A Review of Hybrid Prognostics Applicationsfor Power & Energy Systems

• Jose I. Aizpurua ^[2] ; Jokin Alzibar ^[1] ; Ibai Ramirez ^[1] ; Ekhi Zugasti ^[1] ; Joel Pino ^[1]
  1. [1] Universidad de Mondragón/Mondragon Unibertsitatea

     Universidad de Mondragón/Mondragon Unibertsitatea

     Mondragón, España

     
  2. [2] Mondragon Unibertsitatea, Electronics & Computer Science Department, and Ikerbasque, Basque Foudation for Science
• Localización: Digital Maintenance in the Digital Twin Era: Proceedings of the 64th ESReDA Seminar& Doctoral Workshop / coord. por Aitor Goti Elordi, Antonio J. Guillén, Juan Chiachío Ruano, Manuel Chiachio Ruano, 2024, ISBN 978-84-1325-228-5, págs. 301-332
• Idioma: inglés
• Enlaces
  • Texto Completo Libro (pdf)
• Resumen

  • Prognostics is the ability to acquire knowledge about events before they occur. In industrialsettings, prognostics primarily revolves around predicting the remaining useful life (RUL) ofassets. Data-driven (DD) prognostics models capture complex fault-to-failure dynamics butlack of explicability in predictions. Physics-based models make use of physics-of-failure modelsto predict RUL, albeit at the expense of reduced accuracy compared to DD models. In thiscontext, hybrid prognostics models combine both data-driven and physics-based approachesto enhance accuracy and explicability of results. This paper reviews authors’ experience in thedevelopment of hybrid prognostics solutions applied to power and energy assets.