Search

Distributed optical fiber acoustic sensing (DAS) technique has been applied in pipeline health monitoring, and the commonly used sensor is phase-sensitive optical time domain reflectometry. Most DAS monitoring systems can localize leakages of a pipeline but fail to identify potential non-destructive damages like bolt looseness on joints before the leakage occurs. An early damage identification is indispensable to averting severe leakages and secondary disasters. In this study, an optical phase mode analysis method is proposed for identifying pipeline bolt looseness. This method combines structure mode analysis and distributed optical phase demodulation to extract damage-related phase mode parameters. Two algorithms are specially designed for denoising and selecting signals essential for mode analysis. Phase time histories are retrieved from the original optical phase, which are decomposed to acquire phase mode shapes that can localize bolt looseness through Hilbert-Huang transform enhanced with bandwidth restricted empirical mode decomposition. Phase damping ratio is proposed to further quantify the looseness degree. Polarization diversity technique is employed to avoid polarization fading. An experiment was conducted upon a 3.2 m steel pipeline with flange joints. Bolt looseness on three joints are respectively localized even if only one bolt is loosened, obtaining a localization error of 0.07 m and 85.7% recognition ratio. The phase damping ratio shows apparent positive correlation with the number of loose bolts. The error of quantified loose bolt number is 0.79. The present study demonstrates how to localize and quantify pipeline bolt looseness through dynamical mode analysis for distributed optical phase. The developed method can identify potential damages that change the mechanical properties of a pipeline before they get severe, and holds promise in the long-distance health monitoring of other structures. [ABSTRACT FROM AUTHOR]

Deep learning anomaly detection is important in distributed optical fiber acoustic sensing (DAS). However, anomaly detection is more challenging than traditional learning tasks, due to the scarcity of true-positive data and the vast imbalance and irregularity within datasets. Furthermore, it is impossible to catalog all types of anomalies, therefore, the direct application of supervised learning is deficient. To overcome these problems, an unsupervised deep learning method that only learns the normal data features from ordinary events is proposed. First, a convolutional autoencoder is used to extract DAS signal features. A clustering algorithm then locates the feature center of the normal data, and the distance to the new signal is used to determine whether it is an anomaly. The efficacy of the proposed method was evaluated in a real high-speed rail intrusion scenario, and considered all behaviors that may threaten the normal operation of high-speed trains as abnormal. The results show that the threat detection rate of this method reaches 91.5%, which is 5.9% higher than that of the state-of-the-art supervised network and, at 7.2%, the false alarm rate is 0.8% lower than the supervised network. Moreover, using a shallow autoencoder reduces the parameters to 1.34 K, which is significantly lower than the 79.55 K of the state-of-the-art supervised network. [ABSTRACT FROM AUTHOR]

Due to high spatial resolution, low cost, and wide bandwidth, distributed optical fiber acoustic sensing (DAS) is regarded as a potential tool for data acquisition in vertical seismic profile (VSP) surveys. However, in real DAS-VSP records, desired signals are often seriously plagued by various noise, which does not appear in the conventional seismic data received by electronic geophones. Exploring a high-performing attenuation method for the background noise can significantly improve the quality of DAS-VSP records and has essential impacts on the following imaging and interpretation. Deep-learning-based methods, especially convolutional neural network (CNN), have shown remarkable performance in seismic data denoising. However, the conventional CNN-based methods may degrade when dealing with DAS-VSP records in low signal-to-noise ratio. In this study, we propose a novel multi-scale dense-connection denoising network (MDD-Net) to achieve high-accuracy processing of the complex DAS background noise. Unlike conventional multi-scale networks, MDD-Net utilizes widen convolution block to capture the multi-scale features of the analyzed data. On this basis, dense connection operations are employed to fuse the features and improve the network efficiency. Meanwhile, an enhanced spatial attention (ESA) block is designed to reinforce the features, which are helpful for noise suppression and weak signal recovery. Both synthetic and field DAS-VSP records are processed to verify the effectiveness of MDD-Net. Meanwhile, we also compare the denoising results with other competing methods. The experimental results demonstrate that MDD-Net can significantly attenuate the complex DAS background noise and restore the desired signals, even for the weak upgoing signals.

In recent years, distributed optical fiber acoustic sensing (DAS) technology has been increasingly used for vertical seismic profile (VSP) exploration. Even though this technology has the advantages of high spatial resolution, strong resistance to high temperature and pressure variations, long sensing distance, DAS seismic noise has expanded from random noise to optical abnormal noise, fading noise and horizontal noise, etc. This seriously affects the quality of the seismic data and brings huge challenges to subsequent imaging, inversion and interpretation. Moreover, the noise is more complex and more difficult to simultaneously suppress using traditional methods. Therefore, for the purpose of effectively improving the signal-to-noise ratio (SNR) of DAS seismic data, we introduce a denoising network named attention-guided denoising convolutional neural network (ADNet). The network is composed of four blocks, including a sparse block (SB), a feature enhancement block (FEB), an attention block (AB) and a reconstruction block (RB). The network uses different kinds of convolutions alternately to enlarge the receptive field size and extract global feature of the input. Meanwhile, the attention mechanism is introduced to extract the hidden noise information in the complex background. The network predicts the noise, and denoised data are obtained by subtracting the predicted results from the noisy inputs. In addition, we uniquely construct a large number of complex forward models for pure seismic data training set to enhance the network suitability. The combination design improves the denoising performance and reduces computational cost and memory consumption. The results obtained from both synthetic- and field data illustrate that the network has the ability to denoise the seismic images and retrieve weak effective signals better than conventional methods and common networks.

Deep learning anomaly detection is important in distributed optical fiber acoustic sensing (DAS). However, anomaly detection is more challenging than traditional learning tasks, due to the scarcity of true-positive data and the vast imbalance and irregularity within datasets. Furthermore, it is impossible to catalog all types of anomalies, therefore, the direct application of supervised learning is deficient. To overcome these problems, an unsupervised deep learning method that only learns the normal data features from ordinary events is proposed. First, a convolutional autoencoder is used to extract DAS signal features. A clustering algorithm then locates the feature center of the normal data, and the distance to the new signal is used to determine whether it is an anomaly. The efficacy of the proposed method was evaluated in a real high-speed rail intrusion scenario, and considered all behaviors that may threaten the normal operation of high-speed trains as abnormal. The results show that the threat detection rate of this method reaches 91.5%, which is 5.9% higher than that of the state-of-the-art supervised network and, at 7.2%, the false alarm rate is 0.8% lower than the supervised network. Moreover, using a shallow autoencoder reduces the parameters to 1.34 K, which is significantly lower than the 79.55 K of the state-of-the-art supervised network.

Earthquake early warning can provide important information before the occurrence of destructive disasters to reduce the possible loss, and the requirements for earthquake identification time and accuracy are strict. As a new seismic monitoring technology, distributed optical fiber acoustic sensing (DAS) has the advantages of wide coverage and spatially dense detection. DAS can be deployed in earthquake-prone areas, thus saving the time for seismic waves to reach the sensors. In order to reduce the recognition time and improve the accuracy, this paper proposes a fast earthquake recognition method based on DAS and quickly recognition earthquake network (QRE-net). After preprocessing the seismic data collected in DAS, QRE-net identifies the seismic signals. The effectiveness of the proposed method is tested on natural seismic data collected by DAS, when DAS detect 1s seismic signals, recognition accuracy reaches 88.74%. It is believed that this method will play an important role in earthquake early warning. • Deep learning is used to identify the seismic signals monitored by DAS. • A neural network architecture of QER-net is proposed to identify 1s seismic signals. • It provides an idea for DAS to be used in earthquake early warning. [ABSTRACT FROM AUTHOR]

In intelligent transportation systems, distributed acoustic sensing offers unparalleled advantages in monitoring and analyzing vehicle characteristics and behaviors in real time over the entire optical fiber. In this work, an accurate and efficient φ -optical time domain reflectometer-based load recognition method for light and heavy loaded vehicles is proposed. Before load recognition, wavelet denoising and 1D-mean filtering methods are used to denoise the signals; then the Mel spectrograms of the signals are extracted as the features input to the load recognition model with a backbone of EfficientNet convolutional neural network. The validation results show that, using an ∼47 km sensing optical fiber, the recognition of light and heavy loaded vehicles can well meet the needs of real-time data analysis and decision making of intelligent transportation, with an average recognition accuracy of 97.81% within 14 ms for each recognition. [ABSTRACT FROM AUTHOR]