International Journal of Advanced Studies in Computer Science and Engineering

Open Access Open Access  Restricted Access Subscription Access

Wind Farms Predictive Maintenance Based on Condition Monitoring System


Affiliations
1 Faculty of sciences and technology, Ziane Achour University, 17000 Djelfa, Algeria
2 Ingenium Research Group, Universidad Castilla-La Mancha, 13071 Ciudad Real, Spain
3 Materials, Processes and Environment Research Unit, FSI, M’hamed Bougara University, 35000 Boumerdes, Algeria

This article focuses on a new methodology based on fuzzy logic to predict the maintenance of wind farms. In wind systems, Vibrations present the main cause of the occurrence of failures. The prediction of these failures can be ensured by correct processing of the set of vibration signals collected by the Condition Monitoring System (CMS) and declaration of the appropriate maintenance action by the connected management system SCADA (Supervisory Control And Data Acquisition). In this paper, fuzzy logic is used as a new approach to processing vibration signals by exploiting their features for the generation of probabilities of occurrence of failures. The probabilities generated correspond to alarms (Green, Orange and Red) and the alarms correspond to maintenance actions to be applied.
User
Notifications
Font Size

Abstract Views: 134




  • Wind Farms Predictive Maintenance Based on Condition Monitoring System

Abstract Views: 134  | 

Authors

Tahar Benmessaoud
Faculty of sciences and technology, Ziane Achour University, 17000 Djelfa, Algeria
Fausto Pedro Garcia Marquez
Ingenium Research Group, Universidad Castilla-La Mancha, 13071 Ciudad Real, Spain
Kamal Mohammedi
Materials, Processes and Environment Research Unit, FSI, M’hamed Bougara University, 35000 Boumerdes, Algeria
Alberto Pliego Marugan
Ingenium Research Group, Universidad Castilla-La Mancha, 13071 Ciudad Real, Spain

Abstract


This article focuses on a new methodology based on fuzzy logic to predict the maintenance of wind farms. In wind systems, Vibrations present the main cause of the occurrence of failures. The prediction of these failures can be ensured by correct processing of the set of vibration signals collected by the Condition Monitoring System (CMS) and declaration of the appropriate maintenance action by the connected management system SCADA (Supervisory Control And Data Acquisition). In this paper, fuzzy logic is used as a new approach to processing vibration signals by exploiting their features for the generation of probabilities of occurrence of failures. The probabilities generated correspond to alarms (Green, Orange and Red) and the alarms correspond to maintenance actions to be applied.