Study the mechanism of intermittent grinding burn and the on-line monitoring method of acoustic emission to avoid the phenomenon of product grinding burn. Based on the surface grinding temperature field theory and the mirror heat source method, a temperature field model of the intermittent grinding workpiece edge is established. Based on this model, the intermittent grinding burn mechanism can be studied. In order to verify the effectiveness of the above model, the orthogonal Experiments were designed for different intermittent grinding conditions. Infrared thermal imaging cameras and acoustic emission signals were used to monitor the temperature of the intermittent grinding area on-line, and the pickling method and Barkhausen noise detector were used to burn the surface of the workpiece after grinding. Inspection and verification, by solving the wavelet packet energy of the acoustic emission signal to establish the relationship with the temperature of the grinding zone. This model can effectively reflect the temperature field distribution of the grinding zone at the edge of the workpiece during intermittent grinding. The calculation results show that the edge of the fractured grinding workpiece has a higher temperature and is more likely to cause burns than the grinding zone at other positions; the experiment shows that acoustic emission The total energy of the wavelet packet transform of the signal has a certain correlation with the grinding area. Based on the acoustic emission signal, the intermittent grinding burn can be monitored online. The experimental results prove the effectiveness of the model for analyzing the intermittent grinding burn mechanism and the feasibility of using acoustic emission signals to monitor the intermittent grinding burn online; finally, the actual intermittent grinding burn burn of a certain steering nut product is carried out online Monitoring application, practical results show that this method is more efficient and environmentally friendly than traditional pickling burn detection, and it is of great significance to realize the automation and intelligence of grinding processing burn detection.