Abstract: High strength bolts are important connecting components of steel structures, and they are prone to fracture due to corrosion during high stress service. In order to study its damage evolution law under stress corrosion, real-time monitoring of its damage process was carried out using acoustic emission technology. Combining the improved K-means algorithm and the dynamic threshold, the acoustic emission damage signals were clustered, and the clustered acoustic emission signals are further divided into rust damage stages. The test results show that the damage signals of high strength bolts can be divided into three categories, and the damage stage of the third category of signals is further divided, and finally four stages of stress corrosion are obtained. The test verified the effectiveness of the proposed acoustic emission method, and the acoustic emission method can be used to monitor the health of high-strength bolts in real time.