Use this link to cite:
http://hdl.handle.net/2183/23690 Sistema de detección de alarmas de fallos en el tren mecánico de un aerogenerador
Loading...
Identifiers
Publication date
Authors
López de la Cruz, David
Pantoja Álvarez, Luis
Irigoyen, Eloy
Advisors
Other responsabilities
Journal Title
Bibliographic citation
Lopez, D., Pantoja, L., Irigoyen, E. (2019). Sistema de detección de alarmas de fallos en el tren mecánico de un aerogenerador. En XL Jornadas de Automática: libro de actas, Ferrol, 4-6 de septiembre de 2019 (pp. 192-199). DOI capítulo: https://doi.org/10.17979/spudc.9788497497169.192. DOI libro: https://doi.org/10.17979/spudc.9788497497169
Type of academic work
Academic degree
Abstract
[Resumen] Este trabajo presenta un sistema automático de detección de alarmas en el tren mecánico de un aerogenerador. Este sistema se basa en un gemelo digital, como un modelo físico simulado del conjunto de estudio (sistema de transmisión), para entrenar con datos simulados al sistema. A esto se une un modelo de detección entrenado basado en técnicas de clasificación, que, a partir de datos reales, devuelve las alarmas de fallos del dispositivo. De esta forma, se maximiza la cobertura ante fallos, minimizando falsos positivos y negativos del sistema. El trabajo abarca fallos en un rodamiento, con un procedimiento extrapolable al tren mecánico.
[Abstract] This project develops an automatic system for detecting alarms in the mechanical train of a wind turbine. This system is based on a digital twin, that is, a simulated physical model of the study set (system), to train with simulated data to the system. To this is added a trained detection model based on classification techniques, which, based on real data, returns the possible failure alarms of the study element. In this way, coverage is maximized against failures, minimizing false positives and negatives of the system. The project covers faults in a bearing, with a procedure that can be extrapolated to the entire mechanical train.
[Abstract] This project develops an automatic system for detecting alarms in the mechanical train of a wind turbine. This system is based on a digital twin, that is, a simulated physical model of the study set (system), to train with simulated data to the system. To this is added a trained detection model based on classification techniques, which, based on real data, returns the possible failure alarms of the study element. In this way, coverage is maximized against failures, minimizing false positives and negatives of the system. The project covers faults in a bearing, with a procedure that can be extrapolated to the entire mechanical train.
Description
Editor version
Rights
Atribución-NoComercial-CompartirIgual 4.0


