Open Access

ARTICLE

Parameters Identification for Extended Debye Model of XLPE Cables Based on Sparsity-Promoting Dynamic Mode Decomposition Method

Weijun Wang^1,*, Min Chen¹, Hui Yin¹, Yuan Li²

1 Guiyang Power Supply Bureau, Guizhou Power Grid Co., Ltd., Guiyang, 550004, China
2 College of Electrical Engineering, Sichuan University, Chengdu, 610065, China

* Corresponding Author: Weijun Wang. Email:

Energy Engineering 2023, 120(10), 2433-2448. https://doi.org/10.32604/ee.2023.028620

Received 22 December 2022; Accepted 10 March 2023; Issue published 28 September 2023

Abstract

To identify the parameters of the extended Debye model of XLPE cables, and therefore evaluate the insulation performance of the samples, the sparsity-promoting dynamic mode decomposition (SPDMD) method was introduced, as well the basics and processes of its application were explained. The amplitude vector based on polarization current was first calculated. Based on the non-zero elements of the vector, the number of branches and parameters including the coefficients and time constants of each branch of the extended Debye model were derived. Further research on parameter identification of XLPE cables at different aging stages based on the SPDMD method was carried out to verify the practicability of the method. Compared with the traditional differential method, the simulation and experiment indicated that the SPDMD method can effectively avoid problems such as the relaxation peak being unobvious, and possessing more accuracy during the parameter identification. And due to the polarization current being less affected by the measurement noise than the depolarization current, the SPDMD identification results based on the polarization current spectral line proved to be better at reflecting the response characteristics of the dielectric. In addition, the time domain polarization current test results can be converted into the frequency domain, and then used to obtain the dielectric loss factor spectrum of the insulation. The integral of the dielectric loss factor on a frequency domain can effectively evaluate the insulation condition of the XLPE cable.

---

Keywords

---