自适应模糊神经网络多信息融合串联型电弧故障检测方法

Adaptive Fuzzy Neural Network Multi-information Fusion Serial Fault Arc Detection Method

针对低压交流串联电弧故障因其特殊性难以识别的问题,在实验室建立了实验平台,采集不同负载下的正常和电弧故障的电流波形数据,提出一种自适应模糊神经网络(adaptive neuro fuzzy inference system,ANFIS)多信息融合串联型电弧故障检测方法,通过多层次特征参数提取,对自适应模糊神经网络进行训练、测试。结果表明,该方法可以有效地克服反向传播(back propagation,BP)神经网络故障诊断中收敛速度慢、误差精度大的缺点,得到更为精准、理想的诊断结果。

In response to the difficulty in identifying low-voltage AC series arc faults due to their particularity,an experimental platform is established in the laboratory to collect current waveform data of normal and arc faults under different loads.An adaptive neuro fuzzy inference system(ANFIS)fault arc detection method based on multi-information fusion and series is proposed.Through multi-level feature parameter extraction,ANFIS adaptive fuzzy neural network is trained and tested.The results show that this method can effectively overcome the shortcomings of slow convergence speed and large error precision in back propagation(BP)neural network fault diagnosis,and get more accurate and ideal diagnosis results,which opens up a new way for the research in this field.