基于小波神经网络的小电流接地故障选线方法研究

发布时间：2018-04-21 14:01

本文选题：小电流接地 + 故障选线　；参考：《山东理工大学》2014年硕士论文

【摘要】：在我国6~35kV供电系统中大多采用中性点不接地系统或经消弧线圈接地系统，通常称为小电流接地系统。输电线路发生单相接地故障的几率非常高。在小电流接地系统中发生单相接地故障时，由于系统的三相线电压仍然对称，因此，，运行规程规定允许单相接地继续运行一到两个小时。由于单相接地时故障电流较小，而且影响因素较多，使得在小电流接地系统中的单相接地故障检测正确率往往比较低。因此，研究有效提高单相接地保护的选择性是非常必要的。因此，如何实现快速精确检测单相接地故障就成为小电流接地系统提高供电可靠性的一个重要课题。对于小电流系统的单相接地故障选线问题，本文在查阅了国内外诸多文献的基础上，总结了现有的多种选线方法，分析了各种选线方法的优缺点。提出了中性点不接地系统和中性点经消弧线圈接地系统出现的稳态信号和暂态信号特征，为研究单相故障检测方法提供了理论基础，分析得出了利用暂态信号进行故障检测的新方法。本文在分析研究了小波变换理论及小波包分解理论基础上，提出了利用小波分析技术中的小波包分解对小电流系统进行故障选线算法，并给出了选线算法及流程。同时分析了神经网络理论，提出了基于小波神经网络的单相故障检测方法，并给出了故障诊断流程，使单相接地故障选线的正确动作率大大提高。通过理论分析和仿真验证，这两种小电流接地故障选线方法，能够提高故障选线的选择性。利用MATLAB中的SIMULINK仿真平台搭建了一个四出线的中性点经消弧线圈接地的小电流系统模型，对接地故障进行了基于小波分析技术和小波神经网络技术的选线仿真，得出了利用小波分析技术和小波神经网络技术进行小电流系统接地故障检测选线具有较高的选择性。本论文对研究小电流接地故障选线方法具有一定的参考价值。
[Abstract]:The neutral ungrounded system or the arc-suppression coil grounding system are mostly used in the 6~35kV power supply system in our country, which is usually called the small current grounding system. The probability of single-phase grounding fault on transmission line is very high. When the single-phase grounding fault occurs in the low-current grounding system, the three-phase line voltage of the system is still symmetrical, so the operation rules allow the single-phase grounding to continue to run for one to two hours. Because the fault current is small and there are many factors affecting the fault detection in single-phase grounding system, the accuracy rate of single-phase grounding fault detection in low-current grounding system is often low. Therefore, it is necessary to study how to improve the selectivity of single-phase grounding protection. Therefore, how to detect single-phase grounding fault quickly and accurately has become an important issue in improving the reliability of power supply in small current grounding system. For the problem of single-phase grounding fault line selection in small current system, based on the reference of many literatures at home and abroad, this paper summarizes the existing methods of line selection, and analyzes the advantages and disadvantages of various methods of line selection. The characteristics of steady state signal and transient signal of neutral point ungrounded system and neutral point grounding system through arc-suppression coil are presented, which provides a theoretical basis for the study of single-phase fault detection method. A new method of fault detection using transient signals is obtained. Based on the analysis of wavelet transform theory and wavelet packet decomposition theory, this paper proposes a fault line selection algorithm for small current system using wavelet packet decomposition in wavelet analysis technology, and gives the route selection algorithm and flow chart. At the same time, the theory of neural network is analyzed, and the single-phase fault detection method based on wavelet neural network is put forward, and the fault diagnosis flow is given, which greatly improves the correct action rate of single-phase grounding fault line selection. Through theoretical analysis and simulation, it is proved that these two methods can improve the selectivity of fault line selection. Based on the SIMULINK simulation platform in MATLAB, a four-line neutral point grounded by arc-suppression coil is built, and the fault line selection simulation based on wavelet analysis and wavelet neural network is carried out. It is concluded that using wavelet analysis technology and wavelet neural network technology to detect grounding fault line of small current system has high selectivity. This paper has certain reference value to study the method of small current grounding fault line selection.
【学位授予单位】：山东理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM862;TP18

【参考文献】