Fourier series provide a powerful way to represent periodic functions as infinite sums of sinusoids. Developed by Jean-Baptiste Joseph Fourier, these series are fundamental to understanding signal processing, electrical systems, wave phenomena, and numerous engineering applications.
The Fourier series represents a breakthrough in mathematical analysis that allows any periodic function to be decomposed into a sum of simple oscillating functions (sines and cosines). This concept revolutionized our understanding of complex waveforms and laid the foundation for modern signal processing and analysis.
In the early 19th century, while studying heat transfer, Jean-Baptiste Joseph Fourier discovered that complicated periodic functions could be expressed as infinite series of trigonometric functions. Initially met with skepticism by mathematicians including Lagrange and Laplace, Fourier's work eventually gained acceptance and has since become fundamental in various fields of science and engineering.
For electrical engineering students, Fourier series are particularly important because:
A function \(f(t)\) is periodic with period \(T\) if for all \(t\):
The fundamental frequency \(\omega_0\) of the function is related to the period by:
The function \(f(t) = \sin(3t)\) has period \(T = \frac{2\pi}{3}\), because:
\(f(t + \frac{2\pi}{3}) = \sin(3t + 2\pi) = \sin(3t) = f(t)\)
Its fundamental frequency is \(\omega_0 = \frac{2\pi}{T} = 3\) rad/s
The Fourier series states that any periodic function \(f(t)\) with period \(T\) can be represented as an infinite sum of sines and cosines:
This is the trigonometric form of the Fourier series. The coefficients \(a_0\), \(a_n\), and \(b_n\) determine the contribution of each frequency component.
The Fourier coefficients are calculated using the following integrals:
Where \(t_0\) can be any starting point for the interval of integration, as long as the interval spans exactly one period \(T\). Common choices are \(t_0 = -\frac{T}{2}\) or \(t_0 = 0\).
For a square wave with period \(T = 2\pi\):
\(f(t) = \begin{cases} 1, & 0 < t < \pi \\ -1, & -\pi < t < 0 \end{cases}\)
The Fourier coefficients are:
\(a_0 = 0\) (no DC component)
\(a_n = 0\) (for all \(n\))
\(b_n = \begin{cases} \frac{4}{n\pi}, & \text{for odd } n \\ 0, & \text{for even } n \end{cases}\)
Therefore, the Fourier series is:
\(f(t) = \frac{4}{\pi}\sum_{k=0}^{\infty} \frac{\sin((2k+1)t)}{2k+1}\)
The Fourier coefficients have important physical interpretations:
The standard trigonometric form of the Fourier series is:
This form clearly separates the even (cosine) and odd (sine) components of the function.
The trigonometric form can be rewritten in the amplitude-phase format:
Where:
This form is particularly useful in engineering applications as it directly shows the amplitude and phase of each harmonic component.
Using Euler's formula, the Fourier series can be expressed in complex exponential form:
Where the complex Fourier coefficients \(c_n\) are given by:
The relationship between the complex coefficients and the trigonometric coefficients is:
For real-valued functions, we have the property \(c_{-n} = c_n^*\) (complex conjugate).
For the square wave defined earlier, the complex coefficients are:
\(c_0 = 0\)
\(c_n = \begin{cases} \frac{2i}{n\pi}, & \text{for odd } n \\ 0, & \text{for even } n \end{cases}\)
\(c_{-n} = -c_n\) (for all \(n\))
Therefore, the complex exponential form is:
\(f(t) = \frac{2i}{\pi}\sum_{k=0}^{\infty} \frac{e^{i(2k+1)t} - e^{-i(2k+1)t}}{2k+1}\)
Which simplifies to:
\(f(t) = \frac{4}{\pi}\sum_{k=0}^{\infty} \frac{\sin((2k+1)t)}{2k+1}\)
A Fourier series converges to the function at points of continuity if the function satisfies the Dirichlet conditions:
At points of discontinuity, the Fourier series converges to the average of the left and right limits:
This phenomenon is known as the Gibbs phenomenon, which manifests as overshoot and ringing near discontinuities even as more terms are added to the series.
A Fourier series converges uniformly if the function is continuous and has a piecewise continuous derivative. Uniform convergence ensures that the series approximates the function equally well at all points.
The rate of convergence of a Fourier series depends on the smoothness of the function:
Consider these two functions:
1. Square wave: \(f(t) = \text{sgn}(\sin(t))\)
2. Triangle wave: \(g(t) = \frac{2}{\pi}|\arcsin(\sin(t))|\)
The square wave has discontinuities, so its Fourier coefficients decay as \(1/n\), making the series converge slowly.
The triangle wave is continuous but has discontinuous derivatives, so its coefficients decay as \(1/n^2\), and the series converges more rapidly.
This means we need fewer terms to achieve a good approximation for the triangle wave compared to the square wave.
The Fourier series decomposes a periodic function into its frequency components, allowing us to analyze its spectrum. The spectrum can be visualized in several ways:
The amplitude spectrum shows the magnitude of each frequency component.
For the complex form, the amplitude spectrum is given by \(|c_n|\).
For a square wave, the amplitude spectrum is:
\(A_n = \begin{cases} \frac{4}{n\pi}, & \text{for odd } n \\ 0, & \text{for even } n \end{cases}\)
This shows that a square wave consists only of odd harmonics, with amplitudes inversely proportional to the harmonic number.
The phase spectrum shows the phase angle of each frequency component.
For the complex form, the phase spectrum is given by \(\arg(c_n)\).
For a square wave centered at the origin, the phase spectrum is:
\(\phi_n = \begin{cases} -\frac{\pi}{2}, & \text{for odd } n \\ \text{undefined}, & \text{for even } n \end{cases}\)
The phase is -90° for all odd harmonics, indicating that the wave consists of sine terms rather than cosine terms.
The power spectrum shows the power distribution across frequencies. For a periodic function, the power in the \(n\)th harmonic is proportional to \(A_n^2\).
In terms of complex Fourier coefficients, the power in the \(n\)th harmonic is:
The total power in the signal is the sum of the powers in all harmonics:
For a square wave with amplitude 1, the power spectrum is:
\(P_n = \begin{cases} \frac{16}{n^2\pi^2}, & \text{for odd } n \\ 0, & \text{for even } n \end{cases}\)
The total power is \(P_{\text{total}} = 1\), which equals the mean square value of the square wave.
The first harmonic (n=1) contains \(\frac{16}{\pi^2} \approx 81.06\%\) of the total power.
There are several ways to visualize the spectrum of a periodic signal:
These visualizations are essential tools for understanding the frequency content of signals in electrical engineering applications.
Parseval's theorem establishes a relationship between the energy of a signal in the time domain and its energy in the frequency domain. For a periodic function with period \(T\), the theorem states:
Using the trigonometric form, this becomes:
For a square wave with amplitude 1 and period \(2\pi\):
The mean square value is \(\frac{1}{2\pi} \int_{0}^{2\pi} f^2(t) \, dt = 1\)
Using Parseval's theorem:
\(1 = \sum_{n=1,3,5,...}^{\infty} \frac{8}{\pi^2 n^2}\)
This confirms the known result that \(\sum_{n=1,3,5,...}^{\infty} \frac{1}{n^2} = \frac{\pi^2}{8}\)
Parseval's theorem is a statement of energy conservation: the energy in the time domain equals the energy in the frequency domain. This principle is fundamental to signal processing and system analysis.
Parseval's theorem has numerous practical applications in engineering:
If we approximate a function \(f(t)\) using the first \(N\) terms of its Fourier series, we get:
\(f_N(t) = c_0 + \sum_{n=1}^{N} (c_n e^{in\omega_0 t} + c_{-n} e^{-in\omega_0 t})\)
The mean square error of this approximation is:
\(\frac{1}{T} \int_{0}^{T} |f(t) - f_N(t)|^2 \, dt = 2 \sum_{n=N+1}^{\infty} |c_n|^2\)
This allows us to determine how many terms are needed to achieve a desired accuracy.
In power systems, non-sinusoidal voltages and currents can be analyzed using Fourier series to determine:
Consider a non-linear load drawing current:
The THD can be calculated as:
This high THD indicates significant harmonic distortion that may require filtering or other power quality improvements.
Fourier series is fundamental to the design and analysis of filters, helping engineers:
If a square wave with Fourier series:
passes through a low-pass filter with cutoff frequency \(\omega_c = 3\omega_0\), the output will be approximately:
This shows how the filter removes higher harmonics while preserving the lower frequency components.
Fourier series is essential in the analysis and design of communication systems:
A message signal \(m(t) = \cos(\omega_m t)\) modulates a carrier \(c(t) = \cos(\omega_c t)\) to produce:
Using trigonometric identities, this can be expressed as:
The Fourier spectrum shows the carrier at \(\omega_c\) and two sidebands at \(\omega_c \pm \omega_m\), illustrating the bandwidth requirements.
In control systems, Fourier series helps analyze:
For a linear system with transfer function \(H(j\omega)\), the steady-state response to a periodic input \(x(t) = \sum_{n=-\infty}^{\infty} c_n e^{jn\omega_0 t}\) is:
This allows engineers to predict how a system will respond to complex periodic inputs by analyzing each frequency component separately.
When solving problems involving Fourier series, follow these steps:
Exploiting symmetry can greatly simplify the calculation of Fourier coefficients:
For an even function \(f(-t) = f(t)\):
Example: \(f(t) = \cos(t)\), \(f(t) = |t|\)
For an odd function \(f(-t) = -f(t)\):
Example: \(f(t) = \sin(t)\), \(f(t) = t\) (over \([-\pi, \pi]\))
For functions with half-wave symmetry \(f(t + T/2) = -f(t)\):
Example: \(f(t) = \sin(t)\), square wave
Combining even/odd symmetry with half-wave symmetry further simplifies calculations
Avoid these common mistakes when working with Fourier series:
Several computational tools can assist with Fourier series calculations:
fft
and ifft
Find the Fourier series for a square wave with period \(T = 2\):
First, we identify that \(\omega_0 = 2\pi/T = \pi\).
Since the function is odd, we know \(a_0 = 0\) and \(a_n = 0\) for all \(n\).
For the sine coefficients:
After integration:
Therefore, the Fourier series is:
Find the Fourier series for a triangle wave with period \(T = 2\):
First, we identify that \(\omega_0 = 2\pi/T = \pi\).
The function has even symmetry around the point \(t=0\), so \(b_n = 0\) for all \(n\).
For the DC component:
For the cosine coefficients:
After integration:
Therefore, the Fourier series is:
Find the Fourier series for a full-wave rectified sine wave:
The period is \(T = \pi/\omega_0\) (half the period of \(\sin(\omega_0 t)\)).
The function has even symmetry, so \(b_n = 0\) for all \(n\).
For the DC component:
For the cosine coefficients:
After integration:
Therefore, the Fourier series is:
Test your understanding by working through these exercises:
Find the Fourier series for the periodic function with period \(T = 2\) defined by:
Find the Fourier series for the sawtooth wave:
A square wave with amplitude 1 and period \(2\pi\) is passed through a low-pass filter that keeps only the first three non-zero harmonics. What is the expression for the output signal?
Using Parseval's theorem, calculate the power in a triangle wave with amplitude 1 and determine what percentage of the power is contained in the first harmonic.
A periodic signal is given by \(f(t) = 2 + 3\cos(100\pi t) + 4\sin(100\pi t) + \cos(300\pi t) + 2\sin(500\pi t)\). Find its fundamental frequency, period, average value, RMS value, and power.
Fourier series provide a powerful mathematical framework for analyzing periodic signals and systems. By decomposing complex waveforms into simple sinusoidal components, they enable engineers to understand frequency content, design filters, analyze power systems, and develop communication systems.
For electrical engineering students, mastering Fourier series is essential for a deeper understanding of signals, systems, and their behavior in both time and frequency domains. The concepts introduced here form the foundation for more advanced topics such as Fourier transforms, Laplace transforms, and spectral analysis.
By applying the principles and techniques covered in this chapter, you'll be equipped to analyze complex periodic phenomena in various electrical engineering applications and develop innovative solutions to engineering challenges.