分布式并网发电系统孤岛检测研究

发布时间：2018-01-19 01:28

本文关键词： 分布式发电系统孤岛检测频率偏移检测盲区　出处：《湘潭大学》2014年硕士论文　论文类型：学位论文

【摘要】：工业经济的高速发展促使人类社会加快了对能源的开采和利用，而随之而来的能源问题也逐渐被社会所关注。能源的枯竭和环境问题的凸显，迫使各国在不同程度上加大了对新能源技术的投资力度。而分布式供电技术的发展，已成为新能源技术的开发和利用中不可或缺的环节。在分布式发电技术里，一般将风力发电和光伏发电作为最主要的电源，其供电技术的研发也较为成熟。但分布式供电系统与周围负载进行并网运行时，却存在一定的风险。其中之一便是分布式系统与电网连接端口因突发状况而断开，分布式发电系统未能及时检测到此现象的发生，并继续向本地用户输送电。这种运行方式被称为孤岛运行。这种计划之外的孤岛运行会对电力检修人员或者电力设备造成极大的危害。所以，国际标准中都明确规定并网运行的分布式供电系统中必须具备孤岛检测功能。本文主要研究分布式并网系统中的孤岛检测方法，对分布式发电技术的概念，孤岛现象和孤岛发生的机理、分类进行了阐述。检测方式的参数优化和缩小盲区等方面做了大量分析。目前热点方法进行了分类总结，对比各种方法的优缺点，并针对其优缺点选定PFAFD法进行优化。传统的PFAFD检测法是基于最基础的AFD检测法上加以前馈而来，目前对于PFAFD法的研究相对成熟，有大量的经验可以借鉴。但传统的PFAFD法存在着较多不确定性，最主要体现在比如算法对检测盲区的影响较大；初始扰动值c f0和正反馈增益k的选取对于电能质量的干扰较为突出，负载各项参数的优化对于检测速度的提高这几点。而本文正是针对以上几个主要问题，利用Q f0Cnorm盲区描述方法来缩小PFAFD检测法的检测盲区。通过数学逻辑推导以确定扰动频率值和负载电阻值以提高PFAFD检测法的检测速。最后利用MATLAB/SIMULINK软件搭建分布式并网系统的仿真模型，对本文提出的孤岛检测法进行仿真，验证实验结果。同时，为了增加对分布式并网系统中非计划的孤岛运行的检测力度，降低其漏检的可能性，加之PFAFD法对频率这一关键参数的检测便利，，利用STC12C5A60S2型号单片机，采用PWM逆变技术对孤岛效应进行模拟，同时使用IR2110芯片对IGBT进行驱动。通过对PWM波频率和宽度的控制来改变电压、频率和相位从而模拟孤岛效应的发生，得到理论与实际相结合的研究成果。从而在PFAFD算法之外加之频率被动检测法，借此对系统进行二次保护。并搭建硬件的搭建验证其被动检测法的有效性。
[Abstract]:With the rapid development of industrial economy, the human society has accelerated the exploitation and utilization of energy, and the energy problem has gradually been concerned by the society. The development of distributed power supply technology has become an indispensable link in the development and utilization of new energy technology. In the distributed generation technology, wind power generation and photovoltaic power generation are generally regarded as the most important power supply, and the research and development of its power supply technology is more mature, but the distributed power supply system is connected with the surrounding load. But there are some risks. One of them is that the connection port between distributed system and power grid is broken because of the sudden situation, and the distributed generation system can not detect this phenomenon in time. And continue to deliver electricity to local users. This mode of operation is called islanding operation. This unplanned island operation can do great harm to electrical maintenance personnel or power equipment. In the international standards, it is clearly stipulated that the distributed power supply system connected to the grid must have the islanding detection function. This paper mainly studies the islanding detection method in distributed grid-connected system, the concept of distributed generation technology, the phenomenon of islanding and the mechanism of islanding. The classification is expounded. The parameter optimization of the detection method and the reduction of blind area are analyzed. The hot spot methods are classified and summarized, and the advantages and disadvantages of each method are compared. The traditional PFAFD detection method is based on the most basic AFD detection method feedforward, and the current research on PFAFD method is relatively mature. There are a lot of experience to learn from, but there are many uncertainties in the traditional PFAFD method, such as the algorithm has a greater impact on the detection of blind areas. The selection of initial disturbance value c f 0 and positive feedback gain k is very important to the interference of power quality. The optimization of the load parameters improves the detection speed, and this paper is aimed at the above several main problems. Using Q. F0Cnorm blind area description method is used to reduce the detection blind area of PFAFD detection method. The disturbance frequency value and load resistance value are determined by mathematical logic derivation to improve the detection speed of PFAFD detection method. Finally, the simulation model of distributed grid-connected system is built by using MATLAB/SIMULINK software, and the islanding detection method proposed in this paper is simulated to verify the experimental results. In order to increase the detection of unplanned island operation in distributed grid-connected system and reduce the possibility of missing detection, the PFAFD method is convenient to detect the key parameter of frequency. The islanding effect is simulated by using STC12C5A60S2 single chip computer and PWM inverter technology. At the same time, the IR2110 chip is used to drive the IGBT. By controlling the frequency and width of the PWM wave, the voltage, frequency and phase are changed to simulate the occurrence of the islanding effect. The research results of combining theory with practice are obtained, and the frequency passive detection method is added to the PFAFD algorithm. In this way, the system is protected twice, and the hardware is built to verify the effectiveness of the passive detection method.
【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】：TM61

【参考文献】