主动频移法的光伏并网逆变器孤岛检测研究

发布时间：2018-03-27 11:12

本文选题：孤岛检测　切入点：盲区分布　出处：《新疆大学》2017年硕士论文

【摘要】：从工业革命到现在,由于传统化石燃料越来越少和环境问题越发严峻的双重问题,人类加大了对风能、太阳能等清洁能源的利用,分布式发电(generation ddistribute,DG)革新研究工作在当今供电领域中占据了一定的地位。孤岛现象是指当主电网出现故障或人设关断进行检修而处于失电情况下,分布式发电系统没有对主电网断开做出响应,仍持续给予本地负载进行供电形成一种孤立状态,孤岛发生意味着对公共配电网和人员等形成安全危机,即反孤岛检测算法的设计和反孤岛装置的装配是发电系统实际供电必备的。首先依次分别阐述孤岛效应和对应的处理手段产生的具体背景、近况和算法优化的实际价值,从发生机理角度出发简述了海内外关于不同算法的检测性能评判指标,对现阶段运用十分普及的各类检测手段都做了优缺点分析并给出了亟待提高的对应参量指标。给出了孤岛检测盲区(etection-NonZoneD,NDZ)在不同评判坐标系下的区域分布图,并在normf0CQ?坐标系下对AFD、AFDPF在孤岛发生时的检测时间、检测成功率和漏检情况进行了指标定性和定量对比。针对各孤岛检测算法在检测速度、检测成功率、检测盲区和谐波畸变率存在固有矛盾,参数选取上难以单向统一。提出一种新型的智能盲区规避孤岛检测法(IAFDPF),在对负载进行鉴相识别的基础上定向变化初始斩波因子方向,并跟随实时检测频率动态调整盲区分布。通过ULINKMATLAB/SIM软件模拟了简易的协同并网供电模型,通过定时给予BREAK开关设置故障模式。对改进的智能孤岛检测算法(IAFDPF)和AFD以及AFDPF检测算法进行各个评判指标的对比。仿真结果表明:改进的智能算法(IAFDPF)能够较大降低孤岛检测盲区,且能实现品质因素5.20fQ(27)情况下的无盲区检测,并根据负载特性智能改变盲区分布特征。从理论上实现了检测速度、检测成功率、谐波畸变率和检测盲区的协调统一,为孤岛检测系统实物设计和改进提供参考价值。
[Abstract]:Since the industrial revolution, due to the dual problems of less and less fossil fuels and more serious environmental problems, mankind has increased its use of clean energy such as wind energy and solar energy. The research of distributed generation DGG (distributed generation DGG) innovation has occupied a certain position in the field of power supply nowadays. The phenomenon of isolated island is that when the main power network is out of order or when the main power network is shut down and overhauled, it is in the situation of power loss. The distributed generation system does not respond to the disconnection of the main power grid and continues to supply the local load in an isolated state. The occurrence of the isolated island means a security crisis for the public distribution network and the personnel, etc. That is, the design of anti-islanding detection algorithm and the assembly of anti-islanding device are necessary for the actual power supply of power generation system. Firstly, the background of islanding effect and the corresponding processing means, the recent situation and the practical value of algorithm optimization are expounded respectively. From the point of view of occurrence mechanism, the detection performance evaluation indexes of different algorithms at home and abroad are briefly described. The advantages and disadvantages of all kinds of detection methods which are very popular at present are analyzed and the corresponding parameter indexes that need to be improved are given. The regional distribution map of the blind area of isolated island detection in different judgement coordinates is given, and the map is given in normf0CQ? In the coordinate system, the detection time, detection success rate and missed detection rate of AFDF are compared qualitatively and quantitatively in terms of the detection speed and detection success rate of each islanding detection algorithm. There are inherent contradictions between blind area detection and harmonic distortion rate, and the selection of parameters is difficult to be unified.A new intelligent blind zone detection method for avoiding isolated island is proposed, which changes the direction of initial chopper factor on the basis of phase identification of load. The blind area distribution is dynamically adjusted with the real-time detection frequency. A simple cooperative grid-connected power supply model is simulated by ULINKMATLAB/SIM software. The fault mode of BREAK switch is set by timing. The comparison between the improved intelligent islanding detection algorithm (IIAFDPF), AFD and AFDPF detection algorithm is made. The simulation results show that the improved intelligent algorithm can be reduced greatly. Low isolated island detection blind area, And can realize the blind area detection under the condition of the quality factor 5.20fQQ27), and change the blind area distribution characteristic intelligently according to the load characteristic. Theoretically, the detection speed, the detection success rate, the harmonic distortion rate and the detection blind area are coordinated and unified. It provides reference value for the physical design and improvement of islanding detection system.
【学位授予单位】：新疆大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM615;TM464

【参考文献】