K-means Clustering-based Data Mining Methodology to Discover the Prosumers’ Energy Features

In the paper, a Data Mining methodology is proposed to identify the energy features of the prosumers. The K-means clustering algorithm has been used to obtain prosumers' categories considering two specific indicators which can help the Distribution Network Operators (DNOs) in the optimal operating of the networks, namely the injected average hourly power and the total annual energy. Testing the methodology has been done for a database corresponding to the prosumers connected in the low voltage (LV) distribution networks belonging to a Romanian DNO in 2019. The obtained results have confirmed the importance of the Data Mining to extract easy and fast the energy features of the prosumers from the large-size databases and used by the DNOs in the Decision-Making process associated with the optimal operating of the distribution networks.