Combined rotor faults diagnosis in induction motors using MVSA and Intra-Mode Variational Modal Decomposition (IM-VMD)
Résumé
In modern industry, constant production and operation without stoppages is viewed as being vital in ensuring competitiveness and efficiency.To maintain continuity in production, industries make use of advanced diagnostic measures by using of different performance measures such as reliability and risk assessment.In addition, within the context of Industry 4.0, there are also smart devices and Internet of Things (IoT) that can be used to enhance monitoring and optimize production processes intelligently.In such a context, this study seeks to examine the problem of detecting combined faults in three-phase induction machines via vibration signals.The purpose of this research is to diagnose fault signatures as early as possible even in the presence of multiple interactions between the faults.To achieve this objective, an intra-mode variational mode decomposition (IM-VMD) approach that employs an embedded source separation strategy was applied for decomposing vibration signals into multiple intrinsic modes.Using this framework, it became possible to isolate those vibration components associated with faults and increase interpretability of signals in terms of time and frequency domains.The outcomes demonstrated an effective identification of fault signatures; the extracted vibration frequency components match the theoretical frequencies of the defects.More precisely, the coefficient of correlation between extracted frequency components and theoretically calculated ones equals 0.95 to 1.This finding suggests that the proposed algorithm provides reliable results that can be applied in practice for detecting combined faults in three-phase induction machines.It is also possible to highlight that by utilizing the introduced method, it becomes possible to detect fault signatures at the earliest stage of their appearing.
Citer ce document
Accès au document
Texte intégral en lecture en ligne, réservé aux abonnés SPHAERO et aux membres de l'institution. Se connecter
Voir l'article sur le site de la revueAuteur(s)
Statistiques
Consultations : 1
Téléchargements : 0