In order to evaluate the rationality of power system operation mode and the operation status of electric power enterprises with high accuracy, the prediction model of line loss rate of low-voltage distribution network based on dual-mode frequency modulation (FM) decomposition is studied, and the high-precision prediction result of line loss rate is used to improve the operation and management performance of power network. The power transmission signals of low-voltage distribution network are collected, and the dual-mode FM basis function is used to decompose the signals and remove useless signals by dual-mode FM decomposition method. The hierarchical node identification strategy is used to divide the load nodes of the low-voltage distribution network from the decomposed distribution signals, and the injection current of each load node is obtained. Support vector machine (SVM) method is used to establish a prediction model of line loss rate in the same period. The injection current obtained is input into the model, and the optimization result of line loss rate prediction is output. Experimental results show that the prediction results of the model are in good agreement with the actual results, and the mean absolute error and root mean square error are both lower than 0.09%, which proves that the model can effectively predict the line loss rate of low-voltage distribution network in the same period with high prediction accuracy, and can provide an effective basis for the operation and management of intelligent distribution network.