The individuation of the operating conditions of an automatic mechatronic system according to the value of relevant features is useful for many purposes. Just to cite some relevant examples, the automatic identification of the best setting of a machine for the application of interest at the beginning of the operation, to optimize performances of the mechatronic system, or the condition monitoring of the system itself in order to individuate faults or criticalities during its working conditions [1], [2]. General guide-lines can be found in the literature, [2], [3], suggesting features in the time and/or frequency domain that are expected to give reliable and selective information about the correct setting or about the presence of typical defects in the mechatronic devices. Anyway, features are expected to be more and more representative, accurate, selective, resolute with respect to the statuses of complex systems, due to many possible causes. This goal is not a trivial task due to the need of merging huge amounts of data deriving from networks of sensors, of using advanced procedures of data mining and processing and of distributing intelligence according to the precepts of the so-called Industry 4.0 [4], [5]. Finally, features should be easy and light to calculate with reference to the data processing [6]. The aim of this paper is to propose a general methodology able to individuate effective features to define the conditions of mechatronic systems even complex, customized for the specific task. It is based on the possibility of creating a detailed theoretical model of the system, preliminary validated by an experimental activity able to suggest, by a sensitivity analysis, the most suitable features to be processed in order to define the statuses of interest. A practical example is also given, in order to demonstrate the feasibility of the approach.

A THEORETICAL-EXPERIMENTAL APPROACH FOR IDENTIFICATION OF BEST FEATURES FOR STATUS DEFINITION OF A MECHATRONIC SYSTEM

Antonella Gaspari;
2018

Abstract

The individuation of the operating conditions of an automatic mechatronic system according to the value of relevant features is useful for many purposes. Just to cite some relevant examples, the automatic identification of the best setting of a machine for the application of interest at the beginning of the operation, to optimize performances of the mechatronic system, or the condition monitoring of the system itself in order to individuate faults or criticalities during its working conditions [1], [2]. General guide-lines can be found in the literature, [2], [3], suggesting features in the time and/or frequency domain that are expected to give reliable and selective information about the correct setting or about the presence of typical defects in the mechatronic devices. Anyway, features are expected to be more and more representative, accurate, selective, resolute with respect to the statuses of complex systems, due to many possible causes. This goal is not a trivial task due to the need of merging huge amounts of data deriving from networks of sensors, of using advanced procedures of data mining and processing and of distributing intelligence according to the precepts of the so-called Industry 4.0 [4], [5]. Finally, features should be easy and light to calculate with reference to the data processing [6]. The aim of this paper is to propose a general methodology able to individuate effective features to define the conditions of mechatronic systems even complex, customized for the specific task. It is based on the possibility of creating a detailed theoretical model of the system, preliminary validated by an experimental activity able to suggest, by a sensitivity analysis, the most suitable features to be processed in order to define the statuses of interest. A practical example is also given, in order to demonstrate the feasibility of the approach.
978-88-31901-06-2
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/11589/239171
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact