电网重要节点、脆弱线路及骨干网架识别研究

发布时间：2018-08-14 09:22

【摘要】：近年来,多次大规模停电事故给人类社会造成巨大的经济损失,严重威胁社会安全,极大地引起了人们对电网运行安全性的关注。电网大停电事故多是由于过负荷、元件故障、控制保护及通信系统故障等内部因素和自然灾害等外部因素导致电网中某些脆弱线路或重要节点退出运行,进而引发连锁故障导致。在日常运行时,加强对重要节点和脆弱线路的关注有利于提高电网运行的安全性,减少大停电发生的概率;在自然灾害来临前,电网企业通过提高重要节点与脆弱线路的抗灾标准,构建骨干网架,以此保证对重要负荷供电,加快灾后恢复速度,减少大停电事故造成的影响。本文围绕电网节点重要度评估、脆弱线路辨识及骨干网架识别等问题展开研究,对提升电网运行的安全性有一定的指导意义。考虑到电网与互联网同属于复杂网络,本文对两者进行了类比,并基于互联网网页排序Hypertext-Induced Topic Search(HITS)算法,结合电力系统的实际情况,针对电网潮流、电源和负荷容量等影响节点重要度的因素,在电网原始拓扑上,通过引入权重、附加节点等处理对原算法进行相应的适应性改进,给出了一种电网节点重要度评估方法。该方法从节点功率的流入与流出角度定义了权威值和枢纽值来评估节点重要度,并考虑电网与互联网的差异,以两者之和作为电网节点的重要度,使评估结果更加全面和客观。在现有电网大停电事故相关研究的基础上,基于本文定义的单位熵综合负载率指标和风险思想,本文给出一种电网脆弱线路辨识的方法。本文定义单位熵综合负载率来衡量电网运行的脆弱性,该指标反映电网中线路的负载率及其分布情况。本文以该指标在线路开断前后的变化,并用线路两端点的重要度进行修正,作为线路开断后果的表征。在计算线路的相对开断概率时,本文不仅考虑线路自身的故障率,还将线路运行时因负荷波动导致的潮流越限开断的概率予以考虑。后者是基于直流潮流模型,考虑电网拓扑结构、实时运行状态、电源负荷分布及线路传输容量等影响系统安全运行的电气信息进行计算。以上述研究为基础,借助于图论中的最短路径与最小生成树算法,给出一种电网骨干网架识别流程。该方法能够更直接地将重要节点留存在骨干网架之中,同时使得网架的规模最小,即留存在骨干网架中的线路尽可能少且重要,以此兼顾经济性。该方法首先根据电网节点重要度评估,确定必须留存在骨干网架中的重要节点,同时判别这些节点的电源容量与负荷容量是否匹配,并进行相应处理;其次,以各线路的脆弱度评估值作为其权值,计算留存节点之间的最短路径及其对应的线路序列;接着,求解留存节点之间最短路径所构成图的最小生成树,并将最小生成树中的每条边用其对应的线路序列进行替代,得到初步的骨千网架;最后,对骨干网架进行电网安全运行校验,若存在潮流越限的线路,通过调整运行方式或增加合适线路来消除过载线路,保证电网安全运行,得到最终的骨干网架。
[Abstract]:In recent years, many large-scale blackouts have caused enormous economic losses to human society, seriously threatening social security and greatly arousing people's concern about the safety of power grid operation. In daily operation, strengthening the attention to important nodes and vulnerable lines is conducive to improving the security of power grid operation and reducing the probability of major blackouts; before natural disasters, power grid enterprises improve the important nodes and vulnerable nodes by increasing the probability of major blackouts. In order to ensure the power supply for important loads, speed up the recovery speed after disaster and reduce the impact caused by blackouts, the anti-disaster standard of transmission lines and the backbone network framework are constructed. Considering that the power grid and the Internet are both complex networks, this paper makes an analogy between them. Based on the Hypertext-Induced Topic Search (HITS) algorithm and the actual situation of the power system, the power flow, the power supply and the load capacity are considered as the factors that affect the node importance. In the original topology of the power grid, the weight is introduced. This method defines the authoritative value and pivot value from the point of view of the inflow and outflow of node power to evaluate the node importance, and considers the difference between the grid and the Internet, and takes the sum of the two as the importance of the grid node. On the basis of the existing research on power grid blackouts, and based on the unit entropy comprehensive load rate index and risk concept defined in this paper, a method for identifying vulnerable lines in power grid is presented. The unit entropy comprehensive load rate is defined to measure the vulnerability of power grid operation, which is contrary to the index. In this paper, the change of the index before and after the circuit break is taken into account, and the importance of the two ends of the line is modified to represent the consequence of the line break. The latter is based on the DC power flow model, which considers the electrical information which influences the safe operation of the power system, such as the topological structure of the power grid, the real-time operation state, the load distribution of the power supply and the transmission capacity of the line. This method can keep the important nodes in the backbone grid more directly, and make the scale of the grid minimum, that is, the number of lines remaining in the backbone grid is as small and important as possible, so as to give consideration to the economy. There are important nodes in the backbone network, and the power supply capacity and load capacity of these nodes are judged and processed accordingly; secondly, the shortest path between the remaining nodes and the corresponding line sequence are calculated with the vulnerability evaluation value of each line as its weight; and then, the shortest path between the remaining nodes is constructed. The minimum spanning tree is constructed, and each edge of the minimum spanning tree is replaced by its corresponding line sequence to obtain a preliminary skeleton network. Finally, the skeleton network is checked for the safe operation of the power grid. If there are lines whose power flow exceeds the limit, the overloaded lines can be eliminated by adjusting the operation mode or adding appropriate lines to ensure the safety of the power grid. Run and get the final backbone grid.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM732

【相似文献】