TOWARDS A PREDICTIVE MAINTENANCE BASED ON ULTRASONIC GUIDED WAVES NON-DESTRUCTIVE TESTING TECHNIQUE: A CASE STUDY

Mohamed Sfar (1) , Slah Yaacoubi (2) , Sofiene Dellagi (1) , Mahjoub El Mountassir (2) , Sadok Turki (1)

1 LGIPM - Laboratoire de Génie Informatique, de Production et de Maintenance
2 Institut de soudure

This paper concerns the predictive maintenance (PM), which is in increasing use in many fields. It proposes the use of a non-destructive testing technique, namely ultrasonic guided waves, in order to reinforce the PM of a given installation. To achieve PM, various sensors are required to collect data, which are needed as inputs of predictive models. These sensors are employed to monitor the health (said also in other words degradation) of the installation to be maintained in operation. In this study two models based on EDM-ARIMA and neural network are built and applied on data collected on a tube via the ultrasonic guided waves. Residual useful life of the structure is provided in different cases.

Mots clés: Predictive maintenance; Ultrasonic Guided Waves; Prognostic; Residual Useful Life; ARIMA; EMD; Neural Network; Non-destructive testing

Domaines: Automatique / Robotique; Mécanique [physics.med-ph]

Mohamed Sfar, Slah Yaacoubi, Sofiene Dellagi, Mahjoub El Mountassir, Sadok Turki. TOWARDS A PREDICTIVE MAINTENANCE BASED ON ULTRASONIC GUIDED WAVES NON-DESTRUCTIVE TESTING TECHNIQUE: A CASE STUDY. 13ème CONFERENCE INTERNATIONALE DE MODELISATION, OPTIMISATION ET SIMULATION (MOSIM2020), 12-14 Nov 2020, AGADIR, Maroc, Nov 2020, AGADIR (virtual), Morocco. ⟨hal-03192678⟩