| Autor: |
Baoyi, Wang, Weiming, Zhan, Shaomin, Zhang |
| Zdroj: |
2012 2nd International Conference on Applied Robotics for the Power Industry (CARPI); 1/ 1/2012, p175-179, 5p |
| Abstrakt: |
In order to improve the accuracy of power system load forecasting, historical load data must be preprocessed. We use an improved k-means clustering algorithm to fix the bad load data of power system. The algorithm focus on the characteristics of the power system load data, use a new distance calculation method and an evaluation function of clusters number to improve the clustering results. This paper takes daily load data for research object and use a new k-means algorithm to extract daily load characteristic curve, which is used for bad data detection and identification in power system load data. Eventually, we program with Matlab, and then make the simulation analysis. The result shows that this method is effective. [ABSTRACT FROM PUBLISHER] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
