DMCA
A data-driven approach for monitoring blade pitch faults in wind turbines (2011)
| Venue: | IEEE Transactions on Sustainable Energy |
| Citations: | 11 - 6 self |
BibTeX
@ARTICLE{Kusiak11adata-driven,
author = {Andrew Kusiak and Anoop Verma and Student Member},
title = {A data-driven approach for monitoring blade pitch faults in wind turbines},
journal = {IEEE Transactions on Sustainable Energy},
year = {2011},
pages = {87--96}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstract—A data-mining-based prediction model is built to monitor the performance of a blade pitch. Two blade pitch faults, blade angle asymmetry, and blade angle implausibility were analyzed to determine the associations between them and the components/subassemblies of the wind turbine. Five data-mining algorithms have been studied to evaluate the quality of the models for prediction of blade faults. The prediction model derived by the genetic programming algorithm resulted in the best accuracy and was selected to perform prediction at different time stamps. Index Terms—Blade angle asymmetry, blade angle implausibility, cost-sensitive classification, data mining, genetic programming (GP).
Keyphrases
wind turbine blade pitch fault data-driven approach angle implausibility genetic programming index term blade angle asymmetry prediction model data-mining algorithm component subassemblies blade fault data-mining-based prediction model cost-sensitive classification data mining angle asymmetry genetic programming algorithm different time stamp blade pitch
