Location and Size Planning of Distributed Photovoltaic Generation in Distribution network System Based on K-means Clustering Analysis

[摘要] The paper presents a method to generate the planning scenarios, which is based on K-means clustering analysis algorithm driven by data, for the location and size planning of distributed photovoltaic (PV) units in the network. Taken the power losses of the network, the installation and maintenance costs of distributed PV, the profit of distributed PV and the voltage offset as objectives and the locations and sizes of distributed PV as decision variables, Pareto optimal front is obtained through the self-adaptive genetic algorithm (GA) and solutions are ranked by a method called technique for order preference by similarity to an ideal solution (TOPSIS). Finally, select the planning schemes at the top of the ranking list based on different planning emphasis after the analysis in detail. The proposed method is applied to a 10-kV distribution network in Gansu Province, China and the results are discussed.

[发布机构] School of Electrical Engineering, Beijing Jiaotong University, Beijing, China^1;China Electric Power Research Institute, Beijing, China^2

[效力级别] 材料科学

[关键词] Decision variables;Distribution network system;K-means clustering analysis;Maintenance cost;Pareto-optimal front;Photovoltaic generation;Self adaptive genetic algorithm;Voltage offsets