Abstract: Aiming at the problem of high cost and low efficiency of manual inspection of low-voltage switchgear, a system that can detect the operating status and faults of low-voltage switchgear online is researched. The system obtaines operating data of low-voltage switchgear through sensors. The data is processed by the method of cloud sample entropy, and the characteristic values of its normal operation and fault operation are obtained. The eigenvalues are calculated by using Mahalanobis distance algorithm. The Mahalanobis distance threshold is obtained through the training samples, and then the Mahalanobis distance from the experimental sample data is calculated. The Mahalanobis distance is between 7-9, and finally the distance is compared with the Mahalanobis distance threshold, realizing real-time and online monitoring of its operating status. Experiments show that the fault detection accuracy rate of this system is greater than 98%.