Toward reduced recovery time after extreme events, near real‐time damage diagnosis of structures is critical to provide reliable information. For this task, a fully convolutional encoder–decoder neural network is developed, which considers the spatial correlation of sensors in the automatic feature extraction process through a grid environment. A cost‐sensitive score function is designed to include the consequences of misclassification in the framework while considering the ground motion uncertainty in training. A 10‐story‐10‐bay reinforced concrete (RC) moment frame is modeled to present the design process of the deep learning architecture. The proposed models achieve global testing accuracies of 96.3% to locate damage and 93.2% to classify 16 damage mechanisms. Moreover, to handle class imbalance, three strategies are investigated enabling an increase of 16.2% regarding the mean damage class accuracy. To evaluate the generalization capacities of the framework, the classifiers are tested on 1,080 different RC frames by varying model properties. With less than a 2% reduction in global accuracy, the data‐driven model is shown to be reliable for the damage diagnosis of different frames. Given the robustness and capabilities of the grid environment, the proposed framework is applicable to different domains of structural health monitoring research and practice to obtain reliable information.
Computer vision leveraging deep learning has achieved significant success in the last decade. Despite the promising performance of the existing deep vision inspection models, the extent of models’ reliability remains unknown. Structural health monitoring (SHM) is a crucial task for the safety and sustainability of structures, and thus, prediction mistakes can have fatal outcomes. In this paper, we use Bayesian inference for deep vision SHM models where uncertainty can be quantified using the Monte Carlo dropout sampling. Three independent case studies for cracks, local damage identification, and bridge component detection are investigated using Bayesian inference. Aside from better prediction results, the two uncertainty metrics, variations in softmax probability and entropy, are shown to have good correlations with misclassifications. However, modifying the decision or triggering human intervention can be challenging based on raw uncertainty outputs. Therefore, the concept of surrogate models is proposed to develop the models for uncertainty‐assisted segmentation and prediction quality tagging. The former refines the segmentation mask and the latter is used to trigger human interventions. The proposed framework can be applied to future deep vision SHM frameworks to incorporate model uncertainty in the inspection processes.
Rapid condition monitoring of structural health is essential for post-earthquake safety assessment. Therefore, information about damage after extreme events could be of great value in resilient communities. This paper proposes a robust framework for the identification of the existence, probable location, and severity of damage using cumulative intensity-based damage features. Taken into account the seismic hazard uncertainties and the undesirable consequences of misclassification in data-driven methods, an objective function based on the confusion score matrix is optimized. This process can enhance the reliability of the prediction model in terms of two conflicting criteria, namely, the general accuracy and conservativeness. Support vector machines are utilized for the task of damage classification where Bayesian optimization is used to select the proper damage-sensitive input features and hyperparameters. A threestory reinforced concrete (RC) moment frame is designed and modeled in OpenSees to examine the performance of the proposed framework. For this purpose, 5,400 incrementally scaled nonlinear time history analyses are conducted considering 180 ground motions. Two different approaches are introduced to distort signals to simulate measurement noise. The robustness of models is also investigated with respect to different objective functions. Furthermore, in an independent experimental case study, the framework is evaluated on a dataset obtained from 44 shake table trials on a three-story RC frame with masonry infill. Given the results obtained from the two case studies, it is shown that the proposed framework is capable of robust and reliable identification of damage in near real-time whereas the concepts of hazard uncertainty and misclassification consequences are properly considered.
Reduced time to recovery is one of the fundamental characteristics of resilient systems. Bridge infrastructures, as critical components in a transportation system, play an important role in large communities. Such structures should maintain their functionality in harsh environments especially after extreme events (e.g., earthquakes). Given the number of aging infrastructure across the united states, structural health monitoring (SHM) techniques have been widely used to periodically monitor the condition of bridges. Visual inspections are one of the most common ways of condition assessment where this task is conventionally performed by dedicated teams. There are several drawbacks for human inspections. Having dedicated teams for this purpose requires time and monetary resources that may not be readily available after disasters. Moreover, bridges are commonly built in harsh geographical locations to facilitate transportation. Critical structural components may not be easily accessible for manual investigations of damage. That being said, most visual inspections are inaccurate and biased (Phares 2001) while reliable information about bridge condition is essential to the decision makers. To address these issues, automated SHM has been the topic of interest in many studies (Spencer et al. 2019). Cameraequipped unmanned aerial vehicles can be effectively used in this regard. However, obtaining useful information from raw images is still challenging.With the rapid progress of the research in the field of artificial intelligence, recent deep learning models have been capable of classifying object within raw images. Proposed algorithms are mainly designed to detect common objects such as pedestrians, roads, etc. With a similar
Optimal sensor placement (OSP) is essential for effective structural health monitoring (SHM). More recently, deep learning algorithms have shown great potential in sensor-based SHM. However, existing optimization frameworks, such as population-based algorithms, are often not suited for data-driven SHM. Evaluating a number of sensor layouts includes training on large datasets, which is computationally expensive. This paper proposes deep generative Bayesian optimization (DGBO) as a solution for a parallel optimization of black-box/expensive OSP objective functions. Conditional variational autoencoders are leveraged as generative models that transform the OSP problem into a lower-dimensional latent space. Additionally, DGBO utilizes a surrogate neural network to capture the probability distribution of the objective function space. The proposed method is validated on two case studies on a nine-story reinforced concrete moment frame. The first one serves as a proof of concept to show that DGBO can find the global optimum configuration. The second case study aims to maximize the semantic damage segmentation (SDS) accuracy using a fully convolutional neural network. Transfer learning is proposed in training the vibration-based SDS model, which reduces the evaluation times by more than 50%. Without compromising the performance, the number of accelerometers can be reduced by 52% and 43%, respectively, for damage location and severity predictions. It is also shown that DGBO can outperform genetic algorithm with the same number of function evaluations. DGBO can serve as a scalable solution to address the highdimensionality challenge in OSP for large-scale civil infrastructure.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2025 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.