Low-Cost Prototype for Bearing Failure Detection Using Tiny ML Through Vibration Analysis

Published: 17 September 2024| Version 1 | DOI: 10.17632/685hm7n8nb.1
Contributor:
Andres Cotrino

Description

Authors: Andres Felipe Cotrino Herrera, Jesús Alfonso López Sotelo, Juan Carlos Blandón Andrade, Alonso Toro Lazo The following files are for a low-cost, open-source device designed to facilitate the learning of technologies like artificial intelligence in embedded systems through vibration analysis. It also aims to enhance students' skills by introducing industrial challenges into the classroom via a scaled-down prototype. This study analyzes the vibrations generated by bearings to classify, using Artificial Intelligence (AI), whether they are defective. The device integrates electronic, mechanical, and software components, leveraging online technologies and platforms like Arduino to support hands-on learning. The document provides detailed instructions on the components used, circuit connections, step-by-step construction, and implementation, allowing replication of the prototype. This device fosters the development of STEM skills, promotes the application of AI and TinyML in real-world contexts, and enriches educational programs by encouraging interdisciplinary learning. Detailed information on the components used, connection circuits, step-by-step construction, and implementation of the device is provided later, enabling anyone interested to replicate this prototype. This device also supports the development of STEM skills and promotes the application of AI and TinyML in practical settings, enriching educational programs and fostering interdisciplinary learning.

Files

Institutions

Universidad Autonoma de Occidente, Universidad Catolica de Pereira

Categories

Artificial Intelligence, Teaching, Machine Learning, Vibration Analysis

Licence