Clustering of Transmission System Loads Based on Monthly Load Class Energy Consumptions

This paper presents an approach for clustering aggregated loads based on load class energy consumption time series, and choosing representative loads for each group. The described approach was applied in a transmission system load modelling study. The goal of the study was to choose representative loads for measurement-based modelling. The work was motivated by the limited number of available measurement devices and available personnel for data processing. The monthly energy consumption of each load class was known for each aggregated bus load. After measurement data pre-processing the larger loads were clustered using K-means algorithm, and smaller assigned to clusters. Representative loads were selected from each cluster.