Application of data-driven technique to multiple faults diagnosis in grounding grids
基于指定元分析的接地网多故障诊断方法研究

This paper introduced designated component analysisDCA to implement multiple faults diagnosis and prediction in grounding grids based on data-driven technique. It established a projection frame, which was the theory foundation for DCA based on multiple faults diagnosis. Under the DCA projection frame, anomaly detection and fault diagnosis problem was transformed into the signifcance detection problem for projection energy of the observation data which was projected to fault subspace spanned by fault patterns defined in advance under condition that the system was abnormal, it could identify faults diagnosis according to the significance detection problem for projection energy. It provided a case study of the diagnosis in grounding grids, and simulation results shows that the efficiency of the DCA method for multiple faults diagnosis.