微电网移频式孤岛检测方法研究

发布时间：2018-06-20 19:32

本文选题：孤岛检测 + 移频式孤岛检测算法　；参考：《东北大学》2014年硕士论文

【摘要】：近年来随着分布式发电技术(Distributed Generation-DG)的迅速发展,越来越多的分布式电源被接入电网,微电网以其智能、灵活的控制特点,不仅解决了DG大规模并网的问题,提高了供电可靠性和能源综合利用率,而且满足了负荷增长需求,符合电力可持续发展的趋势。无论微电网处于非计划孤岛状态还是计划孤岛状态,成功检测出孤岛是前提条件,因此孤岛检测技术的研究具有非常重要的现实意义。论文正是在这样的背景下围绕以下几方面展开孤岛检测技术的研究。(1)在研究微电网孤岛特性和相关孤岛检测技术的基础上,得出孤岛系统有功平衡、负载谐振频率等于电网频率以及Q,=2.5为最严重孤岛的结论。(2)对现有孤岛检测方法进行系统分类,着重研究一些典型方法,对比分析了各自的优缺点。深入研究移频式孤岛检测方法的原理,指出移频式孤岛检测失败的根本原因是检测算法产生的频率偏移被负载阻抗角抵消。(3)检测盲区是评估孤岛检测方法的重要指标,针对现有检测盲区描述空间的不足,论文引用了一种新的空间坐标系Qfg×Cnorm,并在该坐标系下研究了滑膜频率偏移法(SMS)、主动频率偏移法(AFD)、带正反馈主动频率偏移法(AFDPF)的盲区分布图,并以AFD和AFDPF的盲区分布图为例,通过仿真验证了在该坐标系下盲区范围的准确性。针对现有移频式孤岛检测方法的缺陷,论文提出了优化AFDPF算法,并对其进行了仿真研究。(4)引入谐波电流大小是孤岛检测方法有效性评估的另一重要指标,论文研究了AFD和优化AFDPF对电能质量的影响,分析了移频式孤岛检测算法中的参数与引入电流畸变率增量ATHD之间的关系。针对电网正常频率波动,导致算法引入的ATHD增加,论文提出了一种改进的低畸变率移频式孤岛检测算法,并通过仿真对比验证了改进算法的优越性。(5)为了研究在多逆变器并联运行情况下,各逆变器检测性能之间的相互影响,本文首次在Qfg×Cnorm坐标系下研究了被动式孤岛检测法和移频式孤岛检测法检测性能之间的稀释作用,同时研究了频率偏差△f的检测误差对孤岛检测性能的影响,并通过Matlab/Simulink软件搭建的仿真模型进行了理论验证。
[Abstract]:In recent years, with the rapid development of distributed Generation Technology (DG), more and more distributed power sources are connected to the power grid. Microgrid, with its intelligent and flexible control characteristics, not only solves the problem of large-scale DG grid connection. The reliability of power supply and the comprehensive utilization rate of energy are improved, and the demand of load increase is satisfied, which is in line with the trend of sustainable development of electric power. Whether the microgrid is in the state of unplanned islanding or planned islanding, the successful detection of islanding is a prerequisite, so the research of islanding detection technology is of great practical significance. On the basis of studying the characteristics of islanding and the related islanding detection technology in microgrid, the active power balance of islanding system is obtained. The conclusion that the load resonance frequency equals the power grid frequency and the conclusion that QT 2.5 is the most serious isolated island is a systematic classification of the existing islanding detection methods. Some typical methods are emphatically studied and their advantages and disadvantages are compared and analyzed. The principle of frequency-shifted island detection method is deeply studied. It is pointed out that the fundamental reason of the detection failure is that the frequency offset generated by the detection algorithm is offset by the load impedance angle, and the blind area is an important index to evaluate the detection method of the isolated island. In this paper, a new spatial coordinate system, Qfg 脳 Cnormn, is used to study the blind area distribution of synovium frequency migration method, active frequency migration method, and active frequency migration method with positive feedback, aiming at the shortcomings of the existing detection blind area description space, in which a new spatial coordinate system Qfg 脳 Cnormnormal is used to study the blind area distribution of the synovial frequency migration method, the active frequency migration method and the active frequency migration method with positive feedback. The blind area distribution of AFD and AFDPF is taken as an example, and the accuracy of the blind area in the coordinate system is verified by simulation. Aiming at the defects of the existing detection methods of isolated island with frequency shift, this paper proposes an optimized AFDPF algorithm, and makes a simulation study on it. The introduction of harmonic current is another important index to evaluate the effectiveness of the islanding detection method. In this paper, the influence of AFD and AFDPF on power quality is studied, and the relationship between the parameters in the frequency-shifted islanding detection algorithm and the introduction of current distortion rate increment ATHD is analyzed. Aiming at the increase of ATHD caused by the normal frequency fluctuation of power system, an improved low distortion rate frequency shift isolated island detection algorithm is proposed in this paper. The superiority of the improved algorithm is verified by simulation. In order to study the interaction between the detection performance of the multi-inverter parallel operation, In this paper, the dilution effect between passive islanding detection method and frequency-shifted islanding detection method is studied for the first time in Qfg 脳 Cnorm coordinate system, and the influence of detection error of frequency deviation f on the detection performance of isolated island detection is also studied. The simulation model built by Matlab / Simulink software is used to verify the theory.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM727

【相似文献】