Weighted distance based outlier factor identifying and its application in wind data pre-processing

This paper analyses the properties of raw wind data and proposes a novel wind data pre-processing method. Firstly, the raw wind data are divided into six categories according to their attribute magnitudes. The statistical characteristics of the data reveal that the invalid data can be considered as outliers compared to the valid ones. Then the LOF algorithm and a firstly designed weighted distance based outlier factor identifying algorithm (WDOF) are applied to detect and remove the invalid data. WDOF considers sticking close to the equivalent power curve as an auxiliary factor of being valid. Numerical experiments have verified the effectiveness of the proposed algorithm.