Structural Health Monitoring Impact Classification Method Based on Bayesian Neural Network

Abstract

This paper proposes a novel method for multi-class classification and uncertainty quantification of impact events on a flat composite plate with a structural health monitoring (SHM) system by using a Bayesian neural network (BNN). Most of the existing research in passive sensing has focused on deterministic approaches for impact detection and characterization. However, there are variability in impact location, angle and energy in real operational conditions which results in uncertainty in the diagnosis. Therefore, this paper proposes a reliability-based impact characterization method based on BNN for the first time. Impact data are acquired by a passive sensing system of piezoelectric (PZT) sensors. Features extracted from the sensor signals, such as their transferred energy, frequency at maximum amplitude and time interval of the largest peak, are used to develop a BNN for impact classification (i.e., energy level). To test the robustness and reliability of the proposed model to impact variability, it is trained with perpendicular impacts and tested by variable angle impacts. The same dataset is further applied in a method called multi-artificial neural network (multi-ANN) to compare its ability in uncertainty quantification and its computational efficiency against the BNN for validation of the developed meta-model. It is demonstrated that both the BNN and multi-ANN can measure the uncertainty and confidence of the diagnosis from the prediction results. Both have very high performance in classifying impact energies when the networks are trained and tested with perpendicular impacts of different energy and location, with 94% and 98% reliable predictions for BNN and multi-ANN, respectively. However, both metamodels struggled to detect new impact scenarios (angled impacts) when the data set was not used in the development stage and only used for testing. Including additional features improved the performance of the networks in regularization; however, not to the acceptable accuracy. The BNN significantly outperforms the multi-ANN in computational time and resources. For perpendicular impacts, both methods can reach a reliable accuracy, while for angled impacts, the accuracy decreases but the uncertainty provides additional information that can be further used to improve the classification.

Graphene nanoparticles as data generating digital materials in industry 4.0

“…Novel AI-based methods for the inspection of the Automated Fiber Placement (AFP) process have also been presented by several researchers [139][140][141][142][143] . As part of health monitoring of structures, machine/deep learning models have been used for defect/damage detection [144][145][146][147][148][149][150] , characterization of cracks/delamination [151][152][153] and classification of impact levels 154 . Yu et al 154 demonstrated that probabilistic Bayesian and traditional artificial neural networks can successfully classify the energy levels of different impact events based on the signals obtained from a network of piezoelectric sensors.…”

“…As part of health monitoring of structures, machine/deep learning models have been used for defect/damage detection [144][145][146][147][148][149][150] , characterization of cracks/delamination [151][152][153] and classification of impact levels 154 . Yu et al 154 demonstrated that probabilistic Bayesian and traditional artificial neural networks can successfully classify the energy levels of different impact events based on the signals obtained from a network of piezoelectric sensors. Deep learning tools are particularly capable of such tasks when the signal is in the form of 2D/3D fields and maps 56,57 .…”

A Hybrid Convolutional and Recurrent Neural Network for Multi-Sensor Pile Damage Detection with Time Series

“…The first method extracts statistical features from time-series signals and uses them as the input to ML algorithm frameworks [ 28 , 29 , 30 ]. The second method utilizes the one-dimensional (1D) convolutional neural network (CNN) to automatically extract the features [ 31 , 32 , 33 , 34 ]. The 1DCNN has the same advantages as the CNN and is capable of learning more comprehensive features (i.e., less susceptible to subjective experience).…”

Cracking and Fiber Debonding Identification of Concrete Deep Beams Reinforced with C-FRP Ropes against Shear Using a Real-Time Monitoring System

“…Furthermore, most NDT methods are time-consuming, expensive, require access in situ, and cannot apply continuous and real-time monitoring. Infrared thermography [ 30 , 31 ], ultrasonic C-scan detection, acoustic emission (AE), and resistance strain gauges [ 32 ] are commonly used to check the integrity of composite materials [ 33 , 34 , 35 ]. The latter is predominantly destined for static measurements and achieves low transverse sensitivity [ 36 ].…”