Simulation of pollution accumulation characteristics of insulators and prediction of pollution impact rate based on improved neural network
Conference: CAIBDA 2022 - 2nd International Conference on Artificial Intelligence, Big Data and Algorithms
06/17/2022 - 06/19/2022 at Nanjing, China
Proceedings: CAIBDA 2022
Pages: 8Language: englishTyp: PDF
Authors:
Song, Zhen; Cheng, Xuezhen; Niu, Huihui (College of Electrical Engineering and Automation, Shandong University of Science and Technology, Qingdao, China)
Ren, Zheng (College of Electrical Engineering and Automation, Shandong University of Science and Technology, Qingdao, China & State Grid Shandong Electric Power Company Laiyang power supply company, Yantai, China)
Zhang, Xudong (College of Electrical Engineering and Automation, Shandong University of Science and Technology, Qingdao, China & State Grid Shandong Electric Power Company Qingzhou power supply company, Weifang, China)
Wang, Changan (China Mobile Shandong Co., Ltd. Yantai Branch, Yantai, China)
Abstract:
To study the natural pollution accumulation characteristics of insulators under different meteorological conditions, this paper takes XP-70 suspended porcelain insulator as the research object, introducing pollution particle deposition criteria based on humidity and surface energy, and establishes a numerical simulation model of natural pollution accumulation of insulator in COMSOL software. The standard of insulator pollution impact rate is used to characterize the natural contamination state and characteristics of insulators. The feasibility of the simulation model was verified by setting up an experimental platform for natural pollution deposition. Using this model, the influence of different factors on insulator pollution impact rate was analyzed, and particle swarm optimization BP neural network (ALF-PSO-BPNN) based on inverse cosine learning factor was selected to establish the prediction model of insulator pollution impact rate. The simulation data were used to train the prediction model, and the error analysis and performance comparison showed that the prediction model could better predict the insulator pollution impact rate.