Fault Diagnosis in Gears Based on Vibration Signal Analysis Using Artificial Intelligence and/or Statistical Techniques

This project addresses the critical need to reduce operational costs in industry by focusing on optimizing maintenance through proactive strategies, particularly Condition-Based Maintenance (CBM). CBM aims to avoid unnecessary maintenance tasks and reduce uncertainty by predicting imminent failures, which is superior to reactive diagnosis. The main objective is to develop a system for diagnosing gear failures by analyzing their vibration signals. The methodology begins with an exhaustive literature review on the state-of-the-art in failure diagnosis using AI and statistics. Subsequently, a robust database of vibration signals from a spur gear train under various failure conditions will be constructed, establishing rigorous data acquisition protocols. Acquired signals will be processed, transforming data from the time domain to the frequency domain to extract and select the most relevant condition parameters. Finally, a library of artificial intelligence and statistical algorithms will be developed and evaluated to efficiently classify and diagnose failures, culminating in the presentation of this algorithm library for practical application.<br/><br/><b>Goal</b>: <br/>To diagnose gear failures based on vibration signal analysis using artificial intelligence and/or statistical techniques.<br/><br/><b>Research lines</b>: <br/>Control engineering and automation technologies