SISTEMI INTELIGENTNOG PREDIKTIVNOG ODRŽAVANjA U ELEKTROPRIVREDI

Abstract

The development of information and communication technologies in the field of monitoring technical and technological systems enables the collection of a large amount of data related to maintenance assets. Due to the large amount of data and limited time for their analysis, there was a need to replace conventional, human- dependent methods with intelligent early fault diagnosis systems and, consequently, to apply machine learning methods to support the maintenance function. The application of machine learning algorithms in the concept of predictive maintenance leads to a new strategy Intelligent Predictive Maintenance - IPdM and the concept of Maintenance 4.0. This strategy contributes to determining the optimal equipment maintenance period, shortening downtime, increasing system efficiency, and reducing maintenance costs. The paper first presents the strategy concepts and architecture of intelligent predictive maintenance systems. Further, paper provides implementation guidelines and implementation risk analysis. Finally, literature examples of intelligent predictive maintenance strategy application in electric power systems are presented.