Maintenance Digital Twin using vibration data
Raffaele Abbate, Mario Caterino, Marcello Fera, Francesco Caputo
Abstract
Maintenance represents a critical aspect of manufacturing systems. Its optimization may lead companies to save costs related to production losses and can extend the whole lifecycle of assets or components. This paper aims at proposing a methodology for evaluating the behavior of systems for maintenance purpose by means of analysis carried out on real vibration data collected in the work environment. To this purpose, a digital twin approach has been developed and the statistical method of the control charts has been integrated to conduct the analysis about the vibration data. The methodology has been applied to a real case study in order to analyze the behavior of an electric motor used in the industrial sector and the results show that the method is effective in predicting possible failures that can lead to apply preventive maintenance actions.