基于多源数据融合的电网故障智能诊断方法研究

为提高传统电网故障诊断方法的精度以及可靠性等,提出了一种基于电气量和开关量的多源数据融合的电网故障智能诊断方法。该方法首先对故障电流录波数据进行小波变换,以得到电流这一电气量的特征量;然后对开关量的诊断结果进行归一化处理,得到开关量对应的相对故障度证据;最后采用改进的D-S证据理论对电气量和开关量的多源数据进行融合,并得出最终诊断结果。实际电网结果表明,所提的故障诊断方法能够有效提高故障诊断的精度。

Research on Intelligent Fault Diagnosis Method of Power Network Based on Multi-Source Data Fusion

In order to improve the accuracy and reliability of traditional fault diagnosis methods for power grids,a power grid fault intelligent diagnosis method is proposed based on multi-source data fusion of electric quantity and switch quantity. The method firstly performs wavelet transform on the fault current recording data to obtain the characteristic quantity of the current electric quantity;then normalizes the diagnosis result of the switch quantity to obtain the relative fault degree evidence corresponding to the switch quantity;The DS evidence theory fuses multiple sources of electrical quantities and switching quantities and derives the final diagnosis. The actual power grid test results show that the proposed fault diagnosis method can effectively improve the accuracy of fault diagnosis.