Segmentation of Peer-to-Peer Energy Market Using K-Means, Hierarchical, and DBSCAN Clustering Techniques

Peer-to-peer (P2P) energy trading has emerged as a transformative approach to decentralize the energy market. A critical challenge in P2P energy trading lies in efficiently grouping participants based on their common interest, such as price and quantity in order to maximize trade and participant’s satisfaction. Clustering techniques play a pivotal role in grouping buyers and sellers to facilitate optimal matching. This study evaluates and compares three widely used clustering techniques–K-Means, DBSCAN, and Hierarchical Clustering techniques for market segmentation in the context of P2P energy trading. The evaluation is based on clustering quality, trading performance, and computational efficiency. The result shows that the segmentation impacts trade and computation time significantly. The findings highlight that K-Means clustering effectively identifies distinct and well-separated groupings with least processing time, which makes it suitable for real-time applications. K-means based technique has successfully achieved the market segmentation.