An official website of the United States government
This paper evaluates the ability of two different data-driven models to detect and localize simulated structural damage in an in-service bridge for long-term structural health monitoring (SHM). Strain gauge data collected over 4 years is used to characterize the undamaged state of the bridge. The Powder Mill Bridge in Barre, Massachusetts, U.S., which has been instrumented with strain gauges since its opening in 2009, is used as a case study, and the strain gauges used in this study are located at 26 different stations throughout the bridge superstructure. A linear regression (LR) model and an artificial neural network (ANN) model are evaluated based on the following criteria: (a) the ability to accurately predict the strain at each location in the undamaged state of the bridge; (b) the ability to detect simulated structural damage to the bridge superstructure; and (c) the ability to localize simulated structural damage. Both the LR and the ANN models were able to predict the strain at the 26 stations with an average error of less than 5%, indicating that both methodologies were effective in characterizing the undamaged state of the bridge. A calibrated finite element model was then used to simulate damage to the Powder Mill Bridge for three damage scenarios: fascia girder corrosion, girder fracture, and deck delamination. The LR model proved to be just as effective as the ANN model at detecting and localizing damage. A recommended protocol is thus presented for integrating data-driven models into bridge asset management systems.
more »
« less
Chapter 2 - Robust Output Only Health Monitoring of Steel Railway Bridges: Analysis of Applicability of Different Sensors
This chapter extends application of a framework proposed by the authors (73, 74) for automated damage detection using strain measurements to study feasibility of using sensors that can measure accelerations, tilts, and displacements. The study utilized three-dimensional (3D) finite element models of double track, riveted, steel truss span, and girder bridge span under routine train loads. The chapter also includes three instrumentation schemes for each bridge span (65) to investigate the applicability of the framework to other bridge systems and sensor networks. Connection damage was simulated by reducing rotational spring stiffness at member ends and various responses were extracted for each damage scenario. The methodology utilizes Supervised Machine Learning to automatically determine damage location (DL) and intensity (DI). Simulated experiments showed that DLs and DIs were detected accurately for both spans with various structural responses and using different instrumentation plans.
more »
« less
- Award ID(s):
- 1762034
- PAR ID:
- 10278186
- Editor(s):
- Gaur, L.; Solanki, A.; Jain, V.; Khazanchi, D.
- Date Published:
- Journal Name:
- Handbook of Research on Engineering Innovations and Technology Management in Organizations
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
This study focuses on developing and examining the effectiveness of Transfer Learning (TL) for structural health monitoring (SHM) systems that transfer knowledge about damage states from one structure (i.e., the source domain) to another structure (i.e., the target domain). Transfer Learning (TL) is an efficient method for knowledge transfer and mapping from source to target domains. In addition, Proper Orthogonal Modes (POMs), which help classify behavior and health, provide a promising tool for damage identification in structural systems. Previous investigations show that damage intensity and location are highly correlated with POM variations for structures under unknown loads. To train damage identification algorithms based on POMs and ML, one generally needs to use multiple simulations to generate damage scenarios. The developed process is applied to a simply supported truss span in a multi-span railway bridge. TL is first used to obtain relationships between POMs for two modeled bridges: one being a source model (i.e., labeled) and the other being the target modeled bridge (i.e., unlabeled). This technique is then implemented to develop POMs for a damaged, unknown target using TL that links source and target POMs. It is shown that the trained knowledge from one bridge was effectively generalized to other, somewhat similar, bridges in the population.more » « less
-
Bridges as a key component of road networks require periodic monitoring to detect structural degradation for early warning. In term of maintaining the bridge safety, it is essential to estimate the damage location and extent. This paper hypothetically investigates employing the wavelet transform to analysis the signal of a vehicle/bridge system to localize and estimate the damage severity. The paper investigated the feasibility of using direct measurements from the bridge system, in compare with using indirect measurements from a crossing inspection vehicle. The study utilizes an implicit Vehicle-Bridge Interaction (VBI) algorithm to simulate the passage of the instrumented vehicle over the bridge to generate the signal; then the signals are processed using Wavelet Transform. The study found that using the indirect vehicle measurements is more sensitive to bridge damage since the vehicle acts as a moving sensor over the bridge. Further, the paper shows promising results for damage detection using the bridge displacement responses, if the static component of the displacement is removed from the recorded displacement history.more » « less
-
A 2D plane strain extended finite element method (XFEM) model was developed to simulate three-point bending fracture toughness tests for human bone conducted in hydrated and dehydrated conditions. Bone microstructures and crack paths observed by micro-CT imaging were simulated using an XFEM damage model. Critical damage strains for the osteons, matrix, and cement lines were deduced for both hydrated and dehydrated conditions and it was found that dehydration decreases the critical damage strains by about 50%. Subsequent parametric studies using the various microstructural models were performed to understand the impact of individual critical damage strain variations on the fracture behavior. The study revealed the significant impact of the cement line critical damage strains on the crack paths and fracture toughness during the early stages of crack growth. Furthermore, a significant sensitivity of crack growth resistance and crack paths on critical strain values of the cement lines was found to exist for the hydrated environments where a small change in critical strain values of the cement lines can alter the crack path to give a significant reduction in fracture resistance. In contrast, in the dehydrated state where toughness is low, the sensitivity to changes in critical strain values of the cement lines is low. Overall, our XFEM model was able to provide new insights into how dehydration affects the micromechanisms of fracture in bone and this approach could be further extended to study the effects of aging, disease, and medical therapies on bone fracture.more » « less
-
El Mohtar, Chadi; Kulesza, Stacey; Baser, Tugce; Venezia, Michael D. (Ed.)Piles socketed into rock are frequently utilized to carry large loads from long-span bridges and high-rise buildings into solid ground. The pile design is derived from internal shear and moment magnitudes following code recommendation and numerical predictions. Little experimental data exist to validate code prescriptions and design assumptions for piles embedded in rock. To help alleviate the lack of large-scale test data, the lateral response behavior of three 18-in. diameter, 16 ft long, reinforced concrete piles was evaluated. The pile specimens were embedded in a layer of loose sand and fixed in “rock-sockets,” simulated through high strength concrete. The construction sequence simulated soil-pile interface stress conditions of drilled shafts. The pile reinforcement varied to satisfy the internal reaction forces per (1) code requirements, (2) analytical SSI predictions, and (3) structural demands only. The pile specimens were tested to complete structural failure and excavated thereafter. Internal instrumentation along with crack patterns suggested a combined shear-flexural failure, but do not support the theoretically predicted amplification and de-amplification of shear and moment forces at the boundary, respectively.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.4018/978-1-7998-2772-6
