【作者】 苏丹；

【导师】 罗毅；

【作者基本信息】 华中科技大学 ， 电力系统及其自动化， 2011， 硕士

【摘要】 随着人们对供电可靠性和供电质量的需求越来越高,对配电网故障隔离的快速性、准确性提出了更高的要求。目前,配电自动化系统主要采用的是集中控制的处理方式,即由主站收集各FTU的信息,通过对网络整体数据的分析判定,将控制指令发送至各FTU,实现故障隔离。然而,这种方式对子站的依赖性很强,对信息畸变的适应性很差,一个节点的信息缺失就可能引起整个网络的崩溃。而分布式故障隔离方法,能够通过相邻开关节点的故障信息,实现故障隔离,很好的解决了信息畸变带来的问题,具备较高的快速性、准确性和自主性,是故障隔离未来发展的新方向。本文提出了一种基于分组的配电网故障隔离分布式算法。该算法在对配电网络进行分区后,选取组长,由组长依据组内成员的故障信息,实现故障隔离,提高了故障处理的速度,克服了对子站的依赖问题。文章分别在信息完备和信息发生畸变两种情况下,对算法进行了分析。并以一个辐射状网络为例,论证了算法的正确性。然后,对这种分布式算法的时序特征进行了分析。讨论了分布式故障隔离算法和现有故障处理方案之间的配合,给出它们之间的时序关系,并通过一个典型手拉手配电网络,对这一时序关系进行了说明。最后,简要介绍了DNP 3.0规约的构成结构、各层帧格式和该规约的特点。并对其在故障隔离分布式算法中的应用进行了分析。更多还原

【Abstract】 As the need of people for reliability and quality of the power supply is higher, more demands for the speed and veracity of fault isolation of distribution network are mentioned. By now, the distribution automation system mainly uses a kind of control based on the centralized communication. A master station collects all the information from FTUs, analyses the location of fault with the information of the whole network, and then sends orders to the right FTU to isolate the fault. However, the implement of this method is deeply depended on the sub-master station. And it has a poor adaptability of information distortion, which means that if some information is distorted or missing, the whole algorithm will fail. The use of distributed algorithm in fault isolation can successfully isolate the fault by analyzing the information from its neighbors and itself. It gives a good solution to the problem that caused by the loss and distortion of the information. This way is rapidity, accuracy and autonomy. And it will be the future direction of the fault isolation.A distributed algorithm based on group for fault isolate of distribution network is brought forward in this paper. The method of this algorithm: partition the distributed network and choose a group leader for each group. The leader of the group members could achieve the fault isolation according to the information that all the members in this group collects. This scheme does not depend on the sub-master station, which improves the speed of processing failure of distributed network. The algorithm is analyzed, when the information that station collects is completed, and also when some of the information is missing. The validity of the algorithm is proved by using it for a radial distributed network.Then, the time series features of this distributed algorithm are also analyzed. The cooperation between the distributed fault isolation algorithm and the solution of fault isolation using now is discussed in this paper. And the time series between them are given and explained by using a typical distribution network as an example.Finally, a brief description of DNP 3.0 is given. And the composition of the structure, the frame format in each layer and the characteristics of the protocol are discussed. In the end, how to use the DNP 3.0 protocol for the distributed algorithm that mentioned before is showed to us.更多还原