Insulator fouling assessment model based on RBF neural network. (August 2022)

• Record Type: Journal Article
• Authors: Liu, Bingcai; Pei, Huikun; Fu, Jiaqing; Wang, Zhenhua; Fang, Zhiwen; Zhou, Te
• Is Part Of: Energy reports. Volume 8(2022)Supplement 5
• Page Start: 1429
• Page End: 1436
• Publication Date: 2022-08
• Subjects: Insulator -- Fouling flashover -- Leakage current -- Real-time data -- RBF Neural Networks; Power resources -- Periodicals; Energy industries -- Periodicals; Power resources; Periodicals; Electronic journals; 621.04205
• DOI: 10.1016/j.egyr.2022.03.127
• Languages: English
• ISSNs: 2352-4847

Abstract

Abstract: The fouling and flashing of line insulators often cause serious damage to transmission lines and makes subsequent maintenance difficult, so being able to assess the fouling status of insulators in real time is an important step in achieving life-cycle control of insulators. For the purpose of predicting the actual operating condition of insulators, this paper uses line insulators as the object of study, takes leakage current, ambient relative humidity and ambient temperature as evaluation parameters, and selects the RBF neural network model, which can reflect the non-linear influence of multiple factors, as the method to assess the degree of insulator fouling, and establishes a real-time data-based operating insulator fouling assessment model. Training data and the model structure are important for the accuracy of the fouling assessment model. Therefore, 440 sets of simulated data are selected as the training data for the model simulation, and the model structure with the best fit to the training results is compared by changing the model parameters. This paper presents a real-time fouling assessment model for transmission line insulators of 110–500 kV voltage level, which provides the basic data for the subsequent whole life cycle control of insulators.