Scribd
Upload
4 views2 pages

Analyze Gearbox Vibration Audio Data

This Python script processes an audio file by loading it, converting it to mono if necessary, and normalizing the signal. It then plots the first 0.1 seconds of the waveform and computes the single-sided magnitude spectrum up to one-third of the sampling frequency, visualizing it as well. Finally, it identifies and prints the frequency with the highest magnitude, excluding DC components.

Uploaded by

sonukumarjha150
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
4 views2 pages

Analyze Gearbox Vibration Audio Data

This Python script processes an audio file by loading it, converting it to mono if necessary, and normalizing the signal. It then plots the first 0.1 seconds of the waveform and computes the single-sided magnitude spectrum up to one-third of the sampling frequency, visualizing it as well. Finally, it identifies and prints the frequency with the highest magnitude, excluding DC components.

Uploaded by

sonukumarjha150
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Python

import numpy as np

import [Link] as plt

import [Link] as wavfile

# Load the audio and convert to mono if stereo.

fs, audio_data = [Link]('gearbox_vibration.wav')

if audio_data.ndim > 1:

audio_data = [Link](audio_data, axis=1)

# Normalize the signal to the range [-1, 1].

audio_data = audio_data / [Link]([Link](audio_data))

# Plot the first 0.1 s of the waveform.

time_axis = [Link](0, len(audio_data)) / fs

end_index = int(0.1 * fs)

[Link]()

[Link](time_axis[:end_index], audio_data[:end_index])

[Link]('Time (s)')

[Link]('Amplitude')

[Link]('First 0.1s of the Waveform')

[Link]()

[Link]()

# Compute and plot the single-sided magnitude spectrum up to Fs/3 Hz.

n_fft = len(audio_data)
fft_result = [Link](audio_data, n=n_fft)

fft_magnitude = [Link](fft_result) / n_fft

frequencies = [Link](n_fft, 1/fs)

single_sided_magnitude = 2 * fft_magnitude[:n_fft//2]

single_sided_frequencies = frequencies[:n_fft//2]

max_freq_index = [Link](single_sided_frequencies <= fs/3)[0][-1]

[Link]()

[Link](single_sided_frequencies[:max_freq_index],
single_sided_magnitude[:max_freq_index])

[Link]('Frequency (Hz)')

[Link]('Magnitude')

[Link]('Single-Sided Magnitude Spectrum up to Fs/3 Hz')

[Link]()

[Link]()

# Identify and print the frequency with the highest magnitude (excluding DC).

start_index = [Link](single_sided_frequencies > 0)[0][0]

peak_frequency_index = [Link](single_sided_magnitude[start_index:max_freq_index])

peak_frequency = single_sided_frequencies[start_index + peak_frequency_index]

print(f'The frequency with the highest magnitude (excluding DC) is: {peak_frequency:.2f} Hz')

You might also like