According to the multi-channel sampling continuity of the multi-loop gateway electricity meter monitoring, the multi-features of the sampled signal and its coupling, an online monitoring method of the multi-loop gateway electricity meter based on the support vector machine (SVM) is proposed in this paper. In this method, the signal acquisition unit, data communication unit, data processing unit and host computer display unit are designed respectively based on the monitoring system. This paper builds multiple fault defect models of the electricity meter, and transmits the three-phase current and voltage real-time data of the defect model to the data processing unit through the signal acquisition unit. The data processing unit performs basic calculations on the collected data to form corresponding active power sample data, and performs off-line analysis of wavelet packet energy spectrum to form a fault feature pool. On this basis, a multi-classification SVM model based on the SVM algorithm is constructed, and the obtained model are embedded in the upper computer display unit; when the monitoring system is running, the upper computer display unit is directly based on the association rules of the model and performs feature matching with the converted sampling information to realize the real-time monitoring of the operation of electricity meter. Verification by examples shows that the proposed method can monitor multi-loop gateway electricity meters online in real time and accurately, and has strong application value and prospects.