摘要详情

ID / 提交时间

143 / 2023-10-19 17:06:26

标题

Non-destructive Testing Method for Breakdown Strength of XLPE Cable Insulation: Near-infrared Spectroscopy

关键字

cable insulation,cross-linked polyethylene,near-infrared spectroscopy,thermo-oxidative ageing,breakdown strength

主题及专题

4.High voltage and insulation technology > (1)、High voltage testing and measurement

状态

全文录用

作者

Xuyang Zhao / Zhejiang Huadian Equipment Testing and Research Institute Co., Ltd.；SGCC-Testing Technology Laboratory of Electrical Equipment Safety Performance

Shengyi Xie / Zhejiang Huadian Equipment Testing and Research Institute Co., Ltd.；SGCC-Testing Technology Laboratory of Electrical Equipment Safety Performance

Fangfang Wu / Zhejiang Huadian Equipment Testing and Research Institute Co., Ltd.；SGCC-Testing Technology Laboratory of Electrical Equipment Safety Performance

Yibo Gao / Zhejiang Huadian Equipment Testing and Research Institute Co., Ltd.；SGCC-Testing Technology Laboratory of Electrical Equipment Safety Performance

Zhigang Ren / State Key Laboratory of Electrical Insulation and Power Equipment, Xi’an Jiaotong University

Haoran Sui / State Key Laboratory of Electrical Insulation and Power Equipment, Xi’an Jiaotong University

Haoyue Jiang / State Key Laboratory of Electrical Insulation and Power Equipment, Xi’an Jiaotong University

Jianying Li / State Key Laboratory of Electrical Insulation and Power Equipment, Xi’an Jiaotong University

摘要

With the rising demand for power cables due to the development of urbanization, it becomes more and more important for full overage sampling to prevent the cables with poor quality from the grid. However, the traditional testing methods are incompetent due to their long period and high cost. To address the problem, in this paper, the non-destructive testing method based on near-infrared (NIR) spectroscopy for breakdown strength of cross-linked polyethylene (XLPE) cable insulation was put forward. The NIR spectra of XLPE specimens with different ageing conditions were obtained and preprocessed by smoothing and second derivation. The competitive adaptive reweighted sampling (CARS) was employed to achieve dimensionality reduction and partial least squares regression (PLSR) was used to establish model for predicting the breakdown strength. The results illustrated that only 2.5% error was presented in the prediction. This work proposed a new method to achieve the non-destructive and accurate testing method for cable insulation.