The widespread application of sophisticated structural health monitoring systems in civil infrastructures produces a large volume of data. As a result, the analysis and mining of structural health monitoring data have become hot research topics in the field of civil engineering. However, the harsh environment of civil structures causes the data measured by structural health monitoring systems to be contaminated by multiple anomalies, which seriously affect the data analysis results. This is one of the main barriers to automatic real-time warning, because it is difficult to distinguish the anomalies caused by structural damage from those related to incorrect data. Existing methods for data cleansing mainly focus on noise filtering, whereas the detection of incorrect data requires expertise and is very time-consuming. Inspired by the real-world manual inspection process, this article proposes a computer vision and deep learning–based data anomaly detection method. In particular, the framework of the proposed method includes two steps: data conversion by data visualization, and the construction and training of deep neural networks for anomaly classification. This process imitates human biological vision and logical thinking. In the data visualization step, the time series signals are transformed into image vectors that are plotted piecewise in grayscale images. In the second step, a training dataset consisting of randomly selected and manually labeled image vectors is input into a deep neural network or a cluster of deep neural networks, which are trained via techniques termed stacked autoencoders and greedy layer-wise training. The trained deep neural networks can be used to detect potential anomalies in large amounts of unchecked structural health monitoring data. To illustrate the training procedure and validate the performance of the proposed method, acceleration data from the structural health monitoring system of a real long-span bridge in China are employed. The results show that the multi-pattern anomalies of the data can be automatically detected with high accuracy.

Structural health monitoring (SHM) is used worldwide for managing and maintaining civil infrastructures. SHM systems have produced huge amounts of data, but the effective monitoring, mining, and utilization of this data still need in‐depth study. SHM data generally includes multiple types of anomalies caused by sensor faults or system malfunctions that can disturb structural analysis and assessment. In the routine data pre‐processing, multiple signal processing techniques are required to detect the anomalies, respectively, which is inefficient. The large variations of extracted features from massive SHM data make the data anomaly detection techniques prone to be over‐processed or under‐processed. Even with expert intervention, the parameter tuning, associated with multiple data preprocessing methods, is still a challenge, making the procedure expensive and inefficient. In addition, one data anomaly detection technique frequently mis‐detects other types of anomaly. In this work, we focus on the anomaly detection in the stage of data pre‐processing that little work has been done based on the real‐world continuous SHM data with multiclass anomalies. We proposed a novel data anomaly detection method based on a convolutional neural network (CNN) that imitates human vision and decision making. First, we split raw time series data into sections, and visualized the data in time and frequency domain, respectively. Then each section's images were stacked as a single dual‐channel image and labeled according to graphical features (multi‐2D image space expression). Second, a CNN was designed and trained for data anomaly classification, during which the descriptions and representations of the anomalies' features were learned by convolution. To validate our work, we considered the effects of balanced and imbalanced training sets and training ratios on actual acceleration data from an SHM system for a long‐span cable‐stayed bridge. The results show that our method could detect the multipattern anomalies of SHM data efficiently with high accuracy. The proposed dual‐information CNN‐based design makes this detection process readily scalable, faster, and more accurate, thereby providing a novel perspective with strong potential for SHM data preprocessing.

Compressive sensing has been studied and applied in structural health monitoring for data acquisition and reconstruction, wireless data transmission, structural modal identification, and spare damage identification. The key issue in compressive sensing is finding the optimal solution for sparse optimization. In the past several years, many algorithms have been proposed in the field of applied mathematics. In this article, we propose a machine learning–based approach to solve the compressive-sensing data-reconstruction problem. By treating a computation process as a data flow, the solving process of compressive sensing–based data reconstruction is formalized into a standard supervised-learning task. The prior knowledge, i.e. the basis matrix and the compressive sensing–sampled signals, is used as the input and the target of the network; the basis coefficient matrix is embedded as the parameters of a certain layer; and the objective function of conventional compressive sensing is set as the loss function of the network. Regularized by l₁-norm, these basis coefficients are optimized to reduce the error between the original compressive sensing–sampled signals and the masked reconstructed signals with a common optimization algorithm. In addition, the proposed network is able to handle complex bases, such as a Fourier basis. Benefiting from the nature of a multi-neuron layer, multiple signal channels can be reconstructed simultaneously. Meanwhile, the disassembled use of a large-scale basis makes the method memory-efficient. A numerical example of multiple sinusoidal waves and an example of field-test wireless data from a suspension bridge are carried out to illustrate the data-reconstruction ability of the proposed approach. The results show that high reconstruction accuracy can be obtained by the machine learning–based approach. In addition, the parameters of the network have clear meanings; the inference of the mapping between input and output is fully transparent, making the compressive-sensing data-reconstruction neural network interpretable.

In structural health monitoring, data quality is crucial to the performance of data-driven methods for structural damage identification, condition assessment, and safety warning. However, structural health monitoring systems often suffer from data imperfection, resulting in some entries being unusable in a data matrix. Discrete missing points are relatively easy to recover based on known adjacent points, whereas segments of continuous missing data are more common and also more challenging to recover in a practical scenario. Formulating the data recovery task as an optimization problem for matrix completion, we present a convolutional neural network to achieve simultaneous recovery for multi-channel data with the awareness of group sparsity. The data recovery process based on compressive sensing is formulated as a regression problem and achieved in the neural network. The basis matrix is utilized as the input and the incomplete data matrix as the output to provide partial information for approximation. Basis coefficient optimization is performed via convolutional operation. Group sparsity regularization is applied while updating the kernel of the convolutional layer. The recovery can be readily obtained after optimization (training) without further validation and testing. The proposed method does not need intact data prepared in advance for training; also, it can handle sporadic data loss and make the most of interrupted information. Recovery ability evaluations on synthetic data, field-test data, and monitoring data of seismic response indicate that the proposed method achieves a good recovery result with high loss ratio and continuous data loss. The code is available at https://github.com/dawnnao/Group-sparsity-aware-CNN .