基于因果时序网络的FOA-GRNN电网故障诊断方法

Study on FOA-GRNN Fault Diagnosis Method of Power Grid Based on Causal Sequential Network

针对电网故障诊断过程常受到警报信息畸变以及保护设备误动或拒动等不确定因素的影响而导致误诊断的问题,提出了基于时序网络的果蝇优化算法-广义回归神经网络电网故障诊断方法。利用系统保护与设备之间存在的时序逻辑关系,对获得的电网故障警报信息甄别后再进行故障诊断。算例分析及测试结果说明,所提方法能够准确地实现电网的故障诊断,并适应电网拓扑结构的变化。

The uncertainties including the information distortion,malfunction and miss operation,have effects on fault diagnosis of the grid.These effects may leads to wrong conclusions given by the fault diagnosis system.The FOA-GR-RN fault diagnosis method of power grid is proposed based on sequential network.Via the characteristic of temporal logic relation between protection system and equipment,the obtained alarm information is initially distinguished and then the faults are diagnosed.By analyzing and testing,the proposed method can make accurate fault diagnosis,and deal with the changes of topology configuration of the grid as well.