Aiming at the problem that the existing methods have low detection accuracy for new ways of electricity theft, this paper proposes a new theft detection method based on multi-level non-negative sparse coding and SVM. This method uses the monthly electricity consumption curve as the detection object. Firstly, the multi-level electricity consumption pattern characteristics of the sample are extracted based on the multi-level non-negative sparse coding; secondly, the numerical statistical characteristics of the sample are extracted based on the electricity theft scenario analysis; then, the fusion detection features of the above two are input into the SVM classifier for electricity theft detection; finally, the example of the Irish smart meter data set is used to verify that the proposed method can improve the accuracy and recall rate of the detection.