Research on Optimization Techniques for Identification of Dual AD Inconsistencies Based on Analysis of Differences in Sampling Values

The current identification method for dual AD inconsistency in protective devices is summarized, and the adaptability of existing method based on field cases is analyzed. The current identification methods have poor ability to identify dual AD inconsistency under low current amplitude conditions. Based on the analysis of the characteristic differences between dual AD data under normal and abnormal operating conditions, a processing approach based on the degree of difference in sampling values is proposed, which includes two methods: sampling value comparison method and correlation coefficient method. The effectiveness of both methods was verified through cases of single channel signal abnormality and dual channel signal abnormality that occurred actually. The verification results show that these two identification methods can effectively improve the identification ability of the protection device for dual AD inconsistent faults under low current amplitude such as less than 0.1A. Among them, the correlation coefficient method is more suitable for situations where single signal is abnormal, and the sampling value comparison method is more suitable for situations where both signals are abnormal. This study provides useful references for supplementing and improving the dual AD inconsistency recognition scheme for protective devices.