基于移动多代理的配电网故障恢复研究

发布时间：2018-08-30 10:42

【摘要】：随着智能电网的推广和发展，越来越多的分布式电源和微网接入配电网，使得配电网的结构愈渐复杂。传统的优化算法已经不能完全满足配电网故障恢复重构的需要，寻找一种新的方法策略去适应配电网的复杂多变显得尤为重要。本文基于配电网故障恢复的研究背景，综合了当前的研究现状，就目前配电网故障恢复中可能存在的问题，进行了如下研究：首先，介绍了配电网故障恢复过程中需要反复进行计算的两部分：拓扑分析和潮流计算。对比了深度优先搜索算法和广度优先搜索算法的优劣，根据配电网的辐射状运行结构，采用基于广度优先搜索的拓扑分析算法。并在对比了几种不同的潮流计算方法的基础上，选择前推回代法作为本文的潮流计算方法。其次，将移动多代理技术应用于含分布式电源的配电网故障恢复研究中。建立了故障恢复的数学模型，并通过构建控制中心代理和子任务代理来实现配电网的故障恢复任务，其中，子任务代理中嵌入蚁群算法进行配电网故障恢复的优化。通过子任务代理的动态联盟，减小蚁群算法的寻优范围，提高了搜索效率。算例仿真表明了该策略的有效性。最后，本文考虑了多类型分布式电源以微网形式接入智能配电网的情况。首先对微网中的多类型分布式电源进行分析，将其随机功率模型转化为分时段或分场景的恒定功率模型。考虑到故障发生背景的不同，分别从连续故障和非连续故障两方面进行故障设置，，并采用基于移动多代理的演化算法对配电网的多目标进行优化，最后结合含微网的算例验证了该算法的有效性。
[Abstract]:With the promotion and development of smart grid, more and more distributed generation and microgrid are connected to distribution network, which makes the structure of distribution network more and more complex. The traditional optimization algorithm can not fully meet the needs of distribution network fault recovery and reconfiguration. It is particularly important to find a new method strategy to adapt to the complex and changeable distribution network. Based on the research background of distribution network fault recovery, this paper synthesizes the current research status, and carries out the following research on the possible problems in distribution network fault recovery: first, This paper introduces two parts, topology analysis and power flow calculation, which need to be calculated repeatedly in the process of distribution network fault recovery. The advantages and disadvantages of depth-first search algorithm and breadth-first search algorithm are compared. According to the radial operation structure of distribution network, the topological analysis algorithm based on breadth-first search is adopted. On the basis of comparing several different power flow calculation methods, the forward pushback method is chosen as the power flow calculation method in this paper. Secondly, the mobile multi-agent technology is applied to the research of distribution network fault recovery with distributed generation. The mathematical model of fault recovery is established, and the control center agent and sub-task agent are constructed to realize the fault recovery task of distribution network, in which ant colony algorithm is embedded in the sub-task agent to optimize the fault recovery of distribution network. The optimization range of ant colony algorithm is reduced by dynamic alliance of subtask agent, and the search efficiency is improved. The simulation results show the effectiveness of the strategy. Finally, this paper considers the situation that multi-type distributed power supply is connected to smart distribution network in microgrid form. Firstly, the multi-type distributed power generation in microgrid is analyzed, and its stochastic power model is transformed into a constant power model with time or scene. Considering the difference of fault occurrence background, fault setting is carried out from two aspects of continuous fault and discontinuous fault, and the multi-objective distribution network is optimized by evolutionary algorithm based on mobile multi-agent. Finally, the effectiveness of the algorithm is verified by an example with microgrid.
【学位授予单位】：燕山大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM711

【参考文献】