Abstract: As China’s distributed energy is still in the development stage, energy transmission loss will inevitably occur in the transmission process from the source end to the load end, In order to reduce the loss of transmission energy, we should also beware of electricity theft. In order to improve the accuracy of electricity theft characteristics established and electricity theft detection, the principle of common electricity theft methods is analyzed, and the ReliefF multivariate characteristics selection algorithm is used to optimize the electricity theft characteristics. The Back Propagation (BP) neural network-based electricity theft detection model is built, and the optimized characteristics are selected as the input of the model. The experiment results show that the detection model has better electricity theft identification accuracy by using the optimized characteristics for electricity theft detection.