覆冰绝缘子闪络预警特征参量及方法研究

发布时间：2018-03-01 01:24

本文关键词： 复合绝缘子覆冰泄漏电流冰闪电压评估紫外光子　出处：《重庆大学》2014年硕士论文　论文类型：学位论文

【摘要】：绝缘子是电力系统中使用量最多的器件，冰灾事故的频繁发生导致输电线路绝缘子覆冰，严重威胁电力系统的安全稳定运行。目前国内外开展了大量绝缘子覆冰的特性研究，但是绝缘子覆冰闪络预警却鲜有研究，目前尚未提出统一的评判覆冰绝缘子运行状态的有效特征量。因此本文对110kV复合绝缘子进行了大量覆冰试验，采集了不同覆冰状态下复合绝缘子融冰过程中运行电压下的泄漏电流以及紫外光子数，提取了能准确反映绝缘子运行状态的泄漏电流和紫外光子特征参量，给出了特征参量同盐密和覆冰量之间的关系，建立了两种评估模型预测冰闪电压，提出了覆冰绝缘子闪络分级预警方案。本文主要的结论为： ①本文对110kV复合绝缘子进行了覆冰试验，共模拟了5种覆冰程度、6种盐密共30种覆冰状态，分析了融冰过程中泄漏电流的变化过程，研究融冰过程中运行电压下泄漏电流同覆冰程度以及盐密之间的关系，提取了有效反映绝缘子覆冰状态的泄漏电流特征参量，得到特征参量同覆冰状态之间的关系。 ②本文采集了110kV复合绝缘子在不同覆冰程度以及盐密下的紫外光子数，分析了融冰过程中紫外光子变化过程，提取了能有效反应放电程度的特征参量：紫外光子数平均值N，给出了紫外光子平均值与盐密和覆冰量之间的关系，并进行了闪络电压试验，得到了不同盐密和覆冰量下的闪络电压值，，拟合了闪络电压同盐密和覆冰量之间的关系。 ③本文提出了两种覆冰绝缘子闪络电压评估模型，分别是广义神经回归网络以及最小支持向量机评估模型，对模型进行训练，并通过实验验证了模型的准确性，结果表明两种模型的最大误差分别为9.04%以及9.24%。提出了覆冰绝缘子覆冰闪络分级预警方法。
[Abstract]:Insulators are the most widely used devices in power system. The frequent occurrence of ice disaster causes the insulators of transmission lines to be covered with ice, which seriously threatens the safe and stable operation of power system. At present, a large number of research on the characteristics of insulator icing has been carried out at home and abroad. However, there is little research on the ice flashover warning of insulators. At present, there is no uniform effective characteristic quantity for judging the operation state of icing insulators. Therefore, a large number of icing tests have been carried out on 110 kV composite insulators in this paper. The leakage current and the number of ultraviolet photons in the operation voltage of the composite insulator in the process of melting the ice under different icing conditions were collected, and the leakage current and the characteristic parameters of the ultraviolet light which could accurately reflect the running state of the insulator were extracted. The relationship between characteristic parameters and salt density and ice content is given, two evaluation models are established to predict the lightning pressure of ice, and an ice insulator flashover classification early warning scheme is proposed. The main conclusions of this paper are as follows:. 1 in this paper, the icing test of 110 kV composite insulator is carried out, and the ice state of 5 kinds of icing degree and 6 kinds of salt density are simulated, and the variation process of leakage current during melting process is analyzed. The relationship between leakage current and icing degree and salt density in the process of melting ice is studied. The characteristic parameters of leakage current which effectively reflect the ice state of insulator are extracted, and the relationship between the characteristic parameters and the ice state of insulator is obtained. In this paper, the ultraviolet photons of 110 kV composite insulators under different icing degree and salt density are collected, and the variation process of ultraviolet photons during melting ice is analyzed. In this paper, the characteristic parameters of the discharge degree of effective reaction are extracted: the average number of ultraviolet photons N, the relationship between the average value of ultraviolet photons and salt density and ice content, and the flashover voltage test is carried out. The flashover voltage values under different salt density and ice content are obtained, and the relationship between flashover voltage and salt density and ice content is fitted. In this paper, two kinds of evaluation models of flashover voltage of icing insulator are proposed, which are generalized neural regression network and minimum support vector machine evaluation model. The model is trained, and the accuracy of the model is verified by experiments. The results show that the maximum errors of the two models are 9.04% and 9.24.The method of ice flashover classification early warning for icing insulators is proposed.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM216

【参考文献】