A comparative analysis of clustering algorithms applied to load profiling

摘要: With the electricity market liberalization, distribution and retail companies are looking for better strategies based on adequate information upon consumption patterns of its customers. A fair insight customers' behavior will permit definition specific contract aspects different patterns. In this paper, we propose a KDD project applied to data from utility client's database. To form classes, find set representative patterns, comparative analysis performance K-means, Kohonen Self-Organized Maps (SOM) Two-Level approach is made. Each customer class be represented by load profile obtained with algorithm best in used.