Effects of the pre-processing algorithms in fault diagnosis of wind turbines

The wind sectors pends roughly 2200M€ in repair the wind turbines failures. These failures do not contribute to the goal of reducing greenhouse gases emissions. The 25–35% of the generation costs are operation and maintenance services. To reduce this amount, the wind turbine industry is backing on t...

Descripción completa

Detalles Bibliográficos
Autores: Marti-Puig, Pere|||0000-0001-6582-4551, Blanco Martínez, Alejandro, Cárdenas Araújo, Juan José, Cusidó Roura, Jordi|||0000-0002-1951-1498, Sole Casals, Jordi
Tipo de recurso: artículo
Fecha de publicación:2018
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:2117/336039
Acceso en línea:https://hdl.handle.net/2117/336039
https://dx.doi.org/10.1016/j.envsoft.2018.05.002
Access Level:acceso abierto
Palabra clave:System failures (Engineering)
Wind turbines--Maintenance and repair
Machine learning
Supervisory control systems
Wind farms
SCADA data
Pre-processing
Outliers
Fault diagnosis
Renewable energy
Aerogeneradors -- Manteniment i reparació
Avaries
Aprenentatge automatic
Àrees temàtiques de la UPC::Energies::Energia eòlica::Aerogeneradors
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial
Descripción
Sumario:The wind sectors pends roughly 2200M€ in repair the wind turbines failures. These failures do not contribute to the goal of reducing greenhouse gases emissions. The 25–35% of the generation costs are operation and maintenance services. To reduce this amount, the wind turbine industry is backing on the Machine Learning techniques over SCADA data. This data can contain errors produced by missing entries, uncalibrated sensors or human errors. Each kind of error must be handled carefully because extreme values are not always produced by data reading errors or noise. This document evaluates the impact of removing extreme values (outliers) applying several widely used techniques like Quantile, Hampel and ESD with the recommended cut-off values. Experimental results on real data show that removing outliers systematically is not a good practice. The use of manually defined ranges (static and dynamic) could be a better filtering strategy.