Abstract: In order to improve the accuracy of abnormal line loss detection results, an automatic line loss monitoring system is designed based on electric energy measurement. In terms of hardware, the transmission signal feedback circuit is designed to receive the abnormal signal in time and find the abnormal source. In terms of software, the communication data transmission management mode is established, and the communication quality evaluation model is constructed under the power metering. Taking the signal quality as the evaluation standard, the k-nearest neighbor(KNN)algorithm is used for signal classification to complete the automatic detection of line loss anomalies. The experimental results show that the designed system can obtain the abnormal situation of line loss in time, and the accuracy of the detection results is improved by 15.86%, which can maintain the reliability of power grid operation.