pearson Correlation
Mean, median, mode, & range ppt
Correlation and regression
Correlation ppt...
Ad

Similar to Partial Differential Equations, 3 simple examples (20)

PPT
M220w07
PDF
Heat problems
PDF
Clarke fourier theory(62s)
PPT
Btech admission in india
PPT
Btech admission in india
PPTX
Fourier series
PPTX
Fourier series
PPTX
forrier serries.pptx
PDF
Notes pde pt3
DOCX
1 d heat equation
PDF
Fourier series
PPTX
Fourier transform
PPT
Conference ppt
PDF
fouriertransform.pdf
PDF
Fourier Series.pdf
PDF
partial defferential_slides_numerical.pdf
PDF
TRANSIENT AND STEADY STATE HEAT CONDUCTION WITH NO LATERAL CONVECTION SOLVED ...
PDF
PPTX
23MA401 Numerical Methods Boundary value problems
PDF
PDE Constrained Optimization and the Lambert W-Function
M220w07
Heat problems
Clarke fourier theory(62s)
Btech admission in india
Btech admission in india
Fourier series
Fourier series
forrier serries.pptx
Notes pde pt3
1 d heat equation
Fourier series
Fourier transform
Conference ppt
fouriertransform.pdf
Fourier Series.pdf
partial defferential_slides_numerical.pdf
TRANSIENT AND STEADY STATE HEAT CONDUCTION WITH NO LATERAL CONVECTION SOLVED ...
23MA401 Numerical Methods Boundary value problems
PDE Constrained Optimization and the Lambert W-Function

Recently uploaded (20)

DOCX
THEORY AND PRACTICE ASSIGNMENT SEMESTER MAY 2025.docx
PPTX
Key-Features-of-the-SHS-Program-v4-Slides (3) PPT2.pptx
PDF
Laparoscopic Imaging Systems at World Laparoscopy Hospital
PPTX
Copy of ARAL Program Primer_071725(1).pptx
PDF
Diabetes Mellitus , types , clinical picture, investigation and managment
PPTX
Approach to a child with acute kidney injury
PPTX
CHROMIUM & Glucose Tolerance Factor.pptx
PPTX
Cite It Right: A Compact Illustration of APA 7th Edition.pptx
PPTX
4. Diagnosis and treatment planning in RPD.pptx
PPTX
Power Point PR B.Inggris 12 Ed. 2019.pptx
PPTX
principlesofmanagementsem1slides-131211060335-phpapp01 (1).ppt
PPTX
Thinking Routines and Learning Engagements.pptx
PDF
Review of Related Literature & Studies.pdf
PDF
Disorder of Endocrine system (1).pdfyyhyyyy
PDF
WHAT NURSES SAY_ COMMUNICATION BEHAVIORS ASSOCIATED WITH THE COMP.pdf
PDF
FYJC - Chemistry textbook - standard 11.
PDF
Chevening Scholarship Application and Interview Preparation Guide
PPTX
Neurology of Systemic disease all systems
PPTX
pharmaceutics-1unit-1-221214121936-550b56aa.pptx
PDF
GSA-Past-Papers-2010-2024-2.pdf CSS examination
THEORY AND PRACTICE ASSIGNMENT SEMESTER MAY 2025.docx
Key-Features-of-the-SHS-Program-v4-Slides (3) PPT2.pptx
Laparoscopic Imaging Systems at World Laparoscopy Hospital
Copy of ARAL Program Primer_071725(1).pptx
Diabetes Mellitus , types , clinical picture, investigation and managment
Approach to a child with acute kidney injury
CHROMIUM & Glucose Tolerance Factor.pptx
Cite It Right: A Compact Illustration of APA 7th Edition.pptx
4. Diagnosis and treatment planning in RPD.pptx
Power Point PR B.Inggris 12 Ed. 2019.pptx
principlesofmanagementsem1slides-131211060335-phpapp01 (1).ppt
Thinking Routines and Learning Engagements.pptx
Review of Related Literature & Studies.pdf
Disorder of Endocrine system (1).pdfyyhyyyy
WHAT NURSES SAY_ COMMUNICATION BEHAVIORS ASSOCIATED WITH THE COMP.pdf
FYJC - Chemistry textbook - standard 11.
Chevening Scholarship Application and Interview Preparation Guide
Neurology of Systemic disease all systems
pharmaceutics-1unit-1-221214121936-550b56aa.pptx
GSA-Past-Papers-2010-2024-2.pdf CSS examination

Partial Differential Equations, 3 simple examples

  • 1. Partial Differential Equations, Part I 2015.11.19 Enrique Valderrama, Ph.D.
  • 2. Table of contents 1 Introduction 2 Laplace’s Equation Steady-State temperature in a rectangular plate Math. Parenthesis: The Fourier Series 3 The Diffusion or Heat Flow Equation Flow of Heat through a slab of thickness 4 The Wave Equation The Vibrating String
  • 3. Introduction Many of the problems of mathematical physics involve the solution of PDE’s. The same PDE may apply to a variety of physical problems; then many of the methods will apply to a bigger set of problems. Laplace’s Equation: 2 u = 0 Poisson’s Equation: 2 u = f (x, y, z) The difussion or heat flow equation: 2 u = 1 α2 ∂u ∂t Wave equation: 2 u = 1 v2 ∂2u ∂t2 Helmoltz equation: 2 F + k2 F = 0 Schröedinger equation: − 2 2m 2 Ψ + V Ψ = i ∂ ∂t Ψ
  • 4. Laplace’s Equation: Example (Steady-State temperature in a rectangular plate) We have a rectangular plate on the region: R : {0 < x < 10 , 0 < y} with border conditions: T(x = 0) = T(x = 10) = T(y = ∞) = 0◦ and T(y = 0) = 100◦ Solution 2 T = 0 ∂2T ∂x2 + ∂2T ∂y2 = 0 T(x, y) = X(x)Y (y) Y ∂2X ∂x2 + X ∂2Y ∂y2 = 0
  • 5. Solution Y ∂2X ∂x2 + X ∂2Y ∂y2 = 0 1 X ∂2X ∂x2 + 1 Y ∂2Y ∂y2 = 0 ⇒ 1 X ∂2X ∂x2 = − 1 Y ∂2Y ∂y2 = const. = −k2 X = −k2 X and Y = +k2 Y (soln’s:) X(x) = A cos kx + B sin kx and Y (y) = Ceky + De−ky
  • 6. Solution T(x, y) = X(x)Y (y) = (A cos kx + B sin kx)(Ceky + De−ky ) B.C. : T(x, ∞) = 0 ⇒ C = 0 T(0, y) = 0 ⇒ A = 0 and T(10, y) = 0 ⇒ sin 10k = 0 ⇒ k = n π 10 ) ⇒ T(x, y) = BD sin n π 10 x e−n π 10 y andT(x, 0) = 100 ⇒ BD sin n π 10 x · 1 = 100
  • 7. Solution T(x, y) = ∞ n=1 bne−n π 10 y sin n π 10 x For y = 0, we must have T = 100, then : T(x, 0) = ∞ n=1 bn sin n π 10 x which is just the Fourier sine series · · ·
  • 8. Fourier Series: Similar to Taylor series, which we approx. a function using the summation of the derivatives of the function and polynomials, a Fourier series will approx. a function with the summation of cosine and sine functions: f (x) = a0 + ∞ k=1 an cos nπx + bn sin nπx where the coefficients are given by the Euler formulas a0 = 1 2 − f (x) dx, an = 1 − f (x) cos nπx dx, n = 1, 2, · · · bn = 1 − f (x) sin nπx dx, n = 1, 2, · · ·
  • 9. Approximation to a square wave Considere the square wave, which first wave is: f = 4 if 0 ≤ x ≤ π f = 0 if π < x < 2π and so on Example Use FS to approximate the square wave function. Solution The Euler coefficients: a0 = 1 2 − f (x) dx = 1 2π 0 −π 0 dx + π 0 4 dx = 2
  • 10. Solution an = 1 − f (x) cos nπx dx = 1 π 0 −π 0 dx + π 0 4 cos nπx π dx = 1 π π 0 4 cos nx dx = 1 π 4 sin nx n π 0 = 0 bn = 1 − f (x) sin nπx dx = 1 π 0 −π 0 dx + π 0 4 sin nπx π dx = 1 π π 0 4 sin nx dx = − 1 π 4 cos nx n π 0 = 4 πn (1 − cos nπ) ⇒ f (x) = 2 + 8 π sin x + 8 3π sin 3x + 8 5π sin 5x + · · ·
  • 11. Let’s come back to Laplace PDE · · · Solution T(x, 0) = 100 = ∞ n=1 bn sin n π 10 x We recognize f (x) = 100, = 10 therefore bn are given by: bn = 1 10 0 −10 0 · sin nπx 10 dx + 10 0 100 sin nπx 10 dx = 10 10 πn − 10 cos πn πn = 200 nπ , n = 1, 3, 5 · · · Finally: T(x, y) = ∞ n=1 bne−n π 10 y sin n π 10 x = 200 π e− π 10 y sin π 10 x + 1 3 e−3π 10 y sin 3π 10 x + · · ·
  • 12. The Diffusion or Heat Flow Equation: Example (Flow of Heat through a slab of thickness ) We have a slab place on the region: R : {0 < x < , −∞ < y < ∞} with border conditions: u(x = 0) = 0 and u(x = ) = 100◦ for t = 0 (beginning, a steady-state temperature distribution) Solution 2 u = 1 α2 ∂u ∂t , (α2 is a const. char. mate. and u the temp.) if u = F(x, y, z)T(t) ⇒ T 2 F = F α2 ∂T ∂t (dividing by u = FT) 1 F 2 F = 1 α2T ∂T ∂t ⇒ 1 F 2 F = −k2 and 1 α2T ∂T ∂t = −k2
  • 13. Solution The solution of the DFQ dependent of the time is trivial: T(t) = e−k2α2t and in our particular problem, where we have a very long slab, the diffusion will be only on the x-direction, then the DFQ dependent on the space coordinates, also become trivial, because we recognize as the SHO. ∂2F(x) ∂x2 + k2 F(x) = 0 soln : F(x) =A cos kx + B sin kx
  • 14. Solution B.C.: u(x = 0) = 0 ⇒ A = 0, and we allow u(x = ) = 0 at a later time(diffusion). Then if u(x = ) = 0 ⇒ sin k = 0 ⇒ k = nπ ⇒ k = nπ (eigenvalues) Then our base functions (eigenfunctions) are then u = e−k2α2t sin nπx and the general solution to our problem will be the series: u = ∞ n=1 bne−k2α2t sin nπx
  • 15. Solution The problem say at the beginning, a steady-state temperature distribution is on the slab, which implies u0 satisfy Laplace’s equation, i.e 2u0 = 0, i.e. ∂u2 0 ∂x2 = 0, and the solution for this equation is u0 = ax + b, then applying border condition we get u0 = 100 x Then for t = 0, u(x, t = 0) = u0, i.e. u(x, t = 0) = ∞ n=1 bn sin nπx = 100 x Which can be solved using F.S., for f (x) = 100 x and the half-period equal to
  • 16. Solution The Euler coefficients: an = 0 (we dont want solutions with cosine functions (B.C.)) bn = 1 − f (x) sin nπx dx = 1 0 − 0 dx + 0 100 x sin nπx dx = 100 2 π 0 x sin nπx dx ⇒ (Integration by parts:) bn = 100(sin πn − πn cos πn) π2n2 = 100 π (−1)n−1 n ⇒ u = 100 π ∞ n=1 (−1)n−1 n e−k2α2t sin nπx
  • 17. The Wave Equation: Example (The Vibrating String) Let a string be stretched tightly and its ends fastened to supports at x = 0 and x = . When the string is vibrating, its vertical displacement y from its equilibrium position along the x-axis depends on x and t. We assume the displacement y is very small and that the slope ∂y/∂t is small at any point at any time. Solution ∂2y ∂x2 = 1 v2 ∂2y ∂t2 , (v = T/µ) if y = X(x)T(t) ⇒ 1 X ∂2X ∂x2 = 1 v2 1 T ∂2T ∂t2 = −k2 (soln’s:) X(x) = A cos kx + B sin kx and T(t) = C cos kvt + D sin kvt
  • 18. Solution B.C.: y(x = 0) = 0 ⇒ A = 0, u(x = ) = 0 ⇒ sin k = 0 ⇒ k = nπ ⇒ k = nπ (eigenvalues) Then: y1(x, t) = B sin nπ x C cos nπ vt + D sin nπ vt but at t = 0 every piece of string is not varying with time, the ∂y/∂t = 0 Then D = 0 as well. Then our base functions (eigenfunctions) are then y1 = BC sin nπ x cos nπ vt and the general solution to our problem will be the series: y(x, t) = ∞ n=1 bn sin nπ x cos nπ vt
  • 19. Solution Then at t = 0, y(x, t) = y0 = f (x), then y(x, t = 0) = ∞ n=1 bn sin nπ x = f (x) · · ·