Topology Identification of Low-voltage Transformer Area Based on Improved Particle Swarm Algorithm. Issue 1 (July 2021)

• Record Type: Journal Article
• Authors: Liu, Han; Huizhu, Xian; Dapeng, Duan; Fengming, Lv; Jian, Sun; Haofei, Mao; Qiang, Wang
• Is Part Of: Journal of physics. Volume 1972:Issue 1(2021)
• Publication Date: 2021-07
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1972/1/012049
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: With the continuous changes in the user-side power environment, the low-voltage distribution network has become more and more complex, which brings great challenges to the line loss management and topology identification of the transformer area. To solve the shortcomings of traditional particle swarm optimization such as poor ergodicity of population initialization and easy to fall into premature convergence, this paper proposed a K-means clustering analysis algorithm based on GA-CPSO. First, on the basis of the traditional particle swarm algorithm, the chaotic shrinkage factor was introduced and the parameters were optimized. Secondly, the genetic algorithm was combined with the chaotic particle swarm optimization algorithm, the crossover and mutation operations of the genetic algorithm were used to establish an information exchange mechanism between particles, and it was combined with the K-means clustering method. The simulation results on the test benchmark function show that the improved particle swarm optimization algorithm in this paper has significantly improved the search speed and optimization accuracy. Finally, taking an actual transformer area as an example, the method was applied to the transformer area topology recognition analysis. Using the electrical parameters such as voltage, current, and active power obtained from the monitoring terminal as sample data, a simulation analysis was carried out to verify the effectiveness and feasibility of the algorithm. The performance parameters of different algorithms had been compared and analyzed through multiple experiments, and it was proved that this method can effectively improve the accuracy of platform topology recognition, and has strong practicability and generalization.