Customer clustering and TDLP (typical daily load profile) generation using the clustering algorithm
暂无分享,去创建一个
KEPCO has installed electronic power meters for customers who have relatively high annual power consumption and operated the AMR (Automatic Meter Reading) system that records the amount of power consumption every 15 minutes. To analyze the load of the distribution line using the data collected by the AMR system, this study suggested a method of generating a typical load profile based on the clustering method using the load data of customers. Currently, typical load profiles are usually generated by calculating the average power consumption as recorded every 15 minutes based on the load profiles of customers with the same contract code. Even though they share the same contract code, however, the load analysis has low accuracy because it often shows different load profiles. This study suggested a means of performing customer clustering and generating the typical load profile of each group using the k-means method as one of the clustering methods and 15-minute AMR data of customers sharing the same contract code.
[1] Anil K. Jain,et al. Algorithms for Clustering Data , 1988 .
[2] G.W. Chang,et al. A Simplified Forward and Backward Sweep Approach for Distribution System Load Flow Analysis , 2006, 2006 International Conference on Power System Technology.
[3] Chongqing Kang,et al. Novel approach considering load-relative factors in short-term load forecasting , 2004 .
[4] Balho H. Kim,et al. Development of Customer Oriented Load Management Software for Savings on Utility Bills in the Electricity Market , 2007 .