引用本文:徐志钮,张翼,胡志伟,胡世勋,律方成,金虎.基于两步优化的介电谱特征参数辨识方法及其应用[J].电力自动化设备,2019,39(7):
XU Zhiniu,ZHANG Yi,HU Zhiwei,HU Shixun,Lü Fangcheng,JIN Hu.Dielectric spectrum characteristic parameter identification method based on two-step optimization and its application[J].Electric Power Automation Equipment,2019,39(7):
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 4184次   下载 1919  
基于两步优化的介电谱特征参数辨识方法及其应用
徐志钮1, 张翼1, 胡志伟1, 胡世勋1, 律方成1, 金虎2
1.华北电力大学 电气与电子工程学院,河北 保定 071003;2.南方电网科学研究院有限责任公司,广东 广州 510663
摘要:
为了能够通过拟合介电谱的方式获得绝缘材料的绝缘状态,在Cole-Cole模型下提出了基于两步优化的绝缘材料介电谱特征参数辨识方法。根据绝缘材料的普适弛豫定律,推导形成了一套较为完整的Cole-Cole模型初值获取方法;采用与拟合误差情况更匹配的目标函数,提出基于全局搜索与局部搜索相结合的两步优化对各特征参数进行优化计算,有效地解决了传统辨识方法容易陷入局部极小点甚至发散的问题。运用3种典型的介电谱特征参数辨识方法和所提方法,对数值产生和实测的介电谱进行计算,验证了所提方法的优越性。实验测量了低密度聚乙烯(LDPE)样本直流加压注入空间电荷后的介电谱,运用所提方法辨识不同去极化时刻LDPE介电谱的特征参数。结果表明:不同去极化时间下,通过所提方法均能较好地逼近实测介电谱,且随着频率的升高,复介电常数的实部、虚部的减小速度存在下降的趋势。
关键词:  绝缘  介电谱  Cole-Cole模型  特征参数辨识  两步优化  低密度聚乙烯
DOI:10.16081/j.issn.1006-6047.2019.07.024
分类号:TM21
基金项目:中央高校基本科研业务费专项资金资助项目(2019MS090,2019MS085,2017MS196)
Dielectric spectrum characteristic parameter identification method based on two-step optimization and its application
XU Zhiniu1, ZHANG Yi1, HU Zhiwei1, HU Shixun1, Lü Fangcheng1, JIN Hu2
1.School of Electrical and Electronic Engineering, North China Electric Power University, Baoding 071003, China;2.Electric Power Research Institute of GSG, Guangzhou 510663, China
Abstract:
In order to obtain the insulation condition of insulating materials by the way of dielectric spectrum fitting, based on two-step optimization and Cole-Cole model, a dielectric spectrum characteristic parameter identification method is proposed for insulating materials. According to the universal relaxation law of insulating materials, a relatively complete initial value acquisition method of Cole-Cole model is formed by derivation. The objective function that better consistent with fitting error is adopted. The two-step optimization combining global search and local search is proposed and used to optimize the characteristic parameters, which effectively solves the problem that the traditional identification methods easily fall into the local minima or even diverge. Three existing typical methods and the proposed method are used to calculate the numerically generated and measured dielectric spectra, and the superiority of the proposed method is verified by the calculation results. The dielectric spectrums of LDPE(Low-Density PolyEthylene) samples injected with space charge by DC voltage are measured, and the characteristic parameters of LDPE dielectric spectrum at different depolarization times are identified by the proposed method. The results reveal that all the measured dielectric spectra with different depolarization times can be fitted well by the proposed method, and the descent speeds of real and imaginary parts of complex permittivity decrease with the increase of frequency.
Key words:  insulation  dielectric spectrum  Cole-Cole model  characteristic parameter identification  two-step optimization  low-density polyethylene

用微信扫一扫

用微信扫一扫