变压器励磁涌流识别方法研究

发布时间：2018-07-09 11:17

本文选题：变压器 + 纵差保护　；参考：《湘潭大学》2014年硕士论文

【摘要】：近些年来，中国国民经济飞速发展，用电量突增，电网规模不断发展和扩大，电压等级不断升高，变压器作为电力系统中的极其重要的传能变压设备，其安全性稳定运行关系着整个电网的供电可靠性，于是对变压器差动保护提出了新的更高的要求。由于变压器产生的励磁涌流会流入变压器纵差保护的差动回路中，有可能导致纵差保护误动作。因此，如何正确有效的识别励磁涌流和内部故障电流是一个重要而困难的问题。论文从以下几方面进行了研究。首先，本文密切结合现场工程实践，开展了变压器励磁涌流对电力系统安全稳定性影响的研究。简要综述了国内外励磁涌流的研究现状，分析了励磁涌流对变压器继电保护带来的不良影响；根据进一步提高变压器励磁涌流识别准确率的要求，阐述了继续研究励磁涌流识别方法具有的理论和实际工程意义。然后，深入研究了变压器差动保护的保护原理以及励磁涌流的产生机理，对常用的励磁涌流识别方法进行了详尽分析，得出了各自的优缺点。对比分析了变压器励磁涌流与内部故障电流的波形特征，进而利用小波理论中的多分辨率分析检测励磁涌流和内部故障电流的时频域特征信息。再次，在以上分析研究的基础上，结合小波分析和信息熵理论，提出了基于Tsallis小波熵（能量熵和时间熵）的励磁涌流识别方法，该方法是先将电流进行小波分解重构，然后利用熵的概念来定量分析其能量的分布情况。并且采用的Tsallis小波能在很大程度上改善频域混叠和能量泄露的问题，因此，该方法能够准确识别励磁涌流。同时还根据变压器瞬时无功功率特性得到另一种基于瞬时差有功无功比率的涌流识别方法，其主要原理是当变压器出现励磁涌流时，变压器两侧的有功功率差变化比较小，，而无功功率差变大相对较大，此方法也能有效识别励磁涌流。最后，通过Matlab/Simulink搭建励磁涌流仿真模型，利用上述两种识别方法对单相三绕组变压器和三相三绕组变压器在各种故障条件下进行了仿真验证，比如空载合闸、单相接地短路以及故障切除恢复供电等等，相应的实例仿真结果表明，在系统各种运行工况下，所提方法都具有良好的识别效果，从而验证所提两种识别方法的可行性和灵敏性，克服了以往保护的不足，提高了变压器纵差保护的保护性能。
[Abstract]:In recent years, with the rapid development of China's national economy, the rapid increase in electricity consumption, the continuous development and expansion of the scale of the power grid, and the constant increase in voltage levels, transformers are regarded as the most important power transmission and transformer equipment in the power system. Its safe and stable operation is related to the power supply reliability of the whole power network, so a new and higher requirement for transformer differential protection is put forward. The inrush current generated by the transformer will flow into the differential circuit of the differential protection of the transformer, which may lead to the misoperation of the differential protection. Therefore, how to correctly and effectively identify inrush current and internal fault current is an important and difficult problem. The thesis has carried on the research from the following several aspects. Firstly, the paper studies the influence of transformer inrush current on power system safety and stability. This paper briefly summarizes the research status of inrush current at home and abroad, analyzes the adverse effect of inrush current on relay protection of transformer, and further improves the accuracy of inrush current identification of transformer. This paper expounds the theoretical and practical engineering significance of continuing to study the method of inrush current identification. Then, the principle of transformer differential protection and the generation mechanism of excitation inrush current are deeply studied, and the common methods of inrush current identification are analyzed in detail, and their advantages and disadvantages are obtained. The waveform characteristics of inrush current and internal fault current of transformer are compared and analyzed, and the time-frequency characteristic information of excitation inrush current and internal fault current is detected by multi-resolution analysis in wavelet theory. Thirdly, on the basis of the above analysis, combined with wavelet analysis and information entropy theory, a Tsallis wavelet entropy (energy entropy and time entropy) based inrush current identification method is proposed. Then the concept of entropy is used to quantitatively analyze its energy distribution. And the Tsallis wavelet can improve the frequency domain aliasing and energy leakage to a great extent, so this method can accurately identify the inrush current. At the same time, according to the instantaneous reactive power characteristics of the transformer, another inrush current identification method based on the instantaneous difference between active and reactive power ratio is obtained. The main principle is that the difference of active power on both sides of the transformer is relatively small when the inrush current occurs in the transformer. However, the reactive power difference is relatively large, and this method can also effectively identify the inrush current. Finally, the simulation model of inrush current is built by Matlab / Simulink, and the simulation of single-phase three-winding transformer and three-phase three-winding transformer under various fault conditions, such as no-load closing, is carried out using the above two identification methods. The simulation results of single-phase grounding short circuit and fault resetting and restoring power supply show that the proposed method has a good recognition effect under various operating conditions of the system. Thus, the feasibility and sensitivity of the two identification methods are verified, the shortcomings of the previous protection are overcome, and the protection performance of the transformer differential protection is improved.
【学位授予单位】：湘潭大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM774

【参考文献】