In order to improve the detection ability of carrier communication module, an improved genetic BP algorithm network model is constructed, which includes BP algorithm model, genetic algorithm model, fitness evaluation module and communication anomaly diagnosis module, and can realize the comprehensive factor evaluation in the process of carrier communication. And through the point cloud registration algorithm based on machine learning to detect carrier communication anomalies, in the face of various factors such as power line conditions, signal transmission distance, signal interference, channel conditions, carrier communication equipment, system design and implementation, can quickly detect carrier communication conditions, greatly improving the detection capability of carrier communication module. Point Cloud Registration (PCR) algorithm is used to detect anomalies in real time, and the carrier communication anomaly diagnosis can be judged. Through the experiment, the error rate of carrier communication is greatly reduced, which lays a foundation for the next research.