ALESSANDRO PAOLO DAGA
Assistant Professor with time contract
Tracking the “signature” of a machine (the characteristic features of the machine extracted from the vibration signal, usually from accelerometers) evolving over time, it is possible to identify the main causes of vibration and preventively recognize damage and wear of the components, thus performing an accurate diagnosis.
Two steps are fundamental:
• Signal processing is exploited to highlight and extract damage-characteristic features.
• Machine learning is employed to automatically infer anomalies in the vibration response from the extracted features. Such anomaly or novelty, in fact, can be put in relation to damage when confounding influences (i.e. different operational or environmental conditions) can be excluded or compensated.
The aim is to create a reliable diagnostic system to be integrated into machine maintenance regimes so as to foster safety while, at the same time, saving on costs
|Scientific branch||ING-IND/13 - MECCANICA APPLICATA ALLE MACCHINE
(Area 0009 - Ingegneria industriale e dell'informazione)
Scopus Author ID: 57193438974
Machine Condition Monitoring and Diagnostics
|Skills and keywords||
ERC sectorsPE1_19 - Control theory and optimisation PE6_11 - Machine learning, statistical data processing and applications using signal processing (e.g. speech, image, video) PE8_7 - Mechanical and manufacturing engineering (shaping, mounting, joining, separation) PE1_10 - ODE and dynamical systems PE1_18 - Scientific computing and data processing PE7_7 - Signal processing PE1_14 - Statistics
SDGGoal 7: Affordable and clean energy Goal 9: Industry, Innovation, and Infrastructure
KeywordsBearings (machine parts) Digital signal processing Machine condition monitoring Machine diagnostics Machine learning Statistical learning Vibration monitoring
|Scientific responsibilities and other assignments||
Awards and Honors