小电流接地系统故障选线与定位的研究

发布时间：2019-05-26 18:23

【摘要】：随着经济的高速发展,电能作为清洁环保的二次能源改变着我们的生活。近些年来电力行业的发展突飞猛进,电网构架复杂多变,装机容量不断增加,并且随着特高压输电线路的全面建成,输电线路在发生故障后,各行各业都会受到不同程度的影响。目前,我国的配电网广泛采用小电流接地方式,当该系统发生接地故障时,及时准确选出故障线路是快速恢复供电的关键。本文分析了系统发生接地故障的特征,分别从稳态特征和暂态特征方面深入研究。首先,详细介绍基于稳态分量和暂态分量的选线方法,如比幅比相法、五次谐波法、零序导纳法、能量法、小波分析法。通过仿真验证并总结每种选线方法的适用范围和局限性,得出单一选线方法存在不足。针对单一选线无法实现适应各种单相接地故障的情况,本文提出基于极限学习机的融合选线方法。首先在Matlab中建立110/35kV的小电流系统模型,用小波包变换算法提取特征分量。采用零序导纳法、自适应陷波法、改进锁相环的五次谐波法分析在不同接地系统、不同接地电阻等故障类型发生故障时的暂态特征,利用极限学习机方法进行融合选线,并与其他融合选线方法进行对比。输电网电压等级越来越高,输送容量也在不断增加,输电线路出现故障后,及时准确的找到故障点不仅可以减轻运检人员的工作量,而且可以及时恢复线路供电。本文通过单端测距原理利用电压反向行波实现定位,在Matlab中搭建小电流系统模型,利用Hear小波3层分解获取电压反向行波,通过大量数据可以验证该方法定位具有较高的准确性,计算绝对误差和相对误差并归纳出现误差的因素。
[Abstract]:With the rapid development of economy, electric energy, as a clean and environmentally friendly secondary energy, has changed our lives. In recent years, with the rapid development of the power industry, the power grid structure is complex and changeable, the installed capacity is increasing, and with the comprehensive completion of UHV transmission lines, transmission lines after failure, All kinds of industries will be affected to varying degrees. At present, the low current grounding mode is widely used in the distribution network of our country. When the grounding fault occurs in the system, the timely and accurate selection of the fault line is the key to the rapid recovery of power supply. In this paper, the characteristics of grounding fault in the system are analyzed, and the steady-state and transient characteristics are deeply studied. Firstly, the line selection methods based on steady and transient components, such as amplitude ratio phase method, fifth harmonic method, zero sequence admittance method, energy method and wavelet analysis method, are introduced in detail. Through simulation verification and summary of the applicable scope and limitations of each line selection method, it is concluded that the single line selection method has shortcomings. In view of the fact that a single line selection can not adapt to all kinds of single-phase grounding faults, a fusion line selection method based on limit learning machine is proposed in this paper. Firstly, the small current system model of 110/35kV is established in Matlab, and the feature components are extracted by wavelet packet transform algorithm. The zero sequence admittance method and adaptive notch method are used to analyze the transient characteristics of fault types in different grounding systems and different grounding resistance by improving the quintic harmonic method of phase-locked loop, and the limit learning machine method is used to fuse and select the line. It is compared with other fusion line selection methods. The voltage level of transmission network is getting higher and higher, and the transmission capacity is also increasing. After the fault of transmission line, finding the fault point in time can not only reduce the workload of operators, but also restore the power supply of the line in time. In this paper, based on the principle of single-terminal ranging, the voltage reverse traveling wave is realized by using the voltage reverse traveling wave, the small current system model is built in Matlab, and the voltage reverse traveling wave is obtained by using the three-layer decomposition of Hear wavelet. Through a large number of data, it can be verified that the method has high accuracy, calculates absolute error and relative error, and sums up the factors of error.
【学位授予单位】：太原科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM862

【参考文献】