基于模糊神经网络的小电流接地系统故障选线

Fault line selection of small current neutral grounding system based on fuzzy neural network

小电流接地系统发生单相接地故障后,为解决传统上单一故障选线方法的局限性,文章利用信息融合技术并结合模糊系统和神经网络的特点,提出将模糊神经网络应用于配电网故障选线方法中。提取稳态时有功功率分量、暂态时衰减直流分量、基波和小波包能量熵极值作为综合选线判据,输入到模糊神经网络,再通过调整网络隶属度函数的初始参数大小,体现出在不同相角下,不同判据对故障选线判别能力的不同,能够将配网自身的特点很好地嵌入该算法中。在Matlab环境下搭建10.5kV的小电流接地系统仿真模型,仿真结果显示此方法具有选线准确率高、适应性强、灵敏度高、抗干扰能力强等优点,具有一定的可行性,能够很好地解决故障选线的难题。

This paper deals with fault line selection as single-phase grounding fault occurs in the small current neutral grounding system .In order to solve the limitations of single fault line selection meth-ods ,the information fusion technology is adopted ,and considering the characteristic of fuzzy system and neural network ,the fuzzy neural network is applied in fault line selection of power distribution network .Steady-state active component , transient decaying DC component , transient fundamental component and the extreme of wavelet packet-energy entropy（WP-EE） are considered as judgments of comprehensive line selection ,and used as the input of the fuzzy neural network .In order to reflect the difference of the abilities of fault judgment of different samples in the situation of different angles ,the initial values of the parameters in membership functions are adjusted ,and then the characteristic of power distribution network itself is perfectly integrated into the algorithms .Finally ,a 10.5 kV small current neutral grounding system simulation model is set up by Matlab .The simulation results illus-trate that this method is feasible ,can achieve high accuracy ,adjustability ,sensitivity ,anti-interfer-ence ability and so on ,and solves the difficulties in fault line selection well .