• Login
    Ver ítem 
    •   AEIPRO Principal
    • Proceedings from the International Congress on Project Management and Engineering
    • CIDIP 2019 (Málaga)
    • Ver ítem
    •   AEIPRO Principal
    • Proceedings from the International Congress on Project Management and Engineering
    • CIDIP 2019 (Málaga)
    • Ver ítem
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Mantenimiento predictivo mediante técnicas de machine learning

    Thumbnail
    Ver/
    AT03-020_2019.pdf (1020.Kb)
    Fecha
    2019
    Autor
    Guerrero Cano, Manuel
    Luque Sendra, Amalia
    Lama Ruiz, Juan Ramón
    Córdoba Roldán, Antonio
    Metadatos
    Mostrar el registro completo del ítem
    Resumen
    Industrial maintenance is a field of engineering with a high impact on costs and manufacturing times for industrial products. This work is part of the work areas of diagnosis and maintenance of industrial processes and explores techniques of detection of incipient anomalies based on automatic learning. The predictive maintenance aims to predict failures in the machinery, so that repairs can be scheduled without interrupting the production process. It consists of an analysis of the operation of the equipment to detect warning signs that indicate that one of its parts is not working in the correct way. The cost of predictive maintenance is less than that of the corrective, due to the expenses generated by the repair of equipment and downtime due to production stoppage. Information technologies are giving rise to a new revolution that is called industry 4.0. One of the fields of application is in the improvement of maintenance. By using process and product data, machine learning techniques could be applied to determine when failures can occur. In this paper, supervised and unsupervised learning techniques (parametric and nonparametric) will be explored and their usefulness for their application in predictive maintenance will be discussed.
    URI
    http://dspace.aeipro.com/xmlui/handle/123456789/2293
    Colecciones
    • CIDIP 2019 (Málaga) [169]

    DSpace software copyright © 2002-2016  AEIPRO
    Contacto | Sugerencias
     

     

    Listar

    Todo AEIPROComunidades & ColeccionesPor fecha de publicaciónAutoresTítulosMateriasEsta colecciónPor fecha de publicaciónAutoresTítulosMaterias

    Mi cuenta

    Acceder

    DSpace software copyright © 2002-2016  AEIPRO
    Contacto | Sugerencias