Machine learning and deep learning based methods toward industry 4.0 predictive maintenance in induction motors: State of the art survey

Purpose: Developments in Industry 4.0 technologies and Artificial Intelligence (AI) have enabled data-driven manufacturing. Predictive maintenance (PdM) has therefore become the prominent approach for fault detection diagnosis (FD/D) of induction motors (IMs). The early FD/D IMs are critical processes, considering that they constitute main power source industrial production environment. Machine learning (ML) methods enhanced performance reliability PdM. Various deep (DL) based emerged recent years, providing automatic feature engineering thereby alleviating drawbacks traditional ML methods. This paper presents a comprehensive survey DL since 2015. An overview architectures used this purpose is also presented. A discussion trends given as well future directions research.Design/methodology/approach: been carried out through all available publication databases using related keywords. Classification reviewed works done according to techniques algorithmsFindings: PdM mainly introduced implemented IM years. Novel on single hybrid techniques. signal preprocessing moreover, combined with algorithms enhance engineering. Publicly datasets mostly test developed methods, however should well. Multi-agent system (MAS) employing classifiers explored. Several investigated multiple faults, however, presence faults occurring simultaneously rarely investigated.Originality/value: review advances