Home > Research > Publications & Outputs > Optimal sensor selection for wind turbine condi...
View graph of relations

Optimal sensor selection for wind turbine condition monitoring using multivariate principal component analysis approach

Research output: Contribution to conference - Without ISBN/ISSN Conference paperpeer-review

Published
Publication date09/2012
Number of pages7
<mark>Original language</mark>English
EventProceedings of the 18th International Conference on Automation & Computing (ICAC’12), Loughborough University, Leicestershire, UK, 8 September 2012 - Loughborough, United Kingdom
Duration: 8/09/2012 → …

Conference

ConferenceProceedings of the 18th International Conference on Automation & Computing (ICAC’12), Loughborough University, Leicestershire, UK, 8 September 2012
Country/TerritoryUnited Kingdom
CityLoughborough
Period8/09/12 → …

Abstract

With the fast growth in wind energy technologies, research into the condition monitoring system for wind turbines has drawn more attentions. Despite the advantages from the condition monitoring systems, there are also several challenges for the application of condition monitoring system for wind turbines. Accurate and adequate information of the wind turbine is needed for the condition monitoring system to carry out analysis, particularly with the growing size of wind farms. Another challenge is the huge amount of data needing to be collected, handled and processed. Minimising the number of sensors whilst still maintaining a sufficient number to assess the system’s conditions is a critical concern for condition monitoring. This paper focuses on the application of Principal Component Analysis (PCA) to the optimization of sensor selection for wind turbine condition monitoring. The principle behind the proposed methodology is presented and the method is also validated using simulation data obtained from wind power generation model in PSCAD/EMTDC.