In this paper the use of features is described in order to individuate the best set-up conditions of a mechatronic system. In particular, features are obtained from data of quantities measured in different positions of the kinematic linkage, the operation of the system is based on. A kinematic and dynamic model of the device is realized and used in order to obtain the features of interest also in positions different from the ones the measurements are carried out throughout the kinematic link. This procedure allows us to merge information from both internal and external sensors, in order to identify the best features for identification of working conditions, taking into account their resolution, selectivity, easiness of calculation, and data processing time and load. The most efficacious features are identified with reference to the time and frequency domain and the most suitable position and quantity the information is originated from. This methodology step is also a preliminary action toward the use of neural networks for smart identification of specific statuses of interest of a mechatronic system, during both the set-up phase and the functioning of the system for its condition monitoring.

Integration of model and sensor data for smart condition monitoring in mechatronic devices

Gaspari A.;
2019

Abstract

In this paper the use of features is described in order to individuate the best set-up conditions of a mechatronic system. In particular, features are obtained from data of quantities measured in different positions of the kinematic linkage, the operation of the system is based on. A kinematic and dynamic model of the device is realized and used in order to obtain the features of interest also in positions different from the ones the measurements are carried out throughout the kinematic link. This procedure allows us to merge information from both internal and external sensors, in order to identify the best features for identification of working conditions, taking into account their resolution, selectivity, easiness of calculation, and data processing time and load. The most efficacious features are identified with reference to the time and frequency domain and the most suitable position and quantity the information is originated from. This methodology step is also a preliminary action toward the use of neural networks for smart identification of specific statuses of interest of a mechatronic system, during both the set-up phase and the functioning of the system for its condition monitoring.
2nd IEEE International Workshop on Metrology for Industry 4.0 and IoT, MetroInd 4.0 and IoT 2019
978-1-7281-0429-4
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11589/239165
Citazioni
  • Scopus 6
  • ???jsp.display-item.citation.isi??? 0
social impact