Quantitative Monitoring of Bolt Looseness Using Multichannel Piezoelectric Active Sensing and CBAM-Based Convolutional Neural Network

Abstract

The bolted connection is widely utilized in engineering to practically and rigidly couple structural components. The integrity of the connection is paramount to the safety of the structure and has prompted the development of many monitoring methods, including the piezoelectricity-based active sensing method. However, the active sensing method cannot quantify bolt looseness due to the unclear relationship between bolt looseness and the single monitoring index typically used in the active sensing method. Thus, the authors propose the unique combination of a one-dimensional convolutional neural network (1DCNN) and multichannel active sensing for quantitative monitoring of bolted connections. In an experiment, piezoelectric ceramic transducer (PZT) patches are bonded on steel plates connected by a bolt. Each patch is wired to a multichannel active sensing monitoring system. After obtaining multichannel stress wave signals at different looseness levels, a looseness vector is calculated to generate training and validation datasets. A baseline 1DCNN model and a novel model improved using the convolutional block attention module (CBAM) are used to monitor the bolt looseness. Finally, the authors verify that the multichannel active sensing method combined with the 1DCNN model can accurately perform quantitative monitoring of bolt looseness, and the monitoring accuracy of the baseline 1DCNN model is above 91.07% in three different specimens. Compared with the baseline 1DCNN model, the monitoring accuracy of the CBAMCNN model improved by approximately 5%. Overall, the method proposed in this article offers a new and highly accurate approach for quantitative monitoring of bolted connections.

Review on recent advances in structural health monitoring paradigm for looseness detection in bolted assemblies

“…They concluded that the CNN model trained using the scalograms obtained from the updated FE model accurately predicted the damage state of the structure. Chen et al 155 combined a 1D CNN model with a multichannel active sensing method to determine bolt looseness in three steel-bolted joint specimens. They divided their approach into four steps: (i) data acquisition, (ii) data preprocessing, (iii) model training and validation, and (iv) bolt looseness monitoring.…”

An efficient robotic-assisted bolt-ball joint looseness monitoring approach using CBAM-enhanced lightweight ResNet

“…From the perspective of the model, CBAM is better in capturing the intrinsic correlation between features. The classification performance of the network is effectively improved by locally focusing observation of the signal to be studied [45].…”

Monitoring a Bolted Vibrating Structure Using Multiple Acoustic Emission Sensors: A Benchmark

“…During the service life of a bolted structure, the torque can evolve according to operational conditions such as loading, in particular when these structures are submitted to vibrations. Monitoring the loosening condition in bolted structures during operation remains challenging because contact and friction in bolted joints induce a nonlinear stochastic behavior [16][17][18][19][20][21]. A common solution consists of measuring the pretightening force using a strain gauge bonded on the top surface of a bolt head [22].…”