基于人工蜂群算法的配电网故障定位

发布时间：2018-07-29 09:40

【摘要】：配电网故障定位的作用在于馈线故障后快速查找出故障区段，它是配电网故障处理的核心环节，直接影响着故障区域隔离和非故障区域供电恢复的正确性及有效性。故障的快速定位为实现智能电网的自愈控制奠定了基础，对缩短故障停电时间、提高配电网供电可靠性具有重要的实际意义。本文借鉴了人工蜂群算法广泛的应用思路，从网络拓扑分析、算法应用及改进、信息畸变或缺失的处理、故障定位模型等几个方面对配网故障定位方法进行了较深入的研究。本文提出了基于标准蜂群算法结合故障分级处理的配电网故障定位方法，该方法能够适应含分布式电源和多电源环状网络的拓扑变化，先利用故障分级处理的判据确定一部分无故障的区段，再通过标准蜂群算法定位出故障区段，极大地提高了故障定位的效率。算例仿真结果验证了提出的故障定位方法的有效性。针对标准蜂群算法存在的容易陷入局部极值、收敛速度慢的缺点，提出了一种改进蜂群算法，从引入免疫算法中基于抗体浓度的多样性保持、改进引领蜂搜索方程、采用多种群协同进化三个方面改进了标准蜂群算法。四个标准测试函数的仿真结果表明改进蜂群算法的收敛精度、收敛速度和稳定性均优于标准蜂群算法。配电网的仿真算例中，两种蜂群算法和遗传算法定位结果的对比也验证了改进蜂群算法的优越性能。在单一故障的前提下，本文提出了故障定位的机会约束规划模型，该模型采用考虑了故障漏报和误报影响的开关函数，将开关期望和实际的状态差异设定为机会约束条件，依据配电网中各种故障情况发生的可能性建立了目标函数。模型的求解采用基于蒙特卡洛仿真的改进蜂群算法，仿真结果表明该模型存在无法准确定位某些小概率故障情况的问题，因此采用比较电气量中电流幅值和正序分量相位的方法，进一步修正了故障定位的结果。
[Abstract]:The function of fault location in distribution network is to find fault section quickly after feeder fault. It is the core link of fault processing in distribution network, which directly affects the correctness and effectiveness of fault area isolation and power supply recovery in non-fault area. The rapid fault location lays a foundation for the realization of self-healing control of smart grid and has important practical significance for shortening the time of failure and improving the reliability of power supply in distribution network. Based on the extensive application of artificial bee colony algorithm, this paper makes a deep research on fault location method of distribution network from several aspects, such as network topology analysis, algorithm application and improvement, information distortion or missing processing, fault location model and so on. In this paper, a fault location method based on standard bee colony algorithm and fault classification is proposed, which can adapt to the topology changes of distributed and multi-source ring networks. First, a part of the fault free section is determined by using the criterion of fault classification, and then the fault section is located by the standard bee colony algorithm, which greatly improves the efficiency of fault location. The simulation results show that the proposed fault location method is effective. Aiming at the shortcomings of standard bee colony algorithm, which is easy to fall into local extremum and slow convergence speed, an improved bee colony algorithm is proposed, which is based on the diversity of antibody concentration in the immune algorithm, and improves the search equation of leading bee. The standard bee colony algorithm is improved from three aspects of multi-colony coevolution. The simulation results of four standard test functions show that the convergence accuracy, convergence speed and stability of the improved bee colony algorithm are better than those of the standard bee colony algorithm. In the simulation example of distribution network, the comparison of the results of two honeybee colony algorithms and genetic algorithm also verifies the superior performance of the improved beecolony algorithm. On the premise of a single fault, this paper presents a chance constraint programming model for fault location. The model uses switching functions that take into account the influence of fault false positives and false positives, and sets the difference between switch expectation and actual state as a chance constraint condition. The objective function is established according to the possibility of various faults in distribution network. The improved bee colony algorithm based on Monte Carlo simulation is used to solve the model. The simulation results show that the model can not locate some small probability faults accurately. Therefore, the results of fault location are further modified by comparing the current amplitude and the phase of positive sequence components in electrical quantities.
【学位授予单位】：湖南大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM73;TP18

【相似文献】