Energy Losses Estimation in the Electric Distribution Networks Using Clustering-Based Selection of the Representative Feeders

The energy loss estimation is essential to evaluate the technical and economic performance corresponding to electric distribution networks (EDNs). The development of some energy loss estimation methodologies represents one of the main objectives included in the strategies implemented by the Distribution Network Operators (DNOs) for an optimal operation and planning of the EDNs. The paper presents an improved methodology for the energy losses estimation in the EDNs using a clustering-based selection of the representative feeders (RFs), considering both the energy and technical features. The annual transmitted energy, total length, cross-sections of each branch, and the capacity of the distribution transformers represented the input data in the clustering process. The RFs are identified for each feature class assigned to the energy patterns obtained from the clustering process. Testing the proposed methodology has been done considering the 20 kV EDNs from a rural area belonging to a Romanian DNO. The results emphasized the accuracy of the proposed method to determine the energy losses in the EDNs. The percentage error between the proposed method and the Backward-Forward Sweep-based method, considered the reference, was 3.5%, compared with 12.99% associated with the average element method.