Distributed photovoltaic siting optimisation considering load centre similarity

Abstract

The load fluctuation law remains challenging for traditional distribution line users. Furthermore, as the number of distributed photovoltaic installations connected to the distribution network rises, the issue of voltage overload on the distribution lines becomes more prevalent. This, in turn, adds complexity to power system operation. To configure the PV access location, whilst ensuring the line voltage is qualified, this paper applies the DBSCAN density clustering and K-means optimization algorithm to classify the distribution network line load. Different characteristics of the load collection are then constructed, and the concept of the load centre distance is proposed. PV access points are selected based on the load center characteristic curve and PV curve similarity. Model simulation is used to verify the conclusions, which demonstrate the rationality and effectiveness of the proposed method. The model simulation is conducted to confirm the soundness and credibility of the conclusions drawn.