The invention discloses a metering automation terminal fault diagnosis method based on deep learning and supporting vector data description, and relates to the technical field of power metering faultdiagnosis. According to the metering automation terminal fault diagnosis method based on deep learning and supporting vector data description, fault data acquired by the metering automation terminal are subjected to feature extraction through a deep confidence network model in deep learning, and carrying out fault diagnosis and classification by using supporting vector data description; the deep confidence network model can directly start from a low-level original signal and obtain a high-level characteristic representation through layer-by-layer greedy training, thereby avoiding manual operation of characteristic extraction and selection, effectively eliminating complexity and uncertainty brought by traditional manual characteristic extraction and selection, and enhancing the intelligenceof the diagnosis process; according to the method, the supporting vector data description is used for classifying and identifying the samples, so that the accuracy and the efficiency of the multi-class classification problem of the fault diagnosis of the metering automation terminal are effectively improved.