Intelligent Monitoring of Rotating Machinery Condition Through the Fusion of Audio, Acoustic Emission, Vibration, and Current Signals

This project addresses the critical need to enhance Predictive Maintenance (PdM) strategies for rotating machinery, including gearboxes, bearings, and shafts, where unexpected failures cause substantial losses. The core problem is the difficulty in inspecting these enclosed components without halting operations. The proposed solution focuses on Intelligent Condition Monitoring (ICM) through the fusion of multiple monitoring signals: acoustic, vibration, acoustic emission, and current. The methodology is structured in five phases: an exhaustive literature review; experimental data acquisition under controlled conditions using vibration test benches and software like LabVIEW and Matlab; extraction of Condition Parameters (CPs) across time, frequency, and time-frequency domains; selection and reduction of the most relevant CPs; and finally, the design and evaluation of diagnostic systems based on data fusion at the source, data, and classifier levels, employing machine learning to increase diagnostic accuracy and system reliability.<br/><br/><b>Goal</b>: <br/>The main objective is to develop an intelligent monitoring system for the condition of rotating machinery (shafts, bearings, gearboxes) by fusing audio, acoustic emission, vibration, and current signals, utilizing data mining and machine learning techniques to achieve more accurate fault diagnosis.<br/><br/><b>Research lines</b>: <br/>Control engineering and automation technologies