27 March 2018 Audio-based bolt-loosening detection technique of bolt joint
Author Affiliations +
Bolt joint, as the commonest coupling structure, is widely used in electro-mechanical system. However, it is the weakest part of the whole system. The increase of preload tension force can raise the reliability and strength of the bolt joint. Therefore, the pretension force is one of the most important factors to ensure the stability of bolt joint. According to the way of generating pretension force, the pretension force can be monitored by bolt torque, degrees and elongation. The existing bolt-loosening monitoring methods all require expensive equipment, which greatly restricts the practicality of the bolt-loosening monitoring. In this paper, a new method of bolt-loosening detection technique based on audio is proposed. The sound that bolt is hit by a hammer is recorded on the Smartphone, and the collected audio signal is classified and identified by support vector machine algorithm. First, a verification test was designed and the results show that this new method can identify the damage of bolt looseness accurately. Second, a variety of bolt-loosening was identified. The results indicate that this method has a high accuracy in multiclass classification of the bolt looseness. This bolt-loosening detection technique based on audio not only can reduce the requirements of technical and professional experience, but also make bolt-loosening monitoring simpler and easier.
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yang Zhang, Yang Zhang, Xuefeng Zhao, Xuefeng Zhao, Wensheng Su, Wensheng Su, Zhigang Xue, Zhigang Xue, "Audio-based bolt-loosening detection technique of bolt joint", Proc. SPIE 10599, Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XII, 1059929 (27 March 2018); doi: 10.1117/12.2296533; https://doi.org/10.1117/12.2296533

Back to Top