📘 Lecture 40 — Fourier Series
📖 Overview: This lecture provides a comprehensive derivation and application of Fourier series coefficients (A₀, aₙ, bₙ) and presents the final formulas for representing periodic functions. It also covers the Dirichlet conditions for convergence and demonstrates the effect of harmonics in reconstructing square waves, showing how adding more terms improves accuracy.
🗂️ Topics Covered
The lecture begins by reviewing the general form of a Fourier series, then derives the Fourier coefficients A₀, aₙ, and bₙ through integration. It presents the final result formulas using a₀, aₙ, and bₙ, followed by a worked example using limits from 0 to 2π. The Dirichlet conditions for convergence are then explained, followed by a second example involving a piecewise function with discontinuities. Finally, the lecture visualizes the effect of harmonics by plotting successive approximations of a square wave.
📝 Lecture Summary
Fourier series
As we know that A₀ + Σ {aₙ cos nx + bₙ sin nx}; n=1 to ∞, which can be written in the expanded form: A₀ + (a₁ cos x + b₁ sin x) + (a₂ cos 2x + b₂ sin 2x) + ... + (aₙ cos nx + bₙ sin nx) + ...
f(x) = A₀ + a₁ cos x + a₂ cos 2x + ... + aₙ cos nx + ... + b₁ sin x + b₂ sin 2x + ... + bₙ sin nx + ...
Fourier coefficients
We have defined Fourier series in the form f(x) = A₀ + Σ {aₙ cos nx + bₙ sin nx}; n a positive integer.
(a) To find A₀, we integrate f(x) with respect to x from -π to π. ∫₋π^π f(x) dx = ∫₋π^π A₀ dx + Σₙ₌₁^∞ {∫₋π^π aₙ cos nx dx + ∫₋π^π bₙ sin nx dx} = [A₀x]₋π^π + Σₙ₌₁^∞ {0 + 0} = 2A₀π
Therefore, 2A₀π = ∫₋π^π f(x) dx, so A₀ = (1/2π) ∫₋π^π f(x) dx = 1/2 a₀, where a₀ = (1/π) ∫₋π^π f(x) dx.
🔑 Definition — A₀ (average term): The constant term in the Fourier series, equal to half the mean value of the function over one period. It represents the average value of the function.
📌 Example: For the function f(x) = x/2 over 0 < x < 2π, a₀ = (1/π)∫₀²π (x/2) dx = π.
(b) To find aₙ, we multiply f(x) by cos mx and integrate from -π to π. ∫₋π^π f(x) cos mx dx = ∫₋π^π A₀ cos mx dx + Σₙ₌₁^∞ {∫₋π^π aₙ cos nx cos mx dx + ∫₋π^π bₙ sin nx cos mx dx} = 0 for n ≠ m, and = aₙπ for n = m.
Therefore, aₙ = (1/π) ∫₋π^π f(x) cos nx dx.
(c) To find bₙ, we multiply f(x) by sin mx and integrate from -π to π. ∫₋π^π f(x) sin mx dx = 0 for n ≠ m, and = bₙπ for n = m.
Therefore, bₙ = (1/π) ∫₋π^π f(x) sin nx dx.
Result For Fourier Series
f(x) = (1/2) a₀ + Σₙ₌₁^∞ {aₙ cos nx + bₙ sin nx}
(a) a₀ = (1/π) ∫₋π^π f(x) dx = 2 × mean value of f(x) over a period
(b) aₙ = (1/π) ∫₋π^π f(x) cos nx dx = 2 × mean value of f(x) cos nx over a period
(c) bₙ = (1/π) ∫₋π^π f(x) sin nx dx = 2 × mean value of f(x) sin nx over a period
In each case, n = 1, 2, 3, ...
📐 Formula: Standard Fourier series: f(x) = a₀/2 + Σₙ₌₁^∞ (aₙ cos nx + bₙ sin nx) → The sum of a constant term (average value) plus an infinite series of cosine and sine harmonics.
Example
Determine the Fourier series to represent the periodic function f(x) = x/2, 0 < x < 2π, f(x) = f(x + 2π). It is more convenient here to take the limits as 0 to 2π.
(a) a₀: a₀ = (1/π)∫₀²π f(x) dx = (1/π)∫₀²π (x/2) dx = (1/4π)[x²]₀²π = π
(b) aₙ: aₙ = (1/π)∫₀²π f(x) cos nx dx = (1/2π)∫₀²π x cos nx dx Using integration by parts, aₙ = 0 → aₙ = 0 (all cosine harmonics vanish because the function is odd about the midpoint of the interval)
(c) bₙ: bₙ = (1/π)∫₀²π f(x) sin nx dx = (1/2π)∫₀²π x sin nx dx Using integration by parts, bₙ = -(1/n) → bₙ = -1/n
📌 Example: Since a₀ = π, aₙ = 0, and bₙ = -1/n, the Fourier series becomes: f(x) = a₀/2 + Σₙ₌₁^∞ {bₙ sin nx} = π/2 - {sin x + (1/2) sin 2x + (1/3) sin 3x + ...}
Therefore, f(x) = π/2 - (sin x + (1/2) sin 2x + (1/3) sin 3x + ...)
Dirichlet Conditions
If the Fourier series is to represent a function f(x), then putting x = x₁ will give an infinite series in x₁, and the value of this should converge to the value of f(x₁) as more and more terms of the series are evaluated. For this to happen, the following conditions must be fulfilled:
(a) The function f(x) must be defined and single-valued.
(b) f(x) must be continuous or have a finite number of finite discontinuities within a periodic interval.
(c) f(x) and f'(x) must be piecewise continuous in the periodic interval.
If these Dirichlet conditions are satisfied, the Fourier series converges to f(x₁), if x = x₁ is a point of continuity.
💡 Why this matters: These conditions ensure the Fourier series actually converges to the function. Without them, the series might diverge or converge to the wrong value at discontinuities.
Example
Find the Fourier series for the function defined by: f(x) = 0 for -π < x < -π/2; f(x) = 4 for -π/2 < x < π/2; f(x) = 0 for π/2 < x < π; f(x) = f(x + 2π).
(a) a₀: a₀ = (1/π)∫₋π^π f(x) dx = (1/π)[∫₋π^{-π/2} 0 dx + ∫₋π/2^{π/2} 4 dx + ∫₋π/2^π 0 dx] = (1/π)[4x]₋π/2^{π/2} = 4 → a₀ = 4
(b) aₙ: aₙ = (1/π)∫₋π^π f(x) cos nx dx = (4/π)∫₋π/2^{π/2} cos nx dx = (8/nπ) sin(nπ/2)
If n is even, aₙ = 0. If n = 1, 5, 9,... aₙ = 8/(nπ). If n = 3, 7, 11,... aₙ = -8/(nπ).
(c) bₙ: bₙ = (4/π)∫₋π/2^{π/2} sin nx dx = -(4/nπ)[cos(nπ/2) - cos(-nπ/2)] = 0 → bₙ = 0 (all sine terms vanish due to symmetry)
📌 Example: The Fourier series is: f(x) = a₀/2 + Σₙ₌₁^∞ {aₙ cos nx + bₙ sin nx} f(x) = 2 + (8/π){cos x - (1/3) cos 3x + (1/5) cos 5x - (1/7) cos 7x + ...}
In this particular example, there are no sine terms (bₙ = 0).
Effect Of Harmonics
It is interesting to see just how accurately the Fourier series represents the function with which it is associated. The complete representation requires an infinite number of terms, but we can see the effect of including the first few terms.
For the square waveform f(x) = 0 for -π < x < -π/2; f(x) = 4 for -π/2 < x < π/2; f(x) = 0 for π/2 < x < π; f(x) = f(x + 2π):
(1) f(x) = 2 + (8/π) cos x
(2) f(x) = 2 + (8/π){cos x - (1/3) cos 3x}
(3) f(x) = 2 + (8/π){cos x - (1/3) cos 3x + (1/5) cos 5x}
(4) f(x) = 2 + (8/π){cos x - (1/3) cos 3x + (1/5) cos 5x - (1/7) cos 7x}
As the number of terms is increased, the graph gradually approaches the shape of the original square waveform. The ripples increase in number and decrease in amplitude, but a perfectly square waveform is unattainable in practice. For practical purposes, the first few terms normally suffice to give an accuracy of acceptable level.
💡 Why this matters: This demonstrates the Gibbs phenomenon — the overshoot at discontinuities persists even with many terms, but the approximation becomes visually acceptable for engineering applications.
⭐ Key Takeaways
The Fourier series formula f(x) = a₀/2 + Σ(aₙ cos nx + bₙ sin nx) requires computing three coefficients: a₀ from the average value, aₙ from multiplying by cosine, and bₙ from multiplying by sine, all integrated over one period. The Dirichlet conditions guarantee convergence at points of continuity, but at discontinuities, the series converges to the midpoint value. The effect of harmonics shows that adding more terms improves the approximation, with ripples diminishing in amplitude, making the series practically useful for signal representation.
🧠 Quick Revision Questions
-
Derive the formula for A₀ by integrating the Fourier series from -π to π. What does A₀ represent physically?
-
Why do the integrals for aₙ and bₙ vanish when n ≠ m, giving only aₙπ or bₙπ for n = m?
-
State the three Dirichlet conditions. What happens if the function fails to satisfy them at a discontinuity?
-
In the square wave example (f(x) = 4 on -π/2 to π/2), why are all sine coefficients bₙ equal to zero?
-
What is the Fourier series approximation using only the first two non-zero cosine harmonics for the square wave with amplitude 4?
📘 Lecture 41 — Examples
📖 Overview: This lecture demonstrates how to compute Fourier series for piecewise-defined functions through detailed examples. It also introduces the important concepts of odd and even functions, their properties, and how these symmetries simplify Fourier series calculations by eliminating either sine or cosine terms.
🗂️ Topics Covered
The lecture covers a complete step-by-step example of finding the Fourier series for a piecewise function defined on $(-\pi, \pi)$. It then introduces definitions and graphical interpretations of odd and even functions, examines products of odd and even functions with examples, presents two key theorems about Fourier series for odd and even functions, and provides additional examples demonstrating these simplifications.
📝 Lecture Summary
Example — Fourier Series for a Piecewise Function
Find the Fourier series for the function defined by $f(x) = -x$ for $-\pi < x < 0$, $f(x) = 0$ for $0 < x < \pi$, and $f(x) = f(x + 2\pi)$.
The general expressions for $a_0$, $a_n$, $b_n$ are:
$$a_0 = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) , dx , \quad a_n = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) \cos nx , dx , \quad b_n = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) \sin nx , dx$$
To find $a_0$:
$$a_0 = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) , dx = \frac{1}{\pi} \int_{-\pi}^{0} (-x) , dx + \frac{1}{\pi} \int_{0}^{\pi} 0 , dx = \frac{1}{\pi} \left[ -\frac{x^2}{2} \right]_{-\pi}^{0} = \frac{\pi}{2}$$
To find $a_n$:
$$a_n = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) \cos nx , dx = \frac{1}{\pi} \int_{-\pi}^{0} (-x) \cos nx , dx + \frac{1}{\pi} \int_{0}^{\pi} 0 , dx = -\frac{1}{\pi} \int_{-\pi}^{0} x \cos nx , dx$$
Using integration by parts:
$$a_n = -\frac{1}{\pi} \left[ \left( x \frac{\sin nx}{n} \right){-\pi}^{0} - \frac{1}{n} \int{-\pi}^{0} \sin nx , dx \right]$$
$$= -\frac{1}{\pi} \left[ (0-0) - \frac{1}{n} \left( -\frac{\cos nx}{n} \right){-\pi}^{0} \right] = -\frac{1}{\pi n^2} \left[ \cos nx \right]{-\pi}^{0}$$
$$= -\frac{1}{\pi n^2} [\cos 0 - \cos(-n\pi)] = -\frac{1}{\pi n^2} [1 - \cos n\pi]$$
Since $\cos n\pi = 1$ for $n$ even and $\cos n\pi = -1$ for $n$ odd:
$$a_n = -\frac{2}{\pi n^2} \text{ (n odd)} \quad \text{and} \quad a_n = 0 \text{ (n even)}$$
To find $b_n$:
$$b_n = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) \sin nx , dx = \frac{1}{\pi} \int_{-\pi}^{0} (-x) \sin nx , dx = -\frac{1}{\pi} \int_{-\pi}^{0} x \sin nx , dx$$
$$= -\frac{1}{\pi} \left[ \left( x \cdot \frac{-\cos nx}{n} \right){-\pi}^{0} + \frac{1}{n} \int{-\pi}^{0} \cos nx , dx \right]$$
$$= -\frac{1}{\pi} \left[ \frac{\pi \cos n\pi}{n} + \frac{1}{n} \left( \frac{\sin nx}{n} \right)_{-\pi}^{0} \right] = -\frac{\cos n\pi}{n}$$
$$b_n = -\frac{1}{n} \text{ (n even)} \quad \text{and} \quad b_n = \frac{1}{n} \text{ (n odd)}$$
Final Fourier series:
$$f(x) = \frac{\pi}{4} - \frac{2}{\pi} \left( \cos x + \frac{1}{9} \cos 3x + \frac{1}{25} \cos 5x + \cdots \right) + \left( \sin x - \frac{1}{2} \sin 2x + \frac{1}{3} \sin 3x - \frac{1}{4} \sin 4x + \cdots \right)$$
This is obtained by substituting $n = 1, 2, 3,$ etc. into the expressions for $a_0$, $a_n$, and $b_n$. This function has a constant term and both sine and cosine terms.
Odd and Even Functions
(a) Even functions
A function $f(x)$ is said to be even if $f(-x) = f(x)$, meaning the function value for a particular negative value of $x$ is the same as that for the corresponding positive value of $x$. The graph of an even function is symmetrical about the y-axis.
- $y = f(x) = x^2$ is an even function since $f(-2) = 4 = f(2)$, $f(-3) = 9 = f(3)$, etc.
- $y = f(x) = \cos x$ is an even function since $\cos(-x) = \cos x$, so $f(-a) = \cos a = f(a)$.
(b) Odd functions
A function $f(x)$ is said to be odd if $f(-x) = -f(x)$, meaning the function value for a particular negative value of $x$ is numerically equal to that for the corresponding positive value of $x$ but opposite in sign. The graph of an odd function is symmetrical about the origin.
- $y = f(x) = x^3$ is an odd function since $f(-2) = -8 = -f(2)$, $f(-5) = -125 = -f(5)$, etc.
- $y = f(x) = \sin x$ is an odd function since $\sin(-x) = -\sin x$, so $f(-a) = -\sin a = -f(a)$.
🔑 Definition — Even function: $f(-x) = f(x)$, symmetrical about the y-axis. 🔑 Definition — Odd function: $f(-x) = -f(x)$, symmetrical about the origin.
Products of Odd and Even Functions
The rules for products of odd and even functions closely resemble the elementary rules of signs:
- (even) × (even) = (even), like (+) × (+) = (+)
- (odd) × (odd) = (even), like (-) × (-) = (+)
- (odd) × (even) = (odd), like (-) × (+) = (-)
(a) Two even functions: Let $F(x) = f(x)g(x)$ where $f(x)$ and $g(x)$ are even. Then $F(-x) = f(-x)g(-x) = f(x)g(x) = F(x)$, so $F(x)$ is even.
(b) Two odd functions: Let $F(x) = f(x)g(x)$ where $f(x)$ and $g(x)$ are odd. Then $F(-x) = f(-x)g(-x) = {-f(x)}{-g(x)} = f(x)g(x) = F(x)$, so $F(x)$ is even.
(c) One odd and one even function: Let $F(x) = f(x)g(x)$ where $f(x)$ is odd and $g(x)$ is even. Then $F(-x) = f(-x)g(-x) = -f(x)g(x) = -F(x)$, so $F(x)$ is odd.
Summary: If both functions are even or both are odd, their product is even. If one is even and the other is odd, their product is odd.
Example — Classifying products:
- $x^2 \sin 2x$: odd (even × odd = odd)
- $x^3 \cos x$: odd (odd × even = odd)
- $\cos 2x \cos 3x$: even (even × even = even)
- $x \sin nx$: even (odd × odd = even)
- $3 \sin x \cos 4x$: odd (odd × even = odd)
- $(2x+3) \sin 4x$: neither (neither × odd = neither)
- $\sin^2 x \cos 3x$: even (even × even = even)
- $x^3 e^x$: neither (odd × neither = neither)
- $(x^4+4) \sin 2x$: odd (even × odd = odd)
Two useful facts emerge from odd and even functions:
(a) For an even function: $\int_{-a}^{a} f(x) , dx = 2 \int_{0}^{a} f(x) , dx$
(b) For an odd function: $\int_{-a}^{a} f(x) , dx = 0$
Theorem 1 — Fourier Series for Even Functions
If $f(x)$ is defined over the interval $-\pi < x < \pi$ and $f(x)$ is even, then the Fourier series for $f(x)$ contains cosine terms only (including $a_0$, which may be regarded as $a_n \cos nx$ with $n=0$).
Proof:
(a) $a_0 = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) , dx = \frac{2}{\pi} \int_{0}^{\pi} f(x) , dx$ (since $f(x)$ is even)
(b) $a_n = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) \cos nx , dx$. Since $f(x)\cos nx$ is the product of two even functions, it is even. Therefore: $$a_n = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) \cos nx , dx = \frac{2}{\pi} \int_{0}^{\pi} f(x) \cos nx , dx$$
(c) $b_n = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) \sin nx , dx$. Since $f(x)\sin nx$ is the product of an even function and an odd function, it is odd. Therefore: $$b_n = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) \sin nx , dx = 0$$
Therefore, there are no sine terms in the Fourier series for an even function.
Example — Even function waveform:
Consider a waveform symmetrical about the y-axis. The function is even, so there will be no sine terms:
$$f(x) = a_0 + \sum_{n=1}^{\infty} a_n \cos nx$$
For the given function with $f(x) = 4$ on $(-\pi/2, \pi/2)$ and zero elsewhere:
(a) $a_0 = \frac{2}{\pi} \int_{0}^{\pi/2} 4 , dx = \frac{8}{\pi} \cdot \frac{\pi}{2} = 4$
(b) $a_n = \frac{2}{\pi} \int_{0}^{\pi/2} 4 \cos nx , dx = \frac{8}{\pi} \left[ \frac{\sin nx}{n} \right]_{0}^{\pi/2} = \frac{8}{\pi n} \sin \frac{n\pi}{2}$
Since $\sin \frac{n\pi}{2} = 0$ for $n$ even, $= 1$ for $n = 1, 5, 9, \ldots$, and $= -1$ for $n = 3, 7, 11, \ldots$:
- $a_n = 0$ (n even)
- $a_n = \frac{8}{\pi n}$ (n = 1, 5, 9, ...)
- $a_n = -\frac{8}{\pi n}$ (n = 3, 7, 11, ...)
(c) $b_n = 0$ since $f(x)$ is even.
Required series: $$f(x) = 2 + \frac{8}{\pi} \left{ \cos x - \frac{1}{3} \cos 3x + \frac{1}{5} \cos 5x - \frac{1}{7} \cos 7x + \cdots \right}$$
Theorem 2 — Fourier Series for Odd Functions
If $f(x)$ is an odd function defined over the interval $-\pi < x < \pi$, then the Fourier series for $f(x)$ contains sine terms only.
Proof:
Since $f(x)$ is odd, $\int_{-\pi}^{0} f(x) , dx = -\int_{0}^{\pi} f(x) , dx$.
(a) $a_0 = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) , dx = 0$ (since $f(x)$ is odd)
(b) $a_n = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) \cos nx , dx$. Since $f(x)$ is odd and $\cos nx$ is even, the product $f(x)\cos nx$ is odd. Therefore: $$a_n = \frac{1}{\pi} \int_{-\pi}^{\pi} (\text{odd function}) , dx = 0$$
(c) $b_n = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) \sin nx , dx$. Since $f(x)$ and $\sin nx$ are each odd, the product $f(x)\sin nx$ is even. Therefore: $$b_n = \frac{1}{\pi} \int_{-\pi}^{\pi} (\text{even function}) , dx = \frac{2}{\pi} \int_{0}^{\pi} f(x) \sin nx , dx$$
So for an odd function: $a_0 = 0$, $a_n = 0$, and $b_n = \frac{2}{\pi} \int_{0}^{\pi} f(x) \sin nx , dx$. The Fourier series contains sine terms only.
Example — Odd function: $$f(x) = -6 \text{ for } -\pi < x < 0, \quad f(x) = 6 \text{ for } 0 < x < \pi, \quad f(x) = f(x+2\pi)$$
This is an odd function, so $a_0 = 0$ and $a_n = 0$.
$$b_n = \frac{1}{\pi} \int_{-\pi}^{\pi} f(x) \sin nx , dx$$
Since $f(x)\sin nx$ is the product of two odd functions, it is even:
$$b_n = \frac{2}{\pi} \int_{0}^{\pi} 6 \sin nx , dx = \frac{12}{\pi} \left[ -\frac{\cos nx}{n} \right]_{0}^{\pi} = \frac{12}{\pi n} [\cos 0 - \cos n\pi] = \frac{12}{\pi n} (1 - \cos n\pi)$$
$$b_n = 0 \text{ (n even)}, \quad b_n = \frac{24}{\pi n} \text{ (n odd)}$$
Series: $$f(x) = \frac{24}{\pi} \left{ \sin x + \frac{1}{3} \sin 3x + \frac{1}{5} \sin 5x + \cdots \right}$$
💡 Why this matters: If a function is neither odd nor even, we must compute $a_0$, $a_n$, and $b_n$ in full without any simplification.
⭐ Key Takeaways
The most critical things to remember are: (1) Even functions satisfy $f(-x)=f(x)$ and have Fourier series with only cosine terms (including $a_0$), while odd functions satisfy $f(-x)=-f(x)$ and have Fourier series with only sine terms and $a_0=0$. (2) For even functions, $a_n = \frac{2}{\pi} \int_0^\pi f(x) \cos nx , dx$ and $b_n=0$; for odd functions, $b_n = \frac{2}{\pi} \int_0^\pi f(x) \sin nx , dx$ and $a_0=a_n=0$. (3) The product rules are: even×even=even, odd×odd=even, and odd×even=odd. (4) For integrals: even functions satisfy $\int_{-a}^a f(x)dx = 2\int_0^a f(x)dx$ and odd functions satisfy $\int_{-a}^a f(x)dx = 0$. (5) If a function is neither odd nor even, all Fourier coefficients must be computed in full without simplification.
🧠 Quick Revision Questions
-
What are the defining properties of even and odd functions, and what does their symmetry look like graphically?
-
State the product rules for odd and even functions: what is the result when multiplying two even functions, two odd functions, and one odd with one even function?
-
For an even function defined on $(-\pi, \pi)$, write simplified formulas for $a_0$, $a_n$, and $b_n$ in terms of integrals from $0$ to $\pi$.
-
For an odd function defined on $(-\pi, \pi)$, what are the values of $a_0$ and $a_n$, and write the simplified formula for $b_n$?
-
In the first example where $f(x) = -x$ on $(-\pi, 0)$ and $f(x) = 0$ on $(0, \pi)$, is this function odd, even, or neither? How can you tell from the Fourier series obtained?
📘 Lecture 42 — Examples
📖 Overview: This lecture presents practical examples of Fourier series calculations for functions with period 2π, including functions that are neither odd nor even, functions with discontinuities, and half-range series expansions. It demonstrates how to compute Fourier coefficients, handle points of discontinuity, and use symmetry to simplify calculations.
🗂️ Topics Covered
The lecture covers a complete worked example of finding Fourier series for a piecewise function that is neither odd nor even, the behavior of Fourier series at points of finite discontinuity, half-range cosine and sine series for functions defined over (0, π), and the application of odd/even function properties to simplify calculations.
📝 Lecture Summary
Example (First Example)
Determine the Fourier series for the function shown which is neither odd nor even, defined as f(x) = (2/π)x for 0 < x < π, and f(x) = 2 for π < x < 2π.
The general Fourier series is: f(x) = ½a₀ + Σ_{n=1}^{∞} {aₙ cos nx + bₙ sin nx}
Finding a₀: a₀ = (1/π) ∫₀²π f(x)dx = (1/π) {∫₀π (2/π)x dx + ∫π²π 2 dx} = (1/π) { [x²/π]₀π + [2x]π²π } = (1/π) {π + 4π - 2π} a₀ = 3
Finding aₙ: aₙ = (1/π) ∫₀²π f(x) cos nx dx = (1/π) {∫₀π (2/π)x cos nx dx + ∫π²π 2 cos nx dx} = (2/π²) { (1/n)[x sin nx]₀π - (1/n) ∫₀π sin nx dx } + (2/π) ∫π²π cos nx dx After integration: aₙ = (2/(π²n²)) {cos nπ - 1} aₙ = 0 (n even); aₙ = -4/(π²n²) (n odd)
Finding bₙ: bₙ = (1/π) ∫₀²π f(x) sin nx dx = (1/π) {∫₀π (2/π)x sin nx dx + ∫π²π 2 sin nx dx} = (2/π²) { (-x cos nx)/n + (1/n) ∫ cos nx dx }₀π + (2/π) ∫π²π sin nx dx After integration: bₙ = (2/π) {(-1/n) cos nπ + 0 - (1/n)(cos 2πn - cos nπ)} = (2/π) {-(1/n) cos 2nπ} Since cos 2nπ = 1: bₙ = -2/(πn)
🔑 Definition — Fourier series: A representation of a periodic function as an infinite sum of sine and cosine functions.
📐 Formula: f(x) = ½a₀ + Σ_{n=1}^{∞} {aₙ cos nx + bₙ sin nx} where a₀, aₙ, bₙ are Fourier coefficients.
📌 Example: The required Fourier series is: f(x) = 3/2 - (4/π²){cos x + (1/9) cos 3x + (1/25) cos 5x + ...} - (2/π){sin x + (1/2) sin 2x + (1/3) sin 3x + (1/4) sin 4x + ...}
Sum of a Fourier series at a point of discontinuity
At a point of finite discontinuity or "jump" of a function y = f(x), the function has two distinct values: y₁ = f(x₁ - 0) approaching from below, and y₂ = f(x₁ + 0) approaching from above. When x = x₁ is substituted into the Fourier series, it converges to the average value ½{f(x₁ - 0) + f(x₁ + 0)} = ½(y₁ + y₂).
🔑 Definition — Point of finite discontinuity: A point where a function has a jump, with different limiting values from the left and right.
📌 Example: Consider f(x) = 0 for -π < x < 0, f(x) = a for 0 < x < π, period 2π. a₀ = a, aₙ = 0, bₙ = 2a/(nπ) for n odd, bₙ = 0 for n even The series: f(x) = a/2 + (2a/π){sin x + (1/3) sin 3x + (1/5) sin 5x + ...} At x = 0 (a discontinuity), all sine terms vanish, giving f(0) = a/2, which is the average of 0 and a.
Half-Range Series
Sometimes a function of period 2π is defined only over the range 0 to π. We can choose to represent it as an even function (giving a cosine series) or an odd function (giving a sine series), known as half-range series.
🔑 Definition — Half-range series: A Fourier series representing a function defined only on (0, π) by extending it as either an even or odd function over (-π, 0).
Half-Range Cosine Series Example: For f(x) = 2x, 0 < x < π, period 2π, to obtain a cosine series, assume the function is even. a₀ = (2/π) ∫₀π 2x dx = 2π aₙ = (2/π) ∫₀π 2x cos nx dx = (4/(πn²))(cos nπ - 1) aₙ = 0 (n even); aₙ = -8/(πn²) (n odd) bₙ = 0 (since function is even) f(x) = π - (8/π){cos x + (1/9) cos 3x + (1/25) cos 5x + ...}
Half-Range Sine Series Example: For f(x) = 1 + x, 0 < x < π, period 2π, to obtain a sine series, assume the function is odd. a₀ = 0, aₙ = 0 bₙ = (2/π) ∫₀π (1+x) sin nx dx = (2/(πn)){1 - (1+π) cos nπ} bₙ = -2/n (n even); bₙ = (4+2π)/(πn) (n odd) f(x) = (4/π + 2){sin x + (1/3) sin 3x + (1/5) sin 5x + ...} - 2{(1/2) sin 2x + (1/4) sin 4x + (1/6) sin 6x + ...}
💡 Why this matters: Using half-range series and knowing odd/even properties saves significant computational work by eliminating unnecessary coefficient calculations.
⭐ Key Takeaways
The Fourier series at a point of discontinuity converges to the average of the left and right limiting values, not to either value individually. For functions defined only on (0, π), half-range series allow choosing between cosine series (even extension) or sine series (odd extension), which eliminates the need to compute the corresponding coefficients. Understanding odd and even function properties (aₙ = 0 for odd functions, bₙ = 0 for even functions) dramatically simplifies Fourier series calculations. The determination of Fourier coefficients requires careful integration by parts for products like x cos nx and x sin nx. When computing bₙ, the integral of sin nx over a full period [0, 2π] is zero, but the integral of x sin nx over [0, π] requires integration by parts.
🧠 Quick Revision Questions
- At a point of discontinuity x = 0, where f(0⁻) = 0 and f(0⁺) = a, what value does the Fourier series converge to?
- For a half-range cosine series, what assumption is made about the function between x = -π and x = 0?
- What is the general formula for aₙ when f(x) = x over (0, π) and the function is extended as an even function?
- In the first example of this lecture, why is bₙ = -2/(πn) and not dependent on n being odd or even?
- For a function defined only on (0, π), what are the two possible half-range series representations called?
📘 Lecture 43 — Functions With Periods Other Than 2π
📖 Overview: This lecture extends Fourier series analysis to functions with arbitrary period T, which is crucial for real-world applications like engineering and physics where functions rarely have a period of exactly 2π. It covers the general Fourier series for period T, the calculation of Fourier coefficients with the new variable, and practical examples including half-range series and half-wave rectifiers.
🗂️ Topics Covered
The lecture covers functions with period T, converting the independent variable using angular velocity ω, deriving Fourier coefficients for general period, worked examples of periodic functions, half-range sine and cosine series for even and odd functions, and the Fourier series of a half-wave rectifier.
📝 Lecture Summary
Functions With Period T
If y = f(x) is defined over an interval of length T (period T), we can convert this to the standard 2π interval by changing the units of the independent variable. For physical oscillations, the independent variable is often time (t), with period T seconds, so f(t) = f(t + T). The frequency is f = 1/T hertz, and the angular velocity ω = 2πf, so ω = 2π/T and T = 2π/ω. The Fourier series in terms of t is: f(t) = 1/2 a₀ + Σ_{n=1}^{∞} [aₙ cos(nωt) + bₙ sin(nωt)] which can also be written as f(t) = 1/2 A₀ + Σ_{n=1}^{∞} Bₙ sin(nωt + φₙ).
Fourier Coefficients
With the new variable, the Fourier coefficients for a function of period T are:
- a₀ = (2/T) ∫₀ᵀ f(t) dt = (ω/π) ∫₀^{2π/ω} f(t) dt
- aₙ = (2/T) ∫₀ᵀ f(t) cos(nωt) dt = (ω/π) ∫₀^{2π/ω} f(t) cos(nωt) dt
- bₙ = (2/T) ∫₀ᵀ f(t) sin(nωt) dt = (ω/π) ∫₀^{2π/ω} f(t) sin(nωt) dt
The limits can be any interval covering a complete period, such as 0 to T, -T/2 to T/2, -π/ω to π/ω, or 0 to 2π/ω.
🔑 Definition — Angular Velocity: ω = 2π/T, where T is the period and ω is measured in radians per second. 📐 Formula: ω = 2π/T → The angular frequency relates the period to the standard circular frequency.
Example (Periodic Function)
Determine the Fourier series for f(t) = 2(1+t) for -1 < t < 0 and f(t) = 0 for 0 < t < 1, with f(t) = f(t+2). Given: T = 2, so ω = 2π/T = π. Coefficients:
- a₀ = (2/T) ∫{-T/2}^{T/2} f(t) dt = (2/2) ∫{-1}^{1} f(t) dt = ∫{-1}^{0} 2(1+t) dt + ∫₀¹ 0 dt = [2t + t²]{-1}^{0} = 1
- aₙ = (2/T) ∫{-1}^{1} f(t) cos(nωt) dt = ∫{-1}^{0} 2(1+t) cos(nπt) dt Using integration by parts: aₙ = 2/n²ω² (1 - cos(nπ)) So aₙ = 0 (n even) and aₙ = 4/n²ω² (n odd)
- bₙ = (2/T) ∫{-1}^{1} f(t) sin(nωt) dt = ∫{-1}^{0} 2(1+t) sin(nπt) dt Using integration by parts: bₙ = -2/nω (since sin(nπ) = 0)
📌 Example: The first few terms of the series give: f(t) = 1/2 + (4/ω²)[cos(ωt) + (1/9)cos(3ωt) + (1/25)cos(5ωt) + ...] - (2/ω)[sin(ωt) + (1/2)sin(2ωt) + (1/3)sin(3ωt) + ...]
Half-Range Series
The theory of half-range sine and cosine series applies with the new variable.
(a) Even function — Half-range cosine series: For y = f(t) with 0 < t < T/2, extended as an even function (symmetrical about y-axis), we have bₙ = 0. f(t) = 1/2 a₀ + Σ_{n=1}^{∞} aₙ cos(nωt) where a₀ = (4/T) ∫₀^{T/2} f(t) dt and aₙ = (4/T) ∫₀^{T/2} f(t) cos(nωt) dt
(b) Odd function — Half-range sine series: For y = f(t) with 0 < t < T/2, extended as an odd function (symmetrical about origin), we have a₀ = 0 and aₙ = 0. f(t) = Σ_{n=1}^{∞} bₙ sin(nωt) where bₙ = (4/T) ∫₀^{T/2} f(t) sin(nωt) dt
Example — Half-Range Cosine Series
A function f(t) = 4 - t, 0 < t < 4. Form a half-range cosine series. Given: T = 8, extended as even function, T/2 = 4, ω = 2π/T = π/4.
- a₀ = (4/8) ∫₀⁴ (4 - t) dt = (1/2)[4t - t²/2]₀⁴ = (1/2)[16 - 8] = 4
- aₙ = (4/8) ∫₀⁴ (4 - t) cos(nωt) dt = (1/2) ∫₀⁴ (4 - t) cos(nπt/4) dt Integration by parts yields: aₙ = (1/2)[-2 sin(4nω)/(nω) - (1/n²ω²)(cos(4nω) - 1)] With ω = π/4, sin(nπ) = 0, cos(nπ) = 1 (n even) or -1 (n odd), so aₙ = 0 (n even) and aₙ = 1/n²ω² (n odd)
📌 Example: The half-range cosine series is: f(t) = 2 + (1/ω²)[cos(ωt) + (1/9)cos(3ωt) + (1/25)cos(5ωt) + ...] where ω = π/4
Example — Half-Range Sine Series
A function f(t) = 3 + t, 0 < t < 2, with f(t) = f(t+4). Obtain half-range sine series. Given: T = 4, extended as odd function, T/2 = 2, ω = 2π/T = π/2.
- bₙ = (4/4) ∫₀² (3 + t) sin(nωt) dt = ∫₀² (3 + t) sin(nπt/2) dt Integration by parts gives: bₙ = (1/nω)(3 - 5 cos(2nω)) + (1/n²ω²) sin(2nω) With ω = π/2, sin(nπ) = 0, cos(nπ) = 1 (n even) or -1 (n odd), so bₙ = -2/nω (n even) and bₙ = 8/n²ω² (n odd)
📌 Example: The half-range sine series is: f(t) = (2/ω)[(4/1)sin(ωt) - (1/2)sin(2ωt) + (4/3)sin(3ωt) - (1/4)sin(4ωt) + ...] where ω = π/2
💡 Why this matters: Half-range series allow representation of functions defined only on half a period, which is common in engineering problems involving piecewise or asymmetric signals.
Half-Wave Rectifier
A sinusoidal voltage E sin(ωt) is passed through a half-wave rectifier that clips the negative portion. Find the Fourier series. Given: u(t) = 0 for -T/2 < t < 0, u(t) = E sin(ωt) for 0 < t < T/2, with T = 2π/ω.
- a₀ = (2/T) ∫₀^{T/2} E sin(ωt) dt = (ω/π) ∫₀^{π/ω} E sin(ωt) dt = 2E/π
- aₙ = (2/T) ∫₀^{π/ω} E sin(ωt) cos(nωt) dt = (ωE/2π) ∫₀^{π/ω} [sin(1+n)ωt + sin(1-n)ωt] dt For n = 1, a₁ = 0. For n = 2, 3, ...: aₙ = 0 (n odd), aₙ = 2E/(π(1-n²)) (n even)
- bₙ = (2/T) ∫₀^{π/ω} E sin(ωt) sin(nωt) dt = -(ωE/2π) ∫₀^{π/ω} [cos(1+n)ωt - cos(1-n)ωt] dt For n = 1: b₁ = E/2. For n ≠ 1: bₙ = 0.
📌 Example: The Fourier series for the half-wave rectifier is: u(t) = E/π + (E/2) sin(ωt) - (2E/π)[(1/1·3)cos(2ωt) + (1/3·5)cos(4ωt) + ...]
💡 Why this matters: Half-wave rectification is fundamental in power electronics, converting AC to pulsating DC, and its Fourier series reveals the harmonic content.
⭐ Key Takeaways
The key concept is adapting Fourier series to any period T by introducing angular velocity ω = 2π/T, with coefficients aₙ and bₙ computed using integrals over one period. Half-range series (cosine for even, sine for odd) simplify representation for functions defined on half a period. The half-wave rectifier example demonstrates how to handle piecewise-defined periodic functions with discontinuities. Mastery of these techniques is essential for analyzing oscillatory systems in electrical engineering, signal processing, and mechanical vibrations.
🧠 Quick Revision Questions
- What is the formula for the angular velocity ω in terms of the period T, and how is it used to convert a function of period T to the standard Fourier series form?
- In the example of the function f(t) = 2(1+t) for -1<t<0, f(t)=0 for 0<t<1, with T=2, what are a₀ and the expressions for aₙ and bₙ?
- How do the formulas for a₀ and aₙ differ between a half-range cosine series and a full-range Fourier series for a periodic function?
- For the half-wave rectifier with input E sin(ωt), what are the values of a₁ and b₁, and why is a₁ zero?
- In the half-range sine series example for f(t) = 3+t (0<t<2, T=4), why are the bₙ coefficients different for odd and even n?
📘 Lecture 44 — Laplace transforms
📖 Overview: This lecture formally introduces the Laplace transform, defines its standard results for elementary functions (exponential, polynomial, trigonometric, and hyperbolic), and demonstrates how to find both transforms and inverse transforms. It also covers complex numbers and Euler’s formula, which are used to derive the transforms of sine and cosine. Understanding these fundamentals is essential for solving differential equations and analyzing systems in engineering and physics.
🗂️ Topics Covered
The lecture covers the formal definition of the Laplace transform, the derivation of transforms for constant functions, exponentials, and polynomials (tⁿ). It introduces complex number operations and Euler’s formula as tools to find the Laplace transforms of sin at and cos at. Standard results are extended to hyperbolic functions sinh at and cosh at. The lecture concludes with examples of linear combinations and a table of standard Laplace transforms.
📝 Lecture Summary
Laplace Transform
The Laplace Transform of a function F(t) is denoted by L{F(t)} and is defined as the integral of F(t) e⁻ˢᵗ between the limits t=0 and t = ∞. ∞ L{F(t)} = ∫₀ e⁻ˢᵗ F(t) dt In all cases, the constant parameter s is assumed to be positive and large enough to ensure that the product F(t)e⁻ˢᵗ converges to zero as t→∞. The result of the transform is always a function of s.
🔑 Definition — Laplace Transform: L{F(t)} = ∫₀∞ e⁻ˢᵗ F(t) dt
Laplace Transform of F(t) = a (constant)
∞ ∞ L(a) = ∫₀ a e⁻ˢᵗ dt = a ∫₀ e⁻ˢᵗ dt = a [e⁻ˢᵗ / -s]₀∞ = -a/s [0 - 1] = a/s
🔑 Definition — Laplace Transform of a constant: L{a} = a/s
Laplace Transform of eᵃᵗ
∞ L(eᵃᵗ) = ∫₀ eᵃᵗ e⁻ˢᵗ dt = ∫₀∞ e⁻⁽ˢ⁻ᵃ⁾ᵗ dt = [e⁻⁽ˢ⁻ᵃ⁾ᵗ / -(s-a)]₀∞ = -1/(s-a) [1/e⁽ˢ⁻ᵃ⁾ᵗ]₀∞ = -1/(s-a)[0 - 1] = 1/(s-a), for s > a
🔑 Definition — Laplace Transform of eᵃᵗ: L{eᵃᵗ} = 1/(s-a)
Complex Numbers Power of i
Every time a factor i⁴ occurs, it can be replaced by the factor 1, so that the power of i is reduced to one of four results: i⁰=1, i¹=i, i²=-1, i³=-i. i⁹ = (i⁴)² i = 1² i = i i²⁰ = (i⁴)⁵ = 1⁵ = 1 i³⁰ = (i⁴)⁷ i² = 1⁷ (-1) = -1 i¹⁵ = (i⁴)³ i³ = 1³ (-i) = -i
Complex Numbers
z = 3 + 5i is a complex number where 3 is the real part and 5 is the imaginary part. In general, z = a + bi, where a is the real part and b is the imaginary part.
Conjugate Complex Numbers
For a complex number a + i b, the complex number a - i b is called its conjugate. Conjugate complex numbers are identical except for the sign of the imaginary part. 🔑 Remember: The product of a complex number and its conjugate is always entirely real. (3+4i)(3-4i) = 9 + 16 = 25. (a+bi)(a-bi) = a² + b².
Euler's Formula
The series expansion of eˣ, cos x, and sin x are given. Replacing x by (it) yields: e^(it) = cos t + i sin t The real part R(e^(it)) = cos t and the imaginary part I(e^(it)) = sin t.
🔑 Definition — Euler's Formula: e^(iθ) = cos θ + i sin θ
The Laplace Transform of F(t) = sin at
L(sin at) = L(I(e^(iat))) = I[∫₀∞ e^(iat) e⁻ˢᵗ dt] = I[∫₀∞ e⁻⁽ˢ⁻ⁱᵃ⁾ᵗ dt] = I{1/(s-ia)} = I[(s+ia)/(s²+a²)] = I[s/(s²+a²) + i a/(s²+a²)] = a/(s²+a²)
🔑 Definition — Laplace Transform of sin at: L{sin at} = a/(s² + a²)
The Laplace Transform of F(t) = cos at
L(cos at) = L(R(e^(iat))) = R[1/(s-ia)] = R[(s+ia)/(s²+a²)] = R[s/(s²+a²) + i a/(s²+a²)] = s/(s²+a²)
🔑 Definition — Laplace Transform of cos at: L{cos at} = s/(s² + a²)
The Transform of F(t) = tⁿ where n is a positive integer
Using integration by parts and a reduction formula, we get: L(tⁿ) = n/s L(tⁿ⁻¹) Repeatedly applying this reduction formula gives: Iₙ = n/s * (n-1)/s * (n-2)/s * ... * 2/s * 1/s * I₀ Since I₀ = L{1} = 1/s, we get L(tⁿ) = n! / sⁿ⁺¹
🔑 Definition — Laplace Transform of tⁿ: L{tⁿ} = n! / sⁿ⁺¹ 📐 Formula: L{tⁿ} = n! / sⁿ⁺¹ → The transform of t raised to a positive integer power n is n factorial divided by s to the power of (n+1). 📌 Example: L{t³} = 3! / s⁴ = 6/s⁴
Laplace Transform of F(t) = sinh at and F(t) = cosh at
Starting from the exponential definitions: sinh at = ½(eᵃᵗ - e⁻ᵃᵗ) and cosh at = ½(eᵃᵗ + e⁻ᵃᵗ)
a) L{sinh at} = L[½(eᵃᵗ - e⁻ᵃᵗ)] = ½[L(eᵃᵗ) - L(e⁻ᵃᵗ)] = ½[1/(s-a) - 1/(s+a)] = ½[(s+a - (s-a))/(s²-a²)] = ½[2a/(s²-a²)] = a/(s²-a²)
b) L{cosh at} = L[½(eᵃᵗ + e⁻ᵃᵗ)] = ½[L(eᵃᵗ) + L(e⁻ᵃᵗ)] = ½[1/(s-a) + 1/(s+a)] = ½[(s+a + s-a)/(s²-a²)] = ½[2s/(s²-a²)] = s/(s²-a²)
🔑 Definition — Laplace Transform of sinh at: L{sinh at} = a/(s² - a²) 🔑 Definition — Laplace Transform of cosh at: L{cosh at} = s/(s² - a²)
Several Standard Results
L{a} = a/s ; L{eᵃᵗ} = 1/(s-a) ; L{tⁿ} = n!/sⁿ⁺¹ L{sin at} = a/(s²+a²) ; L{cos at} = s/(s²+a²) L{sinh at} = a/(s²-a²) ; L{cosh at} = s/(s²-a²)
These transforms can be combined by adding or subtracting as necessary, but they must not be multiplied together to form the transform of a product.
📌 Example a: Find L{2 sin 3t + cos 3t} L{2 sin 3t + cos 3t} = 2L(sin 3t) + L(cos 3t) = 2 * [3/(s²+9)] + [s/(s²+9)] = (6 + s)/(s²+9) = (s+6)/(s²+9)
📌 Example b: Find L{4e²ᵗ + 3 cosh 4t} L{4e²ᵗ + 3 cosh 4t} = 4L(e²ᵗ) + 3L(cosh 4t) = 4*[1/(s-2)] + 3*[s/(s²-16)] = 4/(s-2) + 3s/(s²-16)
⭐ Key Takeaways
(1) The Laplace transform is defined as an integral L{F(t)} = ∫₀∞ e⁻ˢᵗ F(t) dt, and its result is always a function of s. (2) The standard transforms for constants, exponentials, powers of t, sine, cosine, hyperbolic sine, and hyperbolic cosine form a critical "toolkit" for the course; these are derived and must be memorized. (3) Euler's formula e^(iθ) = cos θ + i sin θ is a powerful tool for deriving the transforms of sine and cosine by taking the imaginary or real parts, respectively, of the transform of e^(iat). (4) The product of a complex number with its conjugate always yields a real number (a²+b²), which is essential in simplifying the transforms of sine and cosine. (5) The transform is a linear operator, meaning L{af(t) + bg(t)} = aL{f(t)} + bL{g(t)}. Transforms can be added or subtracted in linear combinations but must NOT be multiplied to find the transform of a product.
🧠 Quick Revision Questions
- State the formal definition of the Laplace transform L{F(t)}.
- What is the Laplace transform of eᵃᵗ? Under what condition on s does this transform hold?
- Derive the Laplace transform of sin(at) using Euler's formula.
- What is the reduction formula used to find L{tⁿ}, and what is the final result?
- State the standard Laplace transforms for both sinh(at) and cosh(at). How are they different from the transforms of sin(at) and cos(at)?
📘 Lecture 45 — Theorems
📖 Overview: This lecture covers three fundamental theorems of Laplace transforms: the First Shift Theorem, Multiplication by t, and Division by t. It then introduces Inverse Laplace transforms, explaining how to recover the original time function from a given transform using partial fractions and a table of standard forms, and concludes by applying these techniques to solve ordinary differential equations.
🗂️ Topics Covered
The lecture begins with the First Shift Theorem, which shows how multiplying a function by e^(-at) shifts the Laplace variable s to s+a. It then covers the theorem for multiplying F(t) by t, which involves differentiating the transform, and the theorem for dividing F(t) by t, which involves integrating the transform. The second half of the lecture thoroughly discusses the inverse Laplace transform process, including rules for partial fraction decomposition, a table of standard inverse transforms, and formulas for transforms of derivatives. The lecture culminates with the step-by-step solution of differential equations using the Laplace transform method, including a detailed example with a second-order ODE.
📝 Lecture Summary
Theorem 1 The First Shift Theorem
The First Shift theorem states that if L{F(t)} = f(s), then L{e^(-at)F(t)} = f(s+a). The transform of the exponentially damped function is the same as the transform of the original function with ‘s’ replaced by (s+a).
🔑 Definition — First Shift Theorem: L{e^(-at)F(t)} = f(s+a)
📐 Formula: L{e^(-at)F(t)} = f(s+a) → The Laplace transform of an exponential multiple of a function is the transform of the function itself with 's' shifted by 'a'.
📌 Example: Given L{sin 2t} = 2/(s² + 4). Then L{e^(-3t) sin 2t} = 2/((s+3)² + 4) = 2/(s² + 6s + 13). Another example: L{t²} = 2/s³, so L{t² e^(4t)} = 2/(s – 4)³.
Theorem 2 Multiplying by t
If L{F(t)} = f(s), then L{t F(t)} = -d/ds {f(s)}. This theorem extends the range of functions we can deal with, and in general, L{tⁿF(t)} = (-1)ⁿ dⁿ/dsⁿ {f(s)}.
🔑 Definition — Multiplying by t Theorem: L{tF(t)} = -d/ds {f(s)}
📐 Formula: L{tF(t)} = -d/ds f(s) → Multiplying a time function by 't' corresponds to taking the negative derivative of its Laplace transform with respect to 's'.
📌 Example: L{sin 2t} = 2/(s² + 4). Then L{t sin 2t} = -d/ds [2/(s² + 4)] = 4s/(s² + 4)². Another example: L{t cos 3t} = -d/ds [s/(s² - 9)] = (s² + 9)/(s² - 9)². This can be taken a step further: L{t² cos 3t} = -d/ds [(s² + 9)/(s² - 9)²] = 2s(s² + 27)/(s² - 9)³.
💡 Why this matters: This theorem allows us to find transforms of functions multiplied by tⁿ, which is crucial for solving differential equations with polynomial coefficients.
Theorem 3 Dividing by t
If L{F(t)} = f(s), then L{F(t)/t} = ∫ₛ^∞ f(s) ds, provided the limit of F(t)/t as t approaches 0 exists.
🔑 Definition — Dividing by t Theorem: L{F(t)/t} = ∫ₛ^∞ f(u) du
📐 Formula: L{F(t)/t} = ∫ₛ^∞ f(s) ds → Dividing a time function by 't' corresponds to integrating its Laplace transform from 's' to infinity.
📌 Example: Determine L{sin(at)/t}. As L{sin at} = a/(s² + a²), then L{sin at/t} = ∫ₛ^∞ a/(s² + a²) ds = [tan⁻¹(s/a)]ₛ^∞ = π/2 - tan⁻¹(s/a) = tan⁻¹(a/s).
📌 Example: Determine L{(1 - cos 2t)/t}. As L{1 - cos 2t} = 1/s - s/(s² + 4), then L{(1 - cos 2t)/t} = ∫ₛ^∞ (1/s - s/(s² + 4)) ds = [ln s - ½ ln(s² + 4)]ₛ^∞ = [½ ln(s²/(s² + 4))]ₛ^∞ = 0 - ½ ln(s²/(s² + 4)) = ½ ln((s² + 4)/s²).
Standard Forms
A table of standard Laplace transforms is essential for both forward and inverse transforms.
| F(t) | L{F(t)} = f(s) |
|---|---|
| a | a/s |
| e^(at) | 1/(s – a) |
| sin at | a/(s² + a²) |
| cos at | s/(s² + a²) |
| sinh at | a/(s² – a²) |
| cosh at | s/(s² – a²) |
| tⁿ (n a positive integer) | n!/s^(n+1) |
Inverse Transforms
Inverse Laplace transforms are the reverse process: given a Laplace transform, we find the function of t to which it belongs. The symbol L⁻¹ indicates the inverse transform.
🔑 Definition — Inverse Laplace Transform: Given a function f(s), the inverse transform L⁻¹{f(s)} finds the original function F(t) such that L{F(t)} = f(s).
📌 Example: L⁻¹{1/(s – 2)} = e^(2t) L⁻¹{s/(s² + 25)} = cos 5t L⁻¹{4/s} = 4 L⁻¹{12/(s² – 9)} = 4 sinh 3t
Often, a transform must be broken into simpler parts using partial fractions. For instance, to find L⁻¹{3s/(s² – s – 6)}, we first decompose 3s/(s² – s – 6) into 1/(s + 2) + 2/(s – 3). Then, L⁻¹{3s/(s² – s – 6)} = L⁻¹{1/(s + 2)} + L⁻¹{2/(s – 3)} = e^(-2t) + 2e^(3t).
Rules of Partial Fractions
- The numerator must be of a lower degree than the denominator. If not, divide out first.
- Factorise the denominator into its prime factors.
- A linear factor (s + a) gives a partial fraction of the form A/(s + a).
- A repeated factor (s + a)² gives A/(s + a) + B/(s + a)².
- Similarly, (s + a)³ gives A/(s + a) + B/(s + a)² + C/(s + a)³.
- A quadratic factor (s² + ps + q) gives (Ps + Q)/(s² + ps + q).
- A repeated quadratic factor (s² + ps + q)² gives (Ps + Q)/(s² + ps + q) + (Rs + T)/(s² + ps + q)².
📌 Example: Determine L⁻¹{(5s + 1)/(s² – s – 12)}. a) The numerator (1st degree) is of lower degree than the denominator (2nd degree). b) Factorise: (s – 4)(s + 3). c) Write identity: (5s + 1)/(s² – s – 12) = A/(s – 4) + B/(s + 3). d) Multiply by denominator: 5s + 1 = A(s + 3) + B(s – 4). e) Substitute s = 4: 21 = A(7) → A = 3. f) Substitute s = -3: -14 = B(-7) → B = 2. g) Therefore, L⁻¹{(5s + 1)/(s² – s – 12)} = L⁻¹{3/(s – 4) + 2/(s + 3)} = 3e^(4t) + 2e^(-3t).
📌 Example: Determine L⁻¹{(9s – 8)/(s² – 2s)}. a) Decompose: (9s – 8)/[s(s – 2)] = A/s + B/(s – 2). b) Multiply by s(s – 2): 9s – 8 = A(s – 2) + B(s). c) Put s = 0: -8 = A(-2) → A = 4. d) Put s = 2: 10 = B(2) → B = 5. e) Therefore, L⁻¹{4/s + 5/(s – 2)} = 4 + 5e^(2t).
Table of Inverse Transforms
A standard table is used for finding inverse transforms. The table shows common f(s) and their corresponding F(t).
| f(s) | F(t) |
|---|---|
| a/s | a |
| 1/(s + a) | e^(-at) |
| n!/s^(n+1) | tⁿ (n a positive integer) |
| 1/sⁿ | t^(n-1)/(n-1)! (n a positive integer) |
| a/(s² + a²) | sin at |
| s/(s² + a²) | cos at |
| a/(s² – a²) | sinh at |
| s/(s² – a²) | cosh at |
Transforms of Derivatives
The Laplace transforms of derivatives are crucial for solving differential equations. Let F'(t), F''(t), etc., denote derivatives.
🔑 Transform of First Derivative: L{F'(t)} = sL{F(t)} – F(0) 🔑 Transform of Second Derivative: L{F''(t)} = s²L{F(t)} – sF(0) – F'(0) 🔑 Transform of Third Derivative: L{F'''(t)} = s³L{F(t)} – s²F(0) – sF'(0) – F''(0) 🔑 Transform of Fourth Derivative: L{Fᶤᵛ(t)} = s⁴L{F(t)} – s³F(0) – s²F'(0) – sF''(0) – F'''(0)
Differential Equation And Its Solution
The process of solving a differential equation using Laplace transforms involves four distinct stages: (a) Rewrite the equation in terms of Laplace transforms. (b) Insert the given initial conditions. (c) Rearrange the equation algebraically to give the transform of the solution. (d) Determine the inverse transform to obtain the particular solution.
📌 Example: Solve the first-order ODE dx/dt – 2x = 4, given x(0) = 1.
- Take Laplace transform: L{dx/dt} – 2L{x} = L{4} → sL{x(t)} – x(0) – 2L{x(t)} = 4/s.
- Substitute initial condition x(0)=1: (s – 2)L{x(t)} – 1 = 4/s → (s – 2)L{x(t)} = (4 + s)/s.
- Solve for L{x(t)}: L{x(t)} = (s + 4)/[s(s – 2)].
- Use partial fractions to invert: (s+4)/[s(s-2)] = A/s + B/(s-2). Solving gives A = -2, B = 3.
- Therefore, x(t) = L⁻¹{-2/s + 3/(s-2)} = -2 + 3e^(2t).
📌 Example: Solve the second-order ODE d²x/dt² – 3 dx/dt + 2x = 2e^(3t), given x(0) = 5 and x'(0) = 7.
- Rewrite: L{x''} – 3L{x'} + 2L{x} = 2L{e^(3t)}.
- Apply derivative formulas: [s²L{x} – s x(0) – x'(0)] – 3[sL{x} – x(0)] + 2L{x} = 2/(s – 3).
- Substitute initial conditions x(0)=5, x'(0)=7: (s²L{x} – 5s – 7) – 3sL{x} + 15 + 2L{x} = 2/(s – 3).
- Rearrange: (s² – 3s + 2)L{x} = 2/(s – 3) + 5s – 8 = (2 + (5s – 8)(s – 3))/(s – 3) = (5s² – 23s + 24)/(s – 3).
- So L{x} = (5s² – 23s + 24)/[(s – 1)(s – 2)(s – 3)].
- Partial fraction decomposition gives L{x} = 3/(s – 1) + 2/(s – 2) + 0/(s – 3).
- Take inverse Laplace: x(t) = 3e^t + 2e^(2t).
⭐ Key Takeaways
The three core theorems—First Shift, Multiplying by t, and Dividing by t—are powerful tools for manipulating Laplace transforms. Mastery of partial fraction decomposition is absolutely critical for finding inverse Laplace transforms of rational functions. The transforms of derivatives directly incorporate initial conditions, which makes the Laplace transform method exceptionally efficient for solving initial value problems of ordinary differential equations, reducing them to algebraic manipulations. The step-by-step process of transform, substitute, rearrange, and invert is a systematic and powerful technique for solving ODEs.
🧠 Quick Revision Questions
- State the First Shift Theorem. If L{F(t)} = 2/(s – 3), what is L{e^(-5t)F(t)}?
- Explain the theorem for L{tF(t)}. Use it to find L{t cos 2t}.
- What is the formula for L{F(t)/t}? Use it to find L{sin(3t)/t}.
- Describe the steps to find the inverse Laplace transform of 1/[s(s+1)].
- Write the Laplace transform formula for the second derivative L{F''(t)} and outline how it is used to solve a second-order initial value problem.