I have some questions regarding interharmonics. What I'm going to do is first ask just a few, and then as people answer them I would expand this post or create a new question.
Harmonics are sinusoids that have a frequency which is an integer multiple of the fundamental frequency of the original signal \$x(t)\$ they represent. Interharmonics (or inter-harmonics) are defined as sinusoids that have a frequency which isn't an integer multiple of the fundamental frequency of the signal \$x(t)\$. First question: in that definition of interharmonics, is it assumed the signal \$x(t)\$ is periodic, or not?¹
I mean, yes, we use Fourier series usually for periodic signal, but I haven't read a single text book on math, circuit analysis, electronics, or signals and systems where they talk about interharmonics. Fourier's theorem never talks about interharmonics. The only place I've seen a brief discussion on interharmonics is in textbooks about power quality and harmonics. So this makes me wonder whether interharmonics even make sense, or that's just a term invented without any mathematical proof. So, before I ask further questions, I'd like to know the answer to the first question above.
I have another question. As you know, there're various ways to represent a Fourier series. One is the trigonometric form, other is the amplitude-phase form, and the other is the complex exponential form. Writing the amplitudes (i.e. the maximum values or peak values) of the harmonics in terms of the RMS values, the amplitude-phase form is:
\$x(t) = X_0 + \sqrt{2} \displaystyle \sum_{n=1}^\infty X_{\text{rms,} n} \cos{(2 \pi n f_0 t + \theta_n)} \tag*{}\$
My second question is if when a signal has an interharmonic of frequency \$m f_0\$, where \$m\$ is a non-integer positive number, do we sum it to the previous expression as a new sinusoid \$\sqrt{2} X_{\text{rms,} m} \cos{(2 \pi m f_0 + \theta_m)}\$? If not, then how does the interharmonic analitically contribute to the signal \$x(t)\$?
Note ¹: Fourier series can be used to represent a periodic signal with an expression valid for all time \$t\$, or to represent a non-periodic signal in a time interval \$\Delta t\$.
Best Answer
Let’s say you have an interharmonic of 1.5. If you scale your assumed fundamental frequency by 0.5 then you will have two integer harmonic sinusoids contributing to the signal - a 2nd harmonic and a 3rd harmonic. Equivalent, of course, to the original signal - just an integer way to look at it.
First case: Fundamental = f1
$$ x(t) = \cos{(1*2 \pi f_1 t + \theta_n)} + \cos{(1.5*2 \pi f_1t + \theta_m )} $$
Second case: Fundamental = f2 = 0.5f1
$$ x(t) = \cos{(2*2 \pi f_2 t + \theta_n)} + \cos{(3*2 \pi f_2t + \theta_m)} $$
These are equivalent signals (and periodic), I just chose to assume a different fundamental frequency for my analysis so i would have integer harmonics.