Wind Data Preprocessing Algorithm Based on Extracting Isolated Points

Provided by: Science & Engineering Research Support soCiety (SERSC)
Topic: Data Management
Format: PDF
In this paper, against the deficiency of existing wind data processing, proposed a data processing method based on special points extraction S-time algorithm, combining the thought of clustering-based on density with preprocessing, preliminary filtering the special points then determine isolated points via the method of outlier determination that this paper proposed for secondary screening, finally use the special points to represent the whole trend of wind power. Through the experiment, the method can accurately filter out the special points of data, the extraction of special points make data distribution more intuitive, can effectively improve the prediction efficiency at the same time, as reduce the amount, greatly reduce the required memory space.

Find By Topic