A portable bolt-loosening detection system with piezoelectric-based nondestructive method and artificial neural networks

Abstract

In general, the bolted joints that connect and secure components together can be easily spotted in our surroundings. This joining method has been commonly used in various areas of engineering (e.g. aerospace, civil, and mechanical engineering) as it has been proven one of the most effective means to join parts together. Although it has its advantages, the vibration that bolted structures endure during service ultimately causes the bolts to loosen. This can, in turn, have a negative effect on the structure’s safety and may, at worst, cause it to fail. Routine inspections of structures are conducted on a regular basis, with some inspection categories requiring heavy equipment in order to acquire certain data. In addition, monitoring systems can be expensive to install and maintain, especially in large infrastructures. In an effort to rectify this, in this study, a piezoelectric transducer–based nondestructive technique is used in conjunction with the application of a reattachable device to investigate the possibility of creating a low-cost inspection system. The acquired data were processed by means of an artificial neural network technique that showed promising results in terms of mitigating bolt loosening.