摘要
在极化去极化电流测试中,对去极化电流进行解谱可以有效评估电缆绝缘状态,针对传统方法中去极化电流的子谱线总数和对应谱线参数难以求解的问题,提出一种基于振型幅值法结合随机子空间识别的去极化电流解谱方法。该方法首先假定去极化电流中子谱线总数,计算各子谱线的振型幅值,相比虚假子谱线,真实子谱线的振型幅值比较大,因此振型幅值图可以准确反映去极化电流中真实子谱线的数目;然后根据真实子谱线数目结合随机子空间识别算法计算各子谱线参数,随机子空间识别算法可以在无需人为干预的情况下准确实现子谱线的参数识别,进而完成整个去极化电流解谱操作;最后,利用该方法对仿真和实测的去极化电流数据进行解谱测试,测试结果表明,相比传统的奇异值法和微分法,该方法能更加准确地对去极化电流进行解谱,进而为有效评估电力电缆绝缘提供重要依据。
In the polarization and depolarization current(PDC)test,the depolarization current spectrum decomposition can be used to assess the insulation state of power cable. In the traditional methods,it is difficult to get the total number and solve the corresponding parameters of the sub-spectral lines of the depolarized current, so a depolarized current spectrum decomposition method on the basis of the combination of the vibration mode amplitude method and stochastic subspace identification(SSI)is proposed. The total number of sub-spectral lines in depolarization current is assumed. The vibration mode amplitudes of each sub-spectral line are calculated. In comparison with those of the false sub-spectral lines,the vibration mode amplitudes of the real sub-spectral lines are larger,so the vibration mode amplitude diagram can accurately present the number of the real sub-spectral lines in the depolarization current. According to the number of the real sub-spectral lines,the parameters of each sub-spectral line are calculated in combination with the SSI algorithm. The SSI algorithm can accurately identify the parameters of each sub-spectral line without human intervention,so as to complete the whole operation of the depolarization current spectrum decomposition. This method is used to perform spectrum decomposition test on the simulated and measured depolarization current data. The test results show that,in comparison with the traditional singular value method and differential method,the proposed method can more accurately perform spectrum decomposition on the depolarization current,so it provides an important basis for the effective evaluation of power cable′s insulation performance.
作者
周凯
王亚兰
ZHOU Kai;WANG Yalan(College of Electrical Engineering,Sichuan University,Chengdu 610065,China)
出处
《现代电子技术》
2022年第1期136-141,共6页
Modern Electronics Technique
基金
国家自然科学基金资助项目(51477106)。
关键词
去极化电流解谱
随机子空间识别
振型幅值法
电力电缆
子谱线参数计算
测试分析
depolarization current spectrum decomposition
SSI
vibration mode amplitude method
power cable
sub-spectral line parameter calculation
testing analysis