Search

In this paper, a structural health monitoring (SHM) system is proposed to provide automatic early warning for detecting damage and its location in composite pipelines at an early stage. The study considers a basalt fiber reinforced polymer (BFRP) pipeline with an embedded Fiber Bragg grating (FBG) sensory system and first discusses the shortcomings and challenges with incorporating FBG sensors for accurate detection of damage information in pipelines. The novelty and the main focus of this study is, however, a proposed approach that relies on designing an integrated sensing-diagnostic SHM system that has the capability to detect damage in composite pipelines at an early stage via implementation of an artificial intelligence (AI)-based algorithm combining deep learning and other efficient machine learning methods using an Enhanced Convolutional Neural Network (ECNN) without retraining the model. The proposed architecture replaces the softmax layer by a k-Nearest Neighbor (k-NN) algorithm for inference. Finite element models are developed and calibrated by the results of pipe measurements under damage tests. The models are then used to assess the patterns of the strain distributions of the pipeline under internal pressure loading and under pressure changes due to bursts, and to find the relationship of strains at different locations axially and circumferentially. A prediction algorithm for pipe damage mechanisms using distributed strain patterns is also developed. The ECNN is designed and trained to identify the condition of pipe deterioration so the initiation of damage can be detected. The strain results from the current method and the available experimental results in the literature show excellent agreement. The average error between the ECNN data and FBG sensor data is 0.093%, thus confirming the reliability and accuracy of the proposed method. The proposed ECNN achieves high performance with 93.33 % accuracy (P%), 91.18 % regression rate (R%) and a 90.54 % F1-score (F%). [ABSTRACT FROM AUTHOR]

• Investigated uncertainties in FE model updating from measurement and modeling errors. • Employed information entropy measure criterion for optimum sensor placement in system identification. • Presented a novel nonlinear FE model updating framework using EKF and UKF. • Investigated uncertainties in nonlinear FE model updating framework. • Used information entropy measure criterion to determine optimal response utilization in FE model updating. In this work, we developed a framework for identifying frame-type structures regarding the measurement uncertainty and the uncertainty involved in inherent and structural parameters. The identification process is illustrated and examined on a one-eight-scale four-story moment-resisting steel frame under seismic excitation using two well-known recursive schemes: the Extended Kalman filter (EKF) and Unscented Kalman Filter (UKF) methods. The nonlinear system equations were assessed by applying a first-order instantaneous linearization approach through the EKF method. In contrast, the UKF algorithm employs several sample points to estimate moments of random variables' nonlinear transformations. A nonlinear transformation is applied to distribute sample points to derive the precise mean and covariance up to the second order of any nonlinearity. Accordingly, it is theoretically expected that the UKF algorithm is more capable of identifying the nonlinear systems and determining the unknown parameters than the EKF algorithm. The capability of the EKF and UKF algorithms was assessed by considering a 4-story moment-resisting steel frame with several inherent uncertainties, including the material behavior model, boundary conditions, and constraints. In addition to these uncertainties, the combination of acceleration and displacement responses of different structural levels is employed to evaluate the capability of the algorithms. The information entropy measure is used to investigate further the uncertainty of a group of established model parameters. As highlighted, a good agreement is observed between the results using the information entropy measure criterion and those using the UKF and EKF algorithms. The results illustrate that using the responses of fewer levels placed in the proper positions may lead to improved outcomes than those of more improperly positioned levels. [ABSTRACT FROM AUTHOR]