Composition Analysis of Operating Insulator Contamination Based on Hyperspectral Imaging Technology

Xia, Changjie; Guo, Chenxi; Ren, Ming; Dong, Ming; Wang, Bin; Wang, Siyun; Duan, Ran

doi:10.1109/ceidp47102.2019.9009921

Cited by 4 publications

(1 citation statement)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Qiu et al [7] first combined HSI and extreme-learning-machine-based classification method for ESDD detection on the insulators surface, achieving an impressive classification accuracy of over 87.5%. Xia et al [5,[15][16][17] and Yin et al [18,19] further employed characteristic bands extraction algorithms and machine-learningbased classification algorithms to facilitate ESDD recognition, which indicated that the narrower swath width in HSI allows for a more precise level of discrimination and detection. Ma et al [20] firstly put forward an MC estimation method for pollution layer using reflection spectra and back propagation neural network.…”

Section: Introductionmentioning

confidence: 99%

Visual characterization of pollution and moisture parameters of the polluted HTV silicone rubber surface with hyperspectral imaging technology

Xia

Ren

Wang

et al. 2023

J. Phys. D: Appl. Phys.

Self Cite

View full text Add to dashboard Cite

The application of high temperature vulcanized (HTV) silicone rubber materials in transmission lines insulators has been extensive. However, with the accumulation of pollution and moisture on the surface of HTV silicone rubber, the insulation reliability gradually decreases, which may even lead to flashover. In this paper, hyperspectral imaging technology (HSI) is utilized as a rapid and effective tool for the visual characterization of pollution and moisture parameters of the polluted HTV silicone rubber. Based on the analysis of reflection spectra, it has been determined that the pollution degree has a positive effect and the moisture content has a negative effect on the reflection characteristics of HTV silicone rubber, respectively. Further quantitative analysis of reflectance values at various moisture conditions reveals that exponential relationships can be observed between reflectance values and moisture contents at a certain wavelength. At the same time, linear correlations between reflectance values and pollution degrees (quantified as equivalent salt deposit density, ESDD) at a certain wavelength are also proposed. By integrating the exponential and linear relationships, a double-exponential model is formulated to concurrently characterize the ESDD and moisture content (MC). Furtherly, from the perspective of engineering application, to enhance the model’s robustness, a set of equations combining spectral indexes and spectral shape feature parameters is initially formulated for the calculating of ESDD and MC. Given the synchronous inversion and visualization capabilities of pollution and moisture parameters of the polluted HTV silicone rubber surface, the visual characterization model serves as a valuable tool for protecting against pollution flashover failures and designing power equipment insulation.

show abstract