Vibration Events Research Articles

Deep learning-based automated identification on vortex-induced vibration of long suspenders for the suspension bridge

Large amplitude vortex-induced vibration (VIV) of long suspenders, caused by vortex-shedding and wake flow, will decrease the service life of high-strength wires and its anchorage systems, hence listed as an essential monitoring indicator of the structural health monitoring system in the suspension bridges. Traditionally, a static vibration amplitude limitation is usually predefined and used to identify VIV from continuously monitored acceleration data of bridge suspenders. Due to the complexity of the starting vibration of the suspenders, it may not be able to flexibly adapt to the suspenders VIV under different conditions. In addition, transient but significant vibration events may not be captured since it relies only on static vibration amplitude limits without considering the temporal information of vibration. To address the above-mentioned issues, a novel framework to autonomously identify the suspenders VIV is proposed using the multimodal fusion techniques with deep neural networks. The main contributions of the methodology are: i) by calculating the wind field and vibration characteristic, two vibration response parameters are identified as indexes for the identification of suspenders VIV based on the significant difference method of big data analysis; ii) Two different modal information of time history images and power spectral density (PSD) sequences are used as inputs to the convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM) to extract the suspenders vibration response features from the time and frequency domain perspectives, respectively; and iii) The multimodal information is concentrated and enhanced by the multi-head attention (MHA) mechanism to achieve automated identification of the suspenders VIV. The proposed framework is validated with data collected from a full-scale long-span suspension bridge in comparison with the base model. The results show that the proposed framework can efficiently identify the full-cycle development process of the suspenders VIV. This study provides a convenient strategy for suspenders VIV identification, which can be used for bridge management and vibration mitigation device design.

Underwater sound produced by operational floating wind turbines

Deployment of offshore wind turbines is growing rapidly. Conventional "fixed bottom" wind turbines are limited to water depths of ~60 m, so deployment of offshore turbines to greater water depths requires floating systems. Both fixed bottom and floating turbines produce operational noise that impacts the marine environment. While the operational noises from fixed bottom turbines are relatively well understood, floating turbines are comparatively novel technologies and there is uncertainty as to the noise they will produce and its impact on the marine environment. Results from acoustic surveys of two floating wind farms off the Scottish Coast are presented; Kincardine wind farm which uses semi-submersible floating systems; and Hywind which uses a floating spar structure. The acoustic signature of each consists of continuous tonal noise related to drivetrain vibration and transient broadband events related to the mooring systems. A machine learning algorithm was used to discriminate between drivetrain noise and mooring noise. Sound propagation modelling was used to determine the source levels for turbines at each wind farm allowing a comparison of the noise signature of different floating structures.

Vision-based digital shadowing to reveal hidden structural dynamics of a real supertall building

Vision-based digital shadowing is a highly efficient way to monitor the health of buildings in use. However, previous studies on digital shadowing have been limited to laboratory experiments. This paper proposes a novel computer-vision-based digital shadow workflow and presents its successful application in a real engineering case. In this case, a 345.8-m supertall building experienced unexpected shaking under normal meteorological conditions. This study established a digital shadow of the building using three-dimensional displacement measurements based on super-resolution monocular vision, revealing the hidden structural dynamics and inherent mechanical reasons for the abnormal shaking. The proposed digital shadowing workflow is a feasible roadmap for developing vision-based digital shadows of real-world structures using low-cost cameras. The abnormal vibration event in the supertall building considered in this study is the first of its type worldwide. The results of this study offer practical strategies and invaluable insights into the prevention and mitigation of this type of global risk, thereby contributing to the lifespan extension of buildings in use worldwide. Furthermore, with the increasing number of general sensing devices, such as surveillance cameras in cities, the proposed method may unleash the immense potential of general sensing devices in achieving the leap from structural health monitoring to city health monitoring.

An improved denoising method for [formula omitted]-OTDR signal based on the combination of temporal local GMCC and ICEEMDAN-WT

A noise reduction method based on the combination of improved complete ensemble empirical mode decomposition with adaptive noise and temporal local generalized maximum cross-correlation entropy and wavelet threshold (ICEEMDAN-TLGMCC-WT) is proposed to improve the noise rejection ratio (NRR) of phase-sensitive optical time-domain reflectometer (φ-OTDR) system signals. Firstly, the principle of the proposed ICEEMDAN-TLGMCC-WT denoising method is introduced. On this basis, experiments are carried out to verify the feasibility of the noise reduction algorithm for non-stationary signals. and the results demonstrate that the vibration event can be detected with an improved average NRR of 72.18 dB and a root mean square error (RMSE) 0.017 within 5 km sensing distance by employing the ICEEMDAN-TLGMCC-WT denoising method, which confirm the effectiveness of proposed scheme for the performance improvement of φ-OTDR system.

Reproducing vortex-induced vibrations of rooftop twin-mast by multi-scale coupled simulation of urban wind fields

The Shenzhen Electronics Group (SEG) Plaza in Shenzhen, China, experienced visible vibrations attributed to significant vortex-induced vibrations of the large spacing twin-mast atop the building. The vibration event occurred under complex urban wind environments. In order to reproduce wind fields around SEG Plaza, a multiscale simulation was carried out by integrating the Weather Research and Forecasting (WRF) and Computational Fluid Dynamics (CFD) models. The aerodynamic characteristics and wake modes of twin-mast are investigated by a downscale CFD simulation in detail. Finally, vortex-induced vibration (VIV) of twin-mast was analyzed considering fluid-structure interaction. This work reveals that wind conditions during the SEG Plaza vibration incident were appropriate for the VIV occurrence of twin-mast, and the results of VIV simulation agree well with the field monitoring results by computer vision identification. This study effectively reproduces the full pictures of VIVs of the rooftop twin-mast during this rare event, offering a critical view for the analysis of this event.

A hand-crafted φ-OTDR event recognition method based on space-temporal graph and morphological object detection

In the long-range pipeline monitoring application of Fiber Optic Distributed Vibration Sensor (FODVS), instantaneous vibration event with high energy generates a pair of invert-V signature in the space-temporal graph, providing clear characteristic for event recognition. This paper proposed an event recognition method utilizing FODVS and hand-crafted morphological object detection algorithm. Phase-sensitive Optical Time Domain Reflectometer (φ-OTDR) was employed as the distributed sensor to collect the real field space-temporal graph data containing invert-V signature. Bandpass filter and Gaussian blur were utilized to suppress random noise and enhance the invert-V signature, then a hand-crafted morphological object detection method was proposed to recognize the invert-V pattern. The proposed method was proved to be able to accurately capture the invert-V signature in the space-temporal graph. This paper provided a concept-proof preliminary demonstration on the hand-crafted FODVS event recognition scheme for pursuing model with good interpretability and strong adaptability to varied environments.

Distributed vibration and temperature sensing system by multiplexed fiber scattering spectra

A new, to the best of our knowledge, distributed optical fiber vibration and temperature hybrid sensing system is proposed and experimentally demonstrated. The proposed system only employs two signal channels, which is more compact and practical. Based on the structure of the optical time domain reflectometer (OTDR), the Rayleigh scattering light and the Raman anti-Stokes scattering light is extracted for vibration and temperature sensing, respectively. For vibration sensing, a new differential location algorithm based on polarization state analysis of the Rayleigh scattering light is proposed to locate the vibration events. It first rectifies the original OTDR traces by fiber attenuation compensation to make each position in it with the same pulse power level. And then, by difference of adjacent traces and threshold discrimination, the vibration positions are identified and located. For temperature sensing, a temperature calibration unit and algorithm are adopted to dynamically correct the trace data and reduce the temperature measurement error caused by the instability of the pulse laser source. The experiment is conducted with a fiber range of about 12 km and laser pulse width of 60 ns, and the experimental results show that the maximum error range for temperature measurement is −0.7∘C to 1.3°C, with a root mean square (RMS) error of 0.85°C in the entire temperature measurement range. Additionally, the spatial resolution (SR) for both vibration and temperature sensing is 6 m.

The Frequency Characteristics of Vibration Events in an Underground Coal Mine and Their Implications on Rock Burst Monitoring and Prevention

The main frequency of microseismic signals has recently been identified as a dominant indicator for characterizing vibration events because it reflects the energy level of these events. Frequency information directly determines whether effective signals can be collected, which has a significant impact on the accuracy of predicting rock burst disasters. In this study, we adopted a characterizing method and developed a monitoring system for capturing rock failure events at various strata in an underground coal mine. Based on the rock break mechanism and energy release level, three types of rock failure events, namely, high roof breaking, low roof breaking, and coal fracture events, were evaluated separately using specific sensors and monitoring systems to optimize the monitoring accuracy and reduce the general cost. The captured vibration signals were processed and statistically analyzed to characterize the main frequency features for different rock failure events. It was found that the main frequency distribution ranges of low roof breaking, high roof breaking, and coal fracture events are 20–400 Hz, 1–180 Hz, and 1–800 Hz, respectively. Therefore, these frequency ranges are proposed to monitor different vibration events to improve detection accuracy and reduce the test and analysis times. The failure mechanism in a high roof is quite different from that of low roof failure and coal fracturing, with the main frequency and amplitude clustering in a limited zone close to the origin. Coal fracturing and lower roof failure show a synergistic effect both in the maximum amplitude and main frequency, which could be an indicator to distinguish failure locations in the vertical direction. This result can support the selection and optimization of the measurement range and main frequency parameters of microseismic monitoring systems. This study also discussed the distribution law of the maximum amplitude and main frequency of different events and the variation in test values with the measurement distance, which are of great significance in expanding the application of optimized microseismic monitoring systems for rock burst monitoring and prevention.

A novel asymmetric tri-stable piezoelectric vibration energy harvester for low-orbit vibration energy harvesting enhancement

This study introduces an innovative asymmetric tri-stable piezoelectric vibration energy harvester (ATPVEH) augmented with an elastic base (EB), aimed at optimizing energy capture from low-orbit vibrations. This design incorporates a uniquely configured asymmetric tri-stable piezoelectric cantilever beam, positioned within a U-shaped block that is further enhanced by an elastic base. A strategically placed spring (kf)–mass (Mf) system, situated between the U-shaped block and the constrained end of the beam, significantly boosts the vertical displacement of the beam during vibrational events. We developed a dynamic model for the ATPVEH+EB utilizing Lagrange’s equations, exploring the impact of various factors—including the asymmetry of the potential well, the stiffness of the elastic base, the mass of the spring–mass system, and the load resistance—on the system’s nonlinear dynamic responses. Our findings indicate that the ATPVEH+EB facilitates more efficient energy harvesting from low-orbit vibrations, demonstrating dual response peaks across its operational frequency spectrum. Notably, the displacement and output voltage amplitudes of the ATPVEH+EB can be enhanced by increasing mf or decreasing kb, whereas the peak output power transitions toward a lower frequency range as the load resistance escalates. Both theoretical analyses and numerical simulations corroborate the ATPVEH+EB’s superior performance in harvesting energy within low-orbit vibrational environments.

Water Pipeline Leakage Detection Based on Coherent φ-OTDR and Deep Learning Technology

Leakage in water supply pipelines remains a significant challenge. It leads to resource and economic waste. Researchers have developed several leak detection methods, including the use of embedded sensors and pressure prediction. The former approach involves pre-installing detectors inside pipelines to detect leaks. This method allows for the precise localization of leak points. The stability is compromised because of the wireless signal strength. The latter approach, which relies on pressure measurements to predict leak events, does not achieve precise leak point localization. To address these challenges, in this paper, a coherent optical time-domain reflectometry (φ-OTDR) system is employed to capture vibration signal phase information. Subsequently, two pre-trained neural network models based on CNN and Resnet18 are responsible for processing this information to accurately identify vibration events. In an experimental setup simulating water pipelines, phase information from both leaking and non-leaking pipe segments is collected. Using this dataset, classical CNN and ResNet18 models are trained, achieving accuracy rates of 99.7% and 99.5%, respectively. The multi-leakage point experiment results indicate that the Resnet18 model has better generalization compared to the CNN model. The proposed solution enables long-distance water-pipeline precise leak point localization and accurate vibration event identification.

Long-range distributed vibration sensing based on internal-modulation OFDR

Presented here is long-range distributed vibration sensing based on internal-modulation optical frequency domain reflectometry (OFDR). In the proposed system with internal modulation, a silicon-based photonic-chip laser is used as the laser source, and by controlling the output voltage curve of an arbitrary waveform generator to induce temperature change in the external cavity of the laser, a 10-GHz optical frequency tuning range is achieved. The complexity of the proposed internal-modulation system is lower than that of the traditional external-modulation OFDR system that combines a narrow-linewidth laser with a single-sideband modulator to achieve wavelength tuning. Cross-correlation analysis is used as a sensing mechanism to evaluate the similarity between Rayleigh scatter signals and to achieve vibration event localization. Experimental comparison is made of the vibration sensing performance of the external- and internal-modulation systems, and for a vibration event generated at a distance of 100.95 km, they locate it with a sensing spatial resolution of 43.0 m and 16.8 m, respectively. The results indicates that the proposed distributed vibration sensing based on internal modulation has better sensing performance and lower complexity compared to the traditional external-modulation system. In addition, the proposed system is single-ended and involves no optical amplification, which makes it very suitable for ultra-long-range sensing.

Simulation and prediction of vortex-induced vibration of a long suspension bridge using SHM-based digital twin technology

Although wind tunnel tests have already become a design basis for improving the stability of long-span bridges under the action of wind loads, vortex-induced vibration (VIV) events of in-service long-span bridges have still been frequently reported in recent years. The reasons behind this could be attributed to the limitations and uncertainties involved in wind tunnel tests as well as the changes and degradations of in-service bridges. In consideration that most long-span bridges have now been equipped with long-term structural health monitoring (SHM) systems and that digital twins could provide a new solution for simulating and predicting the reality, this study aims at developing a digital twin for VIV of a long suspension bridge to simulate and predict its VIV. The design document-based finite element (FE) model of the bridge is first updated using the measured dynamic characteristics. The numerical model for VIV of the bridge is then established using the vortex-induced force (VIF) model established based on wind tunnel test results. The numerical model together with the VIF model are finally updated against the measured vortex-induced responses (VIR) to produce a digital twin for VIV of the bridge. The relationships between the parameters in the VIF model and wind characteristics are further investigated to enable the digital twin to predict future VIV events. The accuracy of the digital twin for predicting the future VIV and the influence of the VIF parameters obtained from wind tunnel tests on the prediction accuracy both are confirmed through comparisons between the predicted and measured results.

Research on intelligent identification algorithm for preventing external damage behavior of power cable based on machine learning

Abstract As the key carrier of power distribution line power transmission and the supporting artery of urban functioning, power cables are widely used in electric power engineering construction. The number of users is increasing, and the distribution of locations is becoming more complex. However, if a power cable failure occurs, if not eliminated in time, it will threaten the safety of the power grid, seriously affecting the production of enterprises and the lives of residents. The article first on the power cable vibration signal recognition algorithm design, extraction of power cable external damage fault signal characteristics, proposed based on PSO-SVM power cable external damage fault waveform recognition algorithm, and the design of power cable anti-external damage intelligent early warning system, to achieve a certain period according to the trend of signal changes in the vibration event to discriminate. Research results show that under the same computer operating conditions, the average running time of the algorithm proposed in this paper is only 3.43. Compared to the other two algorithms, the algorithm has the highest recognition accuracy, the fastest convergence speed, and the optimal comprehensive performance. In the external damage signal recognition and warning test, the recognition rate of the excavator passing through defense zone one and two at a distance of 3 meters is 100%, which is greater than 80%, in line with the expected effect.

Intelligent Machinery Fault Diagnosis With Event-Based Camera

Event-based cameras are the emerging bio-inspired technology in vision sensing. Different from the traditional standard cameras, the event-based cameras asynchronously record the brightness change per pixel, and have the great merits of high temporal resolution, low energy consumption, high dynamic range, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">etc</i> . While the event-based cameras have been initially exploited in several common vision-based tasks in the recent years, the investigation on machine condition monitoring problem is quite limited. This paper offers the first attempt in the current literature on exploring the contactless event vision data for machine fault diagnosis. A vibration event representation is proposed to transform the event records into typical data samples, and a deep convolutional neural network model is used for processing the event information. To enhance the model robustness against environmental noisy vision events, an event data augmentation method is proposed to introduce variations of the event patterns. A deep representation clustering method is further proposed to improve the pattern recognition performance with respect to different machine health conditions. Experiments on the event vision-based rotating machine fault diagnosis problem are carried out. It is extensively validated that high fault diagnosis accuracies can be obtained using the vision data from the event-based cameras, which are competitive with the popular accelerometer data. Considering the properties of flexibility, portability and data recognizability, the event-based cameras thus provide a promising new tool for contactless machine health condition monitoring and fault diagnosis.

Design and Implementation of an Event-Driven Smart Sensor Node for Wireless Monitoring Systems.

In this paper, an event-driven wireless sensor node is proposed and demonstrated. The primary design objective is to devise a wireless sensor node with miniaturization, integration, and high-accuracy recognition ability. The proposed wireless sensor node integrates two vibration-threshold-triggered energy harvesters that sense and power a threshold voltage control circuit for power management, a microcontroller unit (MCU) for system control, a one-dimensional convolutional neural network (1D-CNN) environment data analysis and vibration events distribution, and a radio frequency (RF) digital baseband transmitter with IEEE 802.15.4-/.6 protocols. The dimensions of the wireless sensor node are 4 × 2 × 1 cm3. Finally, the proposed wireless sensor node was fabricated and tested. The alarming time for detecting the vibration event is less than 6 s. The measured recognition accuracy of three events (knock, shake, and heat) is over 97.5%. The experimental results showed that the proposed integrated wireless sensor node is very suitable for wireless environmental monitoring systems.

Traceability System of Sand Core in Casting Production with a Digital-Twin Core Rack

In recent development of casting traceability, it has been noticed that many casting manufacturing processes that involve the use of sand cores often lack proper traceability back to the cores used in their production and the supporting production information. Most marking symbols applied to sand cores using existing marking methods not only exhibit unstable code readability but also pose critical risks to the surface quality of sand cores. In response, we developed and tested a digital-twin core rack aimed at transferring digital codes to sand cores without contact with their surfaces. By utilizing infrared sensors for real-time object detection, the core rack can transfer digital identical codes from the core-making machine to the core rack, and subsequently to the casting machine. This digital code encompasses all relevant process data of a sand core, from core making to de-coring. Moreover, the various sensors embedded in the core rack were designed to monitor air humidity, temperature, and vibration events during the transport and storage of the cores. These parameters are especially critical for maintaining the quality of inorganic sand cores. The study demonstrates the successful application of each component within this conceptual framework. Looking ahead, it is imperative to update the hardware of this concept to ensure its adaptation to the demanding conditions of an industrial environment.

A spatial and temporal signal fusion based intelligent event recognition method for buried fiber distributed sensing system

Due to the characteristics of high sensitivity, fast response speed and multi-point monitoring, Phase-sensitive optical time-domain reflectometry (Φ-OTDR) has attracted attention in the field of perimeter security. However, the intrusion signals are susceptible to interference by ambient signals and difficult to be recognized. In the field of the vibration event recognition of Φ-OTDR system, the deep-learning based methods achieve great recognition ability. However, the previous works are mostly built on single signal source like temporal signal and may not completely adapt to different environment. In this work, a novel phenomenon is reported that a specific variation pattern of light intensity, which is related to the type of vibration source, is hided in the backscattering traces in spatial domain. Inspired by this, a lightweight model and data composition method is proposed to fuse spatial information with temporal correlation information based on end-to-end CNN-LSTM combined model and bicubic scaling. The experiment is conducted on a portable computer (a Nvidia GPU RTX 2080 with 2944 compute unified device architecture cores) with 8 event types and shows that this method can achieve 95.56% validation accuracy through less than 6 min training. Compared with previous method trained by single image structure signal, this lightweight work can achieve higher validation accuracy faster.

Vibration Event Recognition Using SST-Based Φ-OTDR System.

We propose a method based on Synchrosqueezing Transform (SST) for vibration event analysis and identification in Phase Sensitive Optical Time-Domain Reflectometry (Φ-OTDR) systems. SST has high time-frequency resolution and phase information, which can distinguish and enhance different vibration events. We use six tap events with different intensities and six other events as experimental data and test the effect of attenuation. We use Visual Geometry Group (VGG), Vision Transformer (ViT), and Residual Network (ResNet) as deep classifiers for the SST transformed data. The results show that our method outperforms the methods based on Continuous Wavelet Transform (CWT) and Short-Time Fourier Transform (STFT), while ResNet is the best classifier. Our method can achieve high recognition rate under different signal strengths, event types, and attenuation levels, which shows its value for Φ-OTDR system.

Comparison of Amplitude-Frequency Response Characteristics Between DCM and Arctan Algorithms in ϕ-OTDR

In phase-sensitive optical time-domain reflectometry (ϕ-OTDR) systems, DCM and Arctan demodulation algorithms are typically used to recover the information on vibration events along the fiber. The amplitude-frequency response characteristics of the DCM and Arctan algorithms are investigated and compared using a 3×3 coupler demodulation technique. We first testify the demodulation capability of the DCM and Arctan algorithms through experiments in ϕ-OTDR. Then the amplitude-frequency response ranges of the two algorithms are analyzed according to the demodulated phase error distribution from both the experiment and simulation. It is shown that the Arctan algorithm has a wider amplitude-frequency range than the DCM algorithm. The phase error generation mechanism is also investigated and we find that, in a noise-free condition, the phase errors still appear with use of the DCM algorithm, which cannot be explained by the general perspective that the phase errors arise from the sensitivity of the differential operator to noise. For both algorithms the phase error is related to the amplitude (D) and frequency (f) of the vibration signal, and pulse repetition frequency <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$(f_pulse)$</tex-math></inline-formula> . When the phase error tolerance is less than 0.1, the expression of the amplitude-frequency response characteristic of DCM algorithm is 5.7 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$(D+1)f\leq f_pulse$</tex-math></inline-formula> , whereas that of the Arctan algorithm is <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$2Df\leq f_pulse$</tex-math></inline-formula> .

Multi-Event Location Denoising Scheme for φ-OTDR Based on FFDNet Network

In order to improve the signal-to-noise ratio (SNR) of vibration sensing in the phase-sensitive optical time-domain reflectometer (φ-OTDR) system, a fiber sensing signal processing method based on the FFDNet convolutional neural network is proposed in this paper. In the network, the concept of residual learning is introduced, which involves constructing a residual mapping and utilizing multi-layer convolutional neural networks to learn the noise distribution present in the original image. The denoised result can be obtained by subtracting the learned noise from the original image. We have built a φ-OTDR system based on coherent detection, using three PZTs as simulated vibration sources and a series of experiments at 200 Hz, with each experiment simulating a single vibration event or multiple vibration events by setting different intensities. The experimental results demonstrate that the FFDNet based fiber optic sensing signal processing method enhances the SNR to 37.84 dB, 37.11 dB, and 37.31 dB, respectively, while preserving vibration signal details more effectively than wavelet denoising and Gaussian filtering techniques. The trained FFDNet model has great potential for improving the performance of the φ-OTDR system and has some practical application value.