Yakhni, Mohammad (Universite de La Rochelle), Cauet, Sebastien (University of Poitiers), Sakout, Anas (Université de La Rochelle), Assoum, Hassan (Beirut Arab University), El-Gohary, Mohamed (Beirut Arab University)

Online Condition Monitoring of a Vacuum Process Based on Adaptive Notch Filters

Scheduled for presentation during the Regular Session "Diagnosis and Prognosis" (WeA2), Wednesday, November 16, 2022, 16:00−16:20, ROOM - E 210

16th European Workshop on Advanced Control and Diagnosis, November 16-18, 2022, Nancy, France

This information is tentative and subject to change. Compiled on April 25, 2024

Abstract

Vacuum systems play an essential role in industrial installations. This system is subject to various operating conditions, making it susceptible to failures that can cause real damage in the working environment. Online condition monitoring is, therefore, necessary to quickly detect any malfunction and avoid its consequences. It increases the life of system components while reducing maintenance costs and downtime. Various techniques can perform this task. In this paper, we have relied on the transient motor current signature analysis. The system studied is a vacuum system located in the Municipal Technical Center of Poitier, France. Matlab/Simulink program was used to build a digital twin of this system and create several types of faults at different operating speeds. Four Adaptive Notch Filter techniques were developed, based on different structures, which are: Chambers' all-pass structure, Regalia's all-pass solution, Cho, Choi & Lee's all-pass method, and M'Sirdi's structure, to identify the system faults. M'Sirdi's structure, which is the most recent technology, was discussed in detail. The comparison between the results of the four methods was presented, where the simulation results proved their effectiveness in achieving the desired goal, with the superiority of the M'Sirdi structure.