Industrial data-driven monitoring based on incremental learning applied to the detection of novel faults

The detection of uncharacterized events during electromechanical systems operation represents one of the most critical data challenges dealing with condition-based monitoring under the Industry 4.0 framework. Thus, the detection of novelty conditions and the learning of new patterns are considered a...

Descripción completa

Detalles Bibliográficos
Autores: Saucedo Dorantes, Juan Jose, Delgado Prieto, Miquel|||0000-0001-9282-838X, Osornio Rios, Roque A., Romero Troncoso, René de Jesús
Tipo de recurso: artículo
Fecha de publicación:2020
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/328705
Acceso en línea:https://hdl.handle.net/2117/328705
https://dx.doi.org/10.1109/TII.2020.2973731
Access Level:acceso abierto
Palabra clave:Machine learning
Databases
Electromechanical devices
Condition monitoring
Fault detection
Feature extraction
Incremental learning
Novelty detection
Aprenentatge automàtic
Bases de dades
Dispositius electromecànics
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
Descripción
Sumario:The detection of uncharacterized events during electromechanical systems operation represents one of the most critical data challenges dealing with condition-based monitoring under the Industry 4.0 framework. Thus, the detection of novelty conditions and the learning of new patterns are considered as mandatory competencies in modern industrial applications. In this regard, this article proposes a novel multifault detection and identification scheme, based on machine learning, information data-fusion, novelty-detection, and incremental learning. First, statistical time-domain features estimated from multiple physical magnitudes acquired from the electrical motor under inspection are fused under a feature-fusion level scheme. Second, a self-organizing map structure is proposed to construct a data-based model of the available conditions of operation. Third, the incremental learning of the condition-based monitoring scheme is performed adding self-organizing structures and optimizing their projections through a linear discriminant analysis. The performance of the proposed scheme is validated under a complete set of experimental scenarios from two different cases of study, and the results compared with a classical approach.