Abnormal Data Processing Method of Wind Turbine Based on CFSFDP and Quartile
As the wind turbine runs in the harsh environment,there will be abnormal data in the original data collected by the SCADA system.Accurate and effective data preprocessing of the original data is the basis of the follow-up fault early warning work.This paper presents an abnormal data processing method based on the combination of CFSFDP algo-rithm and quartile method.Based on the experimental verification of the measured data of the actual domestic wind farms,firstly,the abnormal state and outlier noise points are detected by the CFSFDP algorithm,and the quartile method is used to deal with the noise data near the normal working conditions.The experimental results show that this method can effec-tively eliminate the abnormal data,improve the data quality and meet the needs of follow-up research.
wind power generationSCADAquartile algorithmCFSFDP algorithm
NETL
NSTL
万方数据
