35kV及以下XLPE电力电缆绝缘老化评估研究

发布时间：2018-05-01 08:21

本文选题：XLPE电缆 + 多参数检测　；参考：《上海交通大学》2014年博士论文

【摘要】：上海气候温和湿润,是一个多地表水的城市,且许多电缆已经运行了20年左右,处于事故多发期,研究中低压电缆在特征环境下的老化状态显得非常重要。本论文通过水树枝观测、介电谱测试、FTIR光谱、拉伸试验、电老化试验、热重分析、差示扫描量热、热老化试验以及加速水树老化试验等,对电缆绝缘老化及其检测方法进行研究。提出水树含量与最大水树枝长度、高频介损峰值与羰基指数、拉伸强度与断裂伸长率、累积击穿强度、活化能、基于等效热历史参数的新参数ln(t/τT)等多个现场老化电缆绝缘老化状态检测量,用于表征电缆水树化程度、热氧老化程度、绝缘降解程度、耐受电老化能力、绝缘热分解难易程度、绝缘老化状态等,以全面反映电缆绝缘中可能存在的各种老化形态。对不同老化状态电缆样品的各检测数据进行了方差分析,结果表明上述检测量均可对不同老化程度的现场老化电缆进行标识。此外,通过水树枝观测发现了大量典型的领结形水树枝,说明水树老化是上海中低压电缆的典型现场老化特征；对电缆径向不同位置处的绝缘进行了红外分析,结果表明氧气是导致电缆热氧老化的主要因素；对热重分析数据进行了热降解动力学分析与回归分析,结果表明60%适合作为Ozawa法(或Toop法)求取交联聚乙烯电缆绝缘热分解活化能时失重百分数的选择标准,20℃/min下的Coast-Redfern法(机理S9)可作为大量现场老化电缆样品绝缘热分解活化能的快速测定方法；加速水树老化试验结果表明,现场运行电缆的水树化过程是加速进行的。运用上述检测方法对上海市近20年内、约300条线路466根现场老化电缆样品进行了实验室理化分析,对每两项检测量进行了回归分析,结果表明：水树枝观测以及拉伸试验是有效且可靠的；水树化程度越高,力学性能倾向于变差；证实固体材料的微观不完整性增大会提高介质的击穿强度；电缆绝缘的现场老化过程是加速进行的；证实热氧化老化会抑制电缆径向的水树生长；绝缘热氧化老化过程伴随着XLPE交联网络的降解；ln(t/τT)能够有效表征现场老化电缆绝缘老化降解程度。对每项检测量进行了T-检验、方差分析,分析了敷设方式、采样区段等对电缆老化降解程度的影响,结果表明：敷设方式对羰基指数与拉伸强度均有显著性的影响,排管电缆比直埋电缆更易发生热氧化,而直埋电缆比排管电缆更易发生绝缘老化降解；能够体现采样区段以及敷设区域显著性影响的检测量只有传统功能性参数拉伸强度和累积击穿强度,并且基于两个检测量的方差分析结果一致,其可靠性得到了进一步的证明。选取回归分析中相关度高、方差分析和T检验中能可靠体现差异性,并且结果合理的检测量作为特征变量,包括水树含量、拉伸强度、累积击穿强度和活化能。将高度相关的两个检测量组合进行二维模糊聚类,所得聚类中心作为特征模糊变量的隶属度函数参数。根据模糊聚类情况与专家经验,建立了诊断规则。采用强度转移法,建立了电缆绝缘模糊聚类诊断模型,实例分析表明,该模型可以很好地反映电缆绝缘的综合老化状态,且电缆绝缘老化状态与投运时间没有必然联系。根据电缆检测数据与绝缘诊断模型,开发了电缆状态评估与老化趋势管理系统软件,并对所有线路电缆样品进行了绝缘诊断,诊断结果的统计分析与电缆状态检测数据统计分析的结论完全一致,验证了模型的可靠性。采用逐步回归法,分析了同一现场运行条件下的正常现场老化电缆样品的绝缘状态检测数据与老化时间的多元线性回归关系。结果显示老化时间与水树含量,累积击穿强度与活化能等三个检测量的多元线性回归关系显著,据此建立了电缆绝缘多变量老化模型,模型参数取决于具体的现场运行条件,对应一定的绝缘降解速率实例分析表明该老化模型是合理的,并且绝缘失效取决于电缆降解速率而非电缆的投运时间。
[Abstract]:Shanghai has a mild and humid climate. It is a city with multi surface water, and many cables have been in operation for about 20 years. It is in a period of frequent accidents. The aging state of low-voltage cables in the study is very important. This paper is based on water branch observation, dielectric spectrum test, FTIR spectrum, tensile test, electrical aging test, thermogravimetric analysis and differential analysis. Scanning calorimetry, thermal aging test and accelerated water tree aging test, the aging of cable insulation and its detection methods were studied. Water tree content and maximum water branch length, high frequency dielectric loss peak and carbonyl index, tensile strength and elongation at break, cumulative breakdown strength, activation energy, new parameter ln (t/ tau T) based on equivalent thermal history parameters were proposed. The measurement of insulation aging state of several field aging cables is used to characterize the degree of cable water tree, the degree of thermal oxygen aging, the degree of insulation degradation, the resistance to electrical aging, the difficulty of insulation thermal decomposition and the aging state of insulation, so as to reflect all kinds of aging forms that may exist in cable insulation. The test data were analyzed by variance. The results showed that the above detection amount could identify the aging cable with different aging degree. In addition, a large number of typical tie shaped water branches were found by water branch observation, indicating that the aging of the water tree is the typical field aging characteristic of the middle and low pressure cables in Shanghai, and the radial direction of the cable is different. The results show that oxygen is the main factor causing the thermal oxygen aging of the cable, and the thermal degradation kinetics analysis and regression analysis of the thermogravimetric analysis data show that 60% is suitable for the selection of the weight loss percentage of the Ozawa method (or Toop method) for the thermal decomposition activation energy of the crosslinked polyethylene cable insulation, 20 The Coast-Redfern method (mechanism S9) under /min can be used as a rapid method for determining the activation energy of insulation thermal decomposition of a large number of field aging cable samples. The results of accelerated water tree aging test show that the water tree process of the field operating cable is accelerated. In the last 20 years, about 300 lines in Shanghai were aged and 466 were aged on site. The results show that water branch observation and tensile test are effective and reliable, and the higher the degree of water tree is, the mechanical properties tend to be worse. It is proved that the micro incompleteness of solid material will increase the breakdown strength of the medium; the cable is absolute. The aging process of the edge is accelerated; it is proved that thermal oxidation aging will inhibit the growth of the radial water tree of the cable; the thermal oxidation aging process of the insulation is accompanied by the degradation of the XLPE crosslinking network; ln (t/ tau T) can effectively characterize the aging degradation degree of the insulation of the field aging cable. The T- test, analysis of variance, and analysis of each measurement are carried out. The effect of laying mode and sampling section on the degradation degree of cable aging shows that the laying method has a significant effect on the carbonyl index and tensile strength, and the pipe cable is more prone to thermal oxidation than the direct buried cable, while the direct buried cable is more prone to insulation degradation than the pipe cable, and can reflect the sampling section and the laying area. The detection amount of the domain saliency is only the tensile strength and the cumulative breakdown strength of the traditional functional parameters, and the reliability is further proved based on the results of the variance analysis of the two tests. The high correlation, the variance analysis and the T test in the regression analysis can reliably reflect the difference, and the result is reasonable. Quantity as a characteristic variable, including the content of water tree, tensile strength, cumulative breakdown strength and activation energy. A two dimensional fuzzy clustering is carried out by combining two highly related tests. The clustering center is used as the membership function parameter of the characteristic fuzzy variables. According to the fuzzy clustering situation and the expert experience, the diagnosis rules are established. The fuzzy clustering diagnosis model of cable insulation is established. The example analysis shows that the model can well reflect the comprehensive aging state of cable insulation, and there is no inevitable connection between the aging state of cable insulation and the time of delivery. The insulation diagnosis of all line cable samples is carried out. The statistical analysis of the diagnosis results is completely consistent with the conclusion of the statistical analysis of the cable state detection data. The reliability of the model is verified by the stepwise regression method, and the insulation state detection data and aging of the samples of the normal field aging cable under the same operation condition are analyzed. The results show that the relationship between the aging time and the content of water tree, the cumulative breakdown strength and the activation energy of the multiple linear regression is significant. According to this, the multi-variable aging model of cable insulation is established. The model parameters are determined by the specific field running pieces, corresponding to the example analysis of the insulation degradation rate. The aging model is reasonable and the insulation failure depends on the degradation rate of the cable rather than the cable's running time.

【学位授予单位】：上海交通大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TM855

【参考文献】