The concept of predictive maintenance, whose application became every day more and more diffused was born some years ago. The fundamental idea on which the predictive maintenance is based on is the monitoring of specific parameters that can supply useful information on the system state of health. In the presented application, vibrational levels represent one of these parameters and relative continuous monitoring is proposed. As a drawback of this approach, the availability of monitoring devices and their correct installation is needed, even if many times not availablr for cost or installation reasons. To avoid such limitations, the present work present an Artificial Neaural Network based approach for the management of “virtual” sensors whose data are derived from a limited set of sub-data. The presented application will show interesting results obtained with reference to a traction converter system as an example of the proposed technique.
ANN based Approach to the Structural Health Monitoring of a Traction Equipment / Viscardi, Massimo; P., Napolitano. - 40:(2014), pp. 189-198. (Intervento presentato al convegno 13th International Conference on Circuits, Systems, Electronics, Control & Signal Processing (CSECS '14) tenutosi a Lisbon, Portugal nel October 30 - November 1, 2014).
ANN based Approach to the Structural Health Monitoring of a Traction Equipment
VISCARDI, MASSIMO;
2014
Abstract
The concept of predictive maintenance, whose application became every day more and more diffused was born some years ago. The fundamental idea on which the predictive maintenance is based on is the monitoring of specific parameters that can supply useful information on the system state of health. In the presented application, vibrational levels represent one of these parameters and relative continuous monitoring is proposed. As a drawback of this approach, the availability of monitoring devices and their correct installation is needed, even if many times not availablr for cost or installation reasons. To avoid such limitations, the present work present an Artificial Neaural Network based approach for the management of “virtual” sensors whose data are derived from a limited set of sub-data. The presented application will show interesting results obtained with reference to a traction converter system as an example of the proposed technique.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.