No abstract

Summary This paper proposed a modified faster region‐based convolutional neural network (faster R‐CNN) for the multitype seismic damage identification and localization (i.e., concrete cracking, concrete spalling, rebar exposure, and rebar buckling) of damaged reinforced concrete columns from images. Four hundred raw images containing different damages and complicated background information are taken by a consumer‐grade camera in various locations and arbitrary perspectives to simulate the diverse situations where real‐world postearthquake damaged structural images are taken by nonprofessionals. Rectangular bounding boxes are obtained to localize multitype structural damages along with the corresponding category labels and classification probabilities. Data augmentation is implemented by rotation at every 90°, vertical and horizontal flipping operations. An interactive labeling process for the ground‐truth regions of the aforementioned damages is performed by a semiautomatic MATLAB program. A four‐step alternating training procedure is adopted on the basis of the mini‐batch stochastic gradient decent algorithm with momentum by backpropagation. Test results show that the trained faster R‐CNN can automatically identify and localize the aforementioned multitype seismic damages and the overall average precision reaches 80%. The relative errors of coordinates of the left‐top point obey minimum extreme value distributions, and those of width and height obey three‐parameter lognormal distributions. The intersection ratio between the identification and ground truth has a mean value of 0.88, and the width–height ratio obeys a two‐parameter lognormal distribution. Updated convolutional kernels in the first layer have shown trending, focusing, and line detectors for the feature extraction of multitype damages. Trending and focusing detectors contribute to the recognition of local damage regions, for example, concrete spalling and rebar exposure, whereas line detectors are more sensitive to the segmentation geometry, that is, concrete cracks.

No abstract

No abstract

Strain is a direct indicator of structural safety. Therefore, strain sensors have been used in most structural health monitoring systems for bridges. However, until now, the investigation of strain response has been insufficient. This paper conducts a comprehensive study of the strain features of the U ribs and transverse diaphragm on an orthotropic steel deck and proposes a statistical paradigm for crack detection based on the features of vehicle-induced strain response by using the densely distributed optic fibre Bragg grating (FBG) strain sensors. The local feature of strain under vehicle load is highlighted, which enables the use of measurement data to determine the vehicle loading event and to make a decision regarding the health status of a girder near the strain sensors via technical elimination of the load information. Time–frequency analysis shows that the strain contains three features: the long-term trend item, the short-term trend item, and the instantaneous vehicle-induced item (IVII). The IVII is the wheel-induced strain with a remarkable local feature, and the measured wheel-induced strain is only influenced by the vehicle near the FBG sensor, while other vehicles slightly farther away have no effect on the wheel-induced strain. This causes the local strain series, among the FBG strain sensors in the same transverse locations of different cross-sections, to present similarities in shape to some extent and presents a time delay in successive order along the driving direction. Therefore, the strain series induced by an identical vehicle can be easily tracked and compared by extracting the amplitude and calculating the mutual ratio to eliminate vehicle loading information, leaving the girder information alone. The statistical paradigm for crack detection is finally proposed, and the detection accuracy is then validated by using dense FBG strain sensors on a long-span suspension bridge in China.

AbstractsPhysicists use differential equations to describe the physical dynamical world, and the solutions of these equations constitute our understanding of the world. During the hundreds of years, scientists developed several ways to solve these equations, i.e., the analytical solutions and the numerical solutions. However, for some complex equations, there may be no analytical solutions, and the numerical solutions may encounter the curse of the extreme computational cost if the accuracy is the first consideration. Solving equations is a high-level human intelligence work and a crucial step towards general artificial intelligence (AI), where deep reinforcement learning (DRL) may contribute. This work makes the first attempt of applying (DRL) to solve nonlinear differential equations both in discretized and continuous format with the governing equations (physical laws) embedded in the DRL network, including ordinary differential equations (ODEs) and partial differential equations (PDEs). The DRL network consists of an actor that outputs solution approximations policy and a critic that outputs the critic of the actor's output solution. Deterministic policy network is employed as the actor, and governing equations are embedded in the critic. The effectiveness of the DRL solver in Schrödinger equation, Navier-Stocks, Van der Pol equation, Burgers' equation and the equation of motion are discussed.

No abstract

Data-driven methods have shown promising results in structural health monitoring (SHM) applications. However, most of these approaches rely on the ideal dataset assumption and do not account for missing data, which can significantly impact their real-world performance. Missing data is a frequently encountered issue in time series data, which hinders standardized data mining and downstream tasks such as damage identification and condition assessment. While imputation approaches based on spatiotemporal relations among monitoring data have been proposed to handle this issue, they do not provide additional helpful information for downstream tasks. This paper proposes a robust deep learning-based method that unifies missing data imputation and damage identification tasks into a single framework. The proposed approach is based on a long short-term memory (LSTM) structured autoencoder (AE) framework, and missing data is simulated using the dropout mechanism by randomly dropping the input channels. Reconstruction errors serve as the loss function and damage indicator. The proposed method is validated using the quasi-static response (cable tension) of a cable-stayed bridge released in the 1st IPC-SHM, and results show that missing data imputation and damage identification can be effectively integrated into the proposed unified framework.