Orderly Clustering of Active Distribution Network Planning Scenarios Based on Change Point Theory

An orderly clustering method 14 active distribution network planning scenarios with distributed generators (DG) based on time series modeling is investigated in this paper. Firstly, the medium and long term planning of active distribution network is simulated in time series, and the time series mathematical model of load growth is established according to different types. Then, the change point theory in statistics is introduced to establish the sub-scenarios of DG and load sequential partition, respectively. Finally, the above-mentioned time series scenarios are fused by the comprehensive change point theory, and the typical time series scenarios are extracted by segments. The simulation results show that, by searching for change points, nearly a million moderately and long-term sequential scenarios can be clustered into 14 continuous-time scenarios in an orderly manner, which greatly reduces the number of scenarios and improves the efficiency.