摘要
基于人工智能算法具有较好的容错性,引入纵横交叉算法(CSO)应用在配电网故障定位过程中。CSO中的横向交叉机制和纵向交叉机制在与竞争算子的配合下提供了较强的搜索能力,能够快速解决多变量非线性优化问题,为准确解决故障定位提供了基础。在多电源分区故障定位中改进适应度函数,对不同区域适应度函数设置区域权值,区域权值由反馈故障电流决定。这种设置方式可以增强算法的容错性,使得输出结果不会因为故障信息在传送过程中发生畸变而误判或者漏判。仿真部分由双电源配电网系统和三电源配电网系统组成,并通过算法进行了验证,每次反馈信息都由一次正常信息和畸变信息组成。从仿真结果可以看出CSO拥有较强的稳定性。
Based on good fault tolerance performance of artificial intelligence algorithm, crisscross optimization algorithm （CSO） is applied in the process of fault location for distribution network. In the CSO algorithm, the horizontal crossover mechanism and the vertical crossover mechanism in cooperation with competitive operator provide strong search ability, which could quickly solve multi variable and nonlinear optimization problems and supply the basis foundation to solve fault location problems. Improving fitness function in the multi-source sections fault location problem and setting section weight values for different section fitness functions, this setting mode can enhance fault tolerance of algorithm, then the output results won＇t erroneous judgment or missing judgment when the fault message distorted during the transformation. The simulation part is composed of the distribution networks system of double and third power sources, which is verified by the algorithm. The regular message and distortion message make up the each time feedback message. The results prove that CSO algorithm has strong stability.
出处
《电力系统保护与控制》
EI
CSCD
北大核心
2016年第21期109-114,共6页
Power System Protection and Control
基金
国家自然科学基金资助项目(51407035)~~
关键词
纵横交叉算法(CSO)
故障定位
适应度函数
容错性
稳定性
crisscross optimization algorithm （CSO）
fault location
fitness function
fault-tolerance
stability
