Fiber optic identification of bolt loosening features based on modal level convergence calculation

The bolt fastening structure of aircraft is prone to loosening in the harsh service environment, and has strong concealment, which makes the maintenance extremely difficult. Therefore, this paper develops a bolt group loosening event discrimination method based on the convergence of modal steps, which is used to determine whether the bolts are loose or not, and solves the problem of global efficient monitoring due to the large number of bolts and dispersed locations in the actual service process. A bolt loosening location discrimination method based on modal level convergence feature calculation is proposed to achieve the bolt loosening location discrimination, breaking through the limitations of time consuming and labour intensive, structural damage susceptibility and limited applicability caused by the need to build a large scale bolt loosening sample library for the purpose of improving the positioning accuracy of the mode recognition class method. A bolt loosening degree discrimination method based on multi-order modal joint calculation is developed, and bolt loosening degree quantification rules are formulated, which provides basic data support for multi-level credibility assessment. On this basis, a bolt loosening monitoring system based on optical frequency domain reflective distributed fibre optic sensors for composite wall plate structures is constructed. The results show that the average identification accuracy of bolt loosening based on modal level convergence calculation is better than 90%. The research results are of great significance for the realisation of aircraft bolted fastening structure health monitoring, maintenance decision-making as appropriate, and the development of digital twin technology for connecting structures.