摘要
可靠的供配电系统是防护工程能够稳定高效运行的重要基础。由于传统集中式系统架构下的配电网在运行、维护中存在诸多问题,近年来学者们将分布式架构的思想引入防护工程配电网的建设中。面向群智能系统架构,提出了一种基于BP神经网络的防护工程配电网故障诊断方法。利用群智能节点自组网能力和BP神经网络的自学习及良好的泛化性能,实现防护工程配电网的故障诊断。仿真结果表明:提出的方法具有训练快速、实现方便等优点,对于防护工程配电网单一和多重故障均能做出准确的定位。
Reliable electrical supply and distribution system plays an important role in stable and efficient operation of protective engineering.Recently,scholars have introduced the idea of decentralized architectures into the construction of the distribution network in the protective engineering,owing to many problems existing in the operation and maintenance of the distribution network under the traditional centralized system architecture.This paper proposes a fault diagnosis method of protective engineering distribution network based on BP neural network,which is realized by using the self-organizing ability of I^2B(insect intelligent building)system and self-learning and good generalization performance of BP neural network.The simulation results show that the proposed method has the merits of fast training and convenient implementation,and can make accurate positioning for single and multiple faults of protective engineering distribution network.
作者
张玉晗
邢建春
赵硕
刘夕广
王世强
ZHANG Yuhan;XING Jianchun;ZHAO Shuo;LIU Xiguang;WANG Shiqiang(College of National Defense Engineering,Army Engineering University of PLA,Nanjing210007,China;Building Energy Conservation Research Center,Tsinghua University,Beijing100084,China)
出处
《防护工程》
2019年第2期54-61,共8页
Protective Engineering
基金
国家重点研发计划项目(2017YFC0704100)
关键词
群智能
神经网络
故障诊断
防护工程
配电网
图论
I^2B(insect intelligent building)
neural network
fault diagnosis
protective engineering
distribution network
graph theory
