The rapid progress in the fields of Big data, Digital technology coupled with growth of Industrial technology has led to the evolution of a vision called Industry 4.0 (fourth industrial revolution). Industry 4.0 has nurtured a lot of research and development in making the factories smart, efficient, and digitally connected. The global competition has made the industrial maintenance very significant. Condition- Based Maintenance is one of the most effective maintenance strategies helping the industries in achieving cost effectiveness while also being reliability centric. Majority of the critical equipment in industries now a days are already equipped standard condition-based maintenance practices. But still there seems to be a lack of applications of CBM when it comes to Linear mechanical system components like Linear bearings and toothed belt. This thesis work is aimed at the implementation of condition monitoring for the linear mechanical components for an already automated assembly line with Linear axis guides. The realization of this Condition Based Maintenance follows the MIMOSA OSA-CBM architecture steps one by one. Vibration analysis of these components are carried with the help wired and wireless accelerometer sensors. The vibration measurements on the assembly line are carried out periodically. The acquired vibration data is subjected to time and frequency domain analysis, filtering along with feature extraction. With the help of some Machine Learning concepts and algorithm some of the faults associated with Linear bearings and toothed belt are identified and a framework is established. Later a GUI is developed for the visualization effective results. Furthermore, the implementation of proposed CBM can be extended to additional components and faults. This scheduled predictive maintenance practice can possibly develop into continuous online monitoring in the future which can be both cost and reliability effective.
Il rapido progresso nei campi dei big data, la tecnologia digitale unita alla crescita della tecnologia industriale ha portato all'evoluzione di una visione chiamata Industria 4.0 (quarta rivoluzione industriale). L'Industria 4.0 ha alimentato molte attività di ricerca e sviluppo per rendere le fabbriche intelligenti, efficienti e connesse digitalmente. La competizione globale ha reso molto significativa la manutenzione industriale. La manutenzione basata sulle condizioni è una delle strategie di manutenzione più efficaci che aiuta le industrie a raggiungere l'efficacia dei costi pur essendo incentrata sull'affidabilità. La maggior parte delle apparecchiature critiche nelle industrie di oggi sono già dotate di pratiche di manutenzione standard basate sulle condizioni. Tuttavia, sembra che manchino ancora applicazioni di CBM quando si tratta di componenti di sistemi meccanici lineari come cuscinetti lineari e cinghia dentata. Questo lavoro di tesi è finalizzato all'implementazione del monitoraggio delle condizioni per i componenti meccanici lineari per una linea di assemblaggio già automatizzata con guide ad assi lineari. La realizzazione di questa Condition Based Maintenance segue i passaggi dell'architettura MIMOSA OSA-CBM uno per uno. L'analisi delle vibrazioni di questi componenti viene effettuata con l'aiuto di sensori accelerometrici cablati e wireless. Le misurazioni delle vibrazioni sulla linea di montaggio vengono eseguite periodicamente. I dati di vibrazione acquisiti vengono sottoposti ad analisi nel dominio del tempo e della frequenza, filtrando insieme all'estrazione delle caratteristiche. Con l'aiuto di alcuni concetti e algoritmi di apprendimento automatico vengono identificati alcuni dei guasti associati ai cuscinetti lineari e alla cinghia dentata e viene stabilita una struttura. Successivamente viene sviluppata una GUI per la visualizzazione dei risultati effettivi. Inoltre, l'implementazione della proposta CBM può essere estesa a componenti aggiuntivi e guasti. Questa pratica di manutenzione predittiva programmata può essere effettuata in futuro con un monitoraggio online continuo che può essere efficace sia in termini di costi che di affidabilità.
Implementation of condition-based maintenance for linear mechanical systems
Agile Nagappa, Ganesh
2020/2021
Abstract
The rapid progress in the fields of Big data, Digital technology coupled with growth of Industrial technology has led to the evolution of a vision called Industry 4.0 (fourth industrial revolution). Industry 4.0 has nurtured a lot of research and development in making the factories smart, efficient, and digitally connected. The global competition has made the industrial maintenance very significant. Condition- Based Maintenance is one of the most effective maintenance strategies helping the industries in achieving cost effectiveness while also being reliability centric. Majority of the critical equipment in industries now a days are already equipped standard condition-based maintenance practices. But still there seems to be a lack of applications of CBM when it comes to Linear mechanical system components like Linear bearings and toothed belt. This thesis work is aimed at the implementation of condition monitoring for the linear mechanical components for an already automated assembly line with Linear axis guides. The realization of this Condition Based Maintenance follows the MIMOSA OSA-CBM architecture steps one by one. Vibration analysis of these components are carried with the help wired and wireless accelerometer sensors. The vibration measurements on the assembly line are carried out periodically. The acquired vibration data is subjected to time and frequency domain analysis, filtering along with feature extraction. With the help of some Machine Learning concepts and algorithm some of the faults associated with Linear bearings and toothed belt are identified and a framework is established. Later a GUI is developed for the visualization effective results. Furthermore, the implementation of proposed CBM can be extended to additional components and faults. This scheduled predictive maintenance practice can possibly develop into continuous online monitoring in the future which can be both cost and reliability effective.| File | Dimensione | Formato | |
|---|---|---|---|
|
2021_12_AGILE_NAGAPPA.pdf
non accessibile
Descrizione: Thesis Document
Dimensione
2.47 MB
Formato
Adobe PDF
|
2.47 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/183935