Vibration Monitoring Data Research Articles

A feature reconstruction and SAE model based diagnosis method for multiple mixed faults

Owing to the merits of automatic feature extraction and depth structure, intelligent fault diagnosis based on deep neural networks has become a great concern. However, the non-fault state monitoring data volume of actual industrial machinery is rich, whereas the fault state data volume is insufficient and weak. Furthermore, achieving multiple mixed-fault diagnoses using skewed data distributions is extremely difficult. A feature reconstruction and sparse auto-encoder (AE) model-based diagnosis method for multiple mixed faults is proposed in this study to bridge these gaps. Such a feature reconstruction algorithm is designed and employed to address the following issues: (1) expensive computing resulting from the long sequential features of vibration monitoring data and (2) the extraction problem caused by the submersion of scarce data features. Furthermore, an adaptive loss function was formulated, and a deep AE network was constructed to identify the health status and determine the fault level. Diagnoses of artificial and real faults verify the availability and superiority of the proposed scheme, demonstrating the adaptability and robustness of these hyperparameters.

Adaptive Vibration Monitoring of Railway Track Structures Using the UWFBG by the Identification of Train-Load Patterns

Due to the capability of multiplexing thousands of sensors on a single optical cable, ultra-weak fiber Bragg grating (UWFBG) vibration sensing technology has been utilized in monitoring the vibration response of large-scale infrastructures, particularly urban railway tracks, and the volume of the collected monitoring data can be huge with the great number of sensors. Even though the train-induced vibration responses of urban railway tracks constitute the most informative and crucial component, they comprised less than 7% of the total operational period. This is mainly attributed to the temporal sparsity of commuting trains. Consequently, the majority of the stored data consisted of low-informative environmental noise and interference excitation data, leading to an inefficient structural health monitoring (SHM) system. To address this issue, this paper introduced an adaptive monitoring strategy for railway track structures, which is capable of identifying train-load patterns by leveraging deep learning techniques. Inspired by image semantic segmentation, a U-net model with one-dimensional convolution layers (U-net-1D) was developed for the pointwise classification of vibration monitoring data. The proposed model was trained and validated using a dataset obtained from an actual urban railway track in China. Results indicated that the proposed method outperforms the traditional dual-threshold method, achieving an Intersection over Union (IoU) of 94.27% on the segmentation task of the test dataset.

Model updating of a shear‐wall tall building using various vibration monitoring data: Accuracy and robustness

SummaryAcceleration measurements are often used for model updating of civil engineering structures, especially in the case of seismic monitoring. It is yet unclear if accelerations alone would generate an accurate and robust finite‐element (FE) model. This study examines this notion and analyzes the possibility of using other vibration monitoring data for model updating of shear‐wall tall buildings. This study compares the accuracy and robustness of the FE models being optimized via accelerations, roof displacement, wall rotations, interstory drift ratios, and the linear combination of these measurements. A numerical case study is analyzed using Timoshenko beams for modeling the lateral vibration of a benchmark 42‐story building under seismic excitations. Results show that the acceleration response of the examined building is mostly governed by its higher vibration modes. Depending on the characteristics of ground motions, using accelerations alone may generate an FE model biased towards higher‐order modes without effectively capturing the lower‐order modes. For instance, the first modal frequency of the updated FE model could be 12.0% lower than the true value, and the reconstructed displacement and rotation responses are noticeably inaccurate. Employing multi‐source monitoring data for model updating, for example, the combinations of roof displacement and acceleration measurements, could reduce the normalized root‐mean‐square errors in displacements by more than 70%. This study also quantifies the robustness of the FE model under various measurement noise levels and 50 pairs of earthquake records. Finally, the effects of multi‐source data on FE model updating are validated via experiments on a 7‐story shear wall building. Analysis reveals that a more accurate and robust FE model can be determined via a combination of accelerations and top displacement than via acceleration alone.

Interpretable multi-domain meta-transfer learning for few-shot fault diagnosis of rolling bearing under variable working conditions

To address the challenges of accurately diagnosing few-shot fault samples obtained from rolling bearings under variable operating conditions, as well as the issues of black box nature and delayed feedback to guide fault handling in intelligent diagnostic models, this paper proposes an interpretable multi-domain meta-transfer learning method. Firstly, vibration monitoring data of rolling bearings under different operating conditions are collected, and time–frequency domain features are extracted to construct multi-channel one-dimensional temporal samples as inputs. A multi-domain meta-transfer learning framework based on deep convolutional neural networks is then built to perform few-shot learning with multiple tasks under different operating conditions. The output results are reverse-reconstructed through a fusion hierarchical class activation mapping, and the feature maps are assigned different weights to obtain saliency maps corresponding to the inputs, thus improving the interpretability of the output results. Finally, the dataset of bearing vibration data under time-varying rotational speed conditions is used to validate the effectiveness of the proposed method. Experimental results show that the proposed method can achieve accurate fault diagnosis results under variable operating conditions with few-shot samples, and the diagnosis results can be fed back to the input for decision-making, enhancing the interpretability of the model. Compared with other models, it also demonstrates better robustness and accuracy.

Modeling of construction vibration impacts in transit project applications

This presentation explores some of the challenges related to predicting the vibration impact from construction activities associated with new mass transit infrastructure. Pre-construction assessment methods are examined and compared against as-measured construction vibration monitoring data. Practical considerations and limitations of construction vibration monitoring are also discussed, citing examples, and insight from the ongoing projects, especially in the Greater Toronto Area.

An Embedded Machine Learning Fault Detection System for Electric Fan Drive

Industrial fans are critical components in industrial production, where unexpected damage of important fans can cause serious disruptions and economic costs. One trending market segment in this area is where companies are trying to add value to their products to detect faults and prevent breakdowns, hence saving repair costs before the main product is damaged. This research developed a methodology for early fault detection in a fan system utilizing machine learning techniques to monitor the operational states of the fan. The proposed system monitors the vibration of the fan using an accelerometer and utilizes a machine learning model to assess anomalies. Several of the most widely used algorithms for fault detection were evaluated and their results benchmarked for the vibration monitoring data. It was found that a simple Convolutional Neural Network (CNN) model demonstrated notable accuracy without the need for feature extraction, unlike conventional machine learning (ML)-based models. Additionally, the CNN model achieved optimal accuracy within 30 epochs, demonstrating its efficiency. Evaluating the CNN model performance on a validation dataset, the hyperparameters were updated until the optimal result was achieved. The trained model was then deployed on an embedded system to make real-time predictions. The deployed model demonstrated accuracy rates of 99.8%, 99.9% and 100.0% for Fan-Fault state, Fan-Off state, and Fan-On state, respectively, on the validation data set. Real-time testing further confirmed high accuracy scores ranging from 90% to 100% across all operational states. Challenges addressed in this research include algorithm selection, real-time deployment onto an embedded system, hyperparameter tuning, sensor integration, energy efficiency implementation and practical application considerations. The presented methodology showcases a promising approach for efficient and accurate fan fault detection with implications for broader applications in industrial and smart sensing applications.

Underwater Radiated Noise Prediction Method of Cabin Structures under Hybrid Excitation

Aiming at the engineering limitations of traditional ship vibration online monitoring and noise prediction methods, this paper proposes a method for online monitoring and underwater radiation noise prediction of cabin structures’ vibration and noise under hybrid excitation of sound and force. The method first constructs the condition test model; based on OTPA technology, the “acoustic-vibration” transfer function between sound and vibration monitoring points in the cabin and the “acoustic/vibration-acoustic” transmission network inside and outside the cabin structure are obtained. Secondly, based on the “acoustic-vibration” transfer function, the online vibration and sound monitoring data are decoupled and processed to obtain the modified vibration and sound monitoring data. Finally, the near-field radiation noise on the conformal hologram surface outside the cabin is predicted based on the “acoustic/vibration-acoustic” transmission network, and the far-field radiation noise of the cabin structure is predicted by the wave superposition method. In this paper, the contribution law of external radiation noise and the coupling characteristics of the monitoring information under the hybrid excitation of sound and force are analyzed theoretically, and the decoupling method of coupling information is also studied. This method makes up for the problem of missing underwater radiation noise caused by sound excitation in traditional vibration monitoring, and it can effectively improve the prediction accuracy of underwater radiation noise. The effectiveness of the method is further verified by the tank model experiments.

Monitoring and analysis of railway-induced vibration and structure-borne noise in a transit oriented development project

With the rapid development of modern cities, the construction of over-track buildings and buildings along the subway line are increasing gradually. In order to evaluate the influence of rail transit on these buildings, the railway-induced vibration and structure-borne noise of subway stations and buildings in a Transit Oriented Development (TOD) project are monitored and analyzed in this paper. The vibration monitoring data in the subway station shows that the vibration acceleration response of the measuring point in the station caused by the train entering and leaving the station is the largest in the vertical direction, followed by the along-track direction, and the vertical-track direction is the smallest. The corresponding predominant frequency of vibration is usually in the range of 100 Hz-200 Hz. The vibration monitoring data of the building foundation along the subway line shows that the railway-induced vibration of the building foundation along the track near the subway station decreases with the reduction of the distance from the subway station. The monitoring vibration results of the over-track building on the metro depot are different from the buildings near the main line. The predominant frequency of the vibration at the measuring point is in the range of 30 Hz-40 Hz. The small vertical vibration acceleration response of the measuring point corresponds to a high vibration level. In addition, the amplitude of railway-induced vibration in some building rooms near the metro depot is basically consistent with the amplitude of background vibration, but the monitoring value of the structure-borne noise is obviously higher than the background noise. The research results of this paper show the influence of railway-induced vibration on the over-track buildings and the buildings along the subway line, which has important reference value for the comprehensive development of rail transit.

Stagnation point-induced vibration on ultra-long stay cables and the vibration control by using a novel stockbridge damper

A peculiar wind-induced vibration with large acceleration amplitude, super-high and narrow frequency bands was observed on ultra-long stay cables after one-year-more field monitoring on the Sutong Bridge. The vibration was found extremely detrimental and cannot be suppressed by a single damper system. Based on this, it is first demonstrated that the vibration is a unique vortex-induced vibration resulting from the damper installation position, and which is named stagnation point-induced vibration (SPIV). Subsequently, a Stockbridge damper (SBD) scheme, usually used on transmission lines, is introduced to suppress the SPIV: the dynamic model of the stay cable-SBD system is established; the additional damping ratios are calculated based on the proposed model and the equations; moreover, parameter optimisation analyses of the SBD are conducted. Then, a real SBD is designed, produced, and finally installed on an actual stay cable with a length of 551 m based on the optimum parameters. Finally, with six months of field vibration monitoring data, the proposed dynamic model and the SBD damping scheme are demonstrated effective to the suppression of the detected SPIV. The study is expected to give a practical vibration control reference for ultra-long stay cables that are more than 500 m.

Structural Vibration Data Anomaly Detection Based on Multiple Feature Information Using CNN-LSTM Model

Structural health monitoring (SHM) system has been operating for a long time in a harsh environment, resulting in various abnormalities in the collected structural vibration monitoring data. Detecting these abnormal data not only requires user interaction but also is quite time-consuming. Inspired by the manual recognition process, a vibration data anomaly detection method based on the combined model of convolutional neural network (CNN) and long short-term memory (LSTM) network is proposed in this paper. This method simulates intelligent human decision making in two steps. First, the original data are reconstructed by two feature sequences with higher universality and smaller size. In the time domain, the residual signal is extracted from the upper and lower peak envelopes of the original data to characterize the symmetry of the data. In the frequency domain, the power spectral density sequence of the original data is extracted to characterize the interpretability of the data. Second, a CNN-LSTM model is constructed and trained which utilizes CNN to extract local high-level features of input sequence and inputs new continuous high-level feature representations into LSTM to learn global long-term dependencies of abnormal data features. For verification, the method was applied to the automatic classification of continuous monitoring data for 42 days of long-span bridge, and the average accuracy of the classification results exceeded 94% and the detection time was 78 minutes. Compared with existing methods, this method can detect abnormal data more accurately and efficiently and has a stronger generalization ability.

Deep learning-based adaptive mode decomposition and instantaneous frequency estimation for vibration signal

Under time-varying environments and loads, vibration monitoring data of structures exhibit nonstationarity with multiple frequency components, and may have obvious nonlinearity under some extreme loads. The time–frequency analysis method is ideal for analyzing these nonlinear and nonstationary signals. However, the vibration signals of large-scale structures typically contain a variety of frequency components, making it challenging for effective mode decomposition without the interference of false mono-components by existing adaptive mode decomposition methods. False mono-components will severely decrease the accuracy of instantaneous frequency. To address the mode mixing and false mono-components in mode decomposition, a data-driven approach based on deep learning for signal adaptive mode decomposition and instantaneous frequency estimation is proposed in this paper. By transforming the time–frequency analysis problem into an optimization task of a deep neural network, built-in redundant adaptive filters with sparse regularization are designed, then the accurate intrinsic mode functions and time-varying frequencies can be estimated. The proposed method is verified by the numerical nonstationary signal and vibration signal of an experimental long cable model. The results demonstrate the superior capability of the proposed method in decomposing a signal into mono-components adaptively without false mono-components and achieving high accuracy of corresponding instantaneous frequencies. From the examples, an accuracy of 98.52% is achieved for a noise-free signal with spectral overlap. The proposed method is more robust to noise, and the instantaneous frequency identification error is 4.90%, which is much lower than EEMD and VMD under a signal-to-noise ratio of 5 dB.

Semi-supervised deep learning for recognizing construction activity types from vibration monitoring data

Monitoring construction-induced vibrations is crucial for mitigating their adverse impacts on surroundings; hence, contractors have collected a large amount of construction-induced vibration data. Unfortunately, corresponding construction activities were usually not recorded during vibration data acquisition. Consequently, such an unclassified historical vibration database cannot be efficiently utilized to establish empirical models for vibration prediction. Therefore, automatic recognition of the corresponding construction activity types from vibration monitoring data has great potential in practice and research. Previous relevant approaches rely primarily on either manual identification or supervised deep learning algorithms. However, these methods are tedious and require a large amount of labeled data that is usually unavailable in actual scenarios. An advanced approach for the recognition of construction activity types based on semi-supervised deep learning is proposed in this paper to solve this problem. The proposed method comprises a one-dimensional (1D) convolutional neural network (CNN) and a semi-supervised Ladder network. The Ladder-CNN was trained and tested on raw acceleration data measured on real construction sites generated by four different piling operations and one excavation operation. Experiments showed that the Ladder-CNN achieved accuracy up to 98.4% with 10% labeled data and 93.8% with only 1% labeled data. The effects of varying amounts of unlabeled data, different noise levels, and different layer weights were also investigated. Another semi-supervised algorithm (i.e., Pseudo-label) and a supervised 1D CNN model were also tested and compared. The experimental study showed that the Ladder-CNN outperforms the Pseudo-label and is comparable to supervised 1D CNN, demonstrating excellent competitiveness and apparent advantages in practical applications.

Application of noise-cancelling and smoothing techniques in road pavement vibration monitoring data

Road pavement surfaces need routine and regular monitoring and inspection to keep the surface layers in high-quality condition. However, the population growth and the increases in the number of vehicles and the length of road networks worldwide have required researchers to identify appropriate and accurate road pavement monitoring techniques. The vibration-based technique is one of the effective techniques used to measure the condition of pavement degradation and the level of pavement roughness. The consistency of pavement vibration data is directly proportional to the intensity of surface roughness. Intense fluctuations in vibration signals indicate possible defects at certain points of road pavement. However, vibration signals typically need a series of pre-processing techniques such as filtering, smoothing, segmentation, and labelling before being used in advanced processing and analyses. This research reports the use of noise-cancelling and data-smoothing techniques, including high pass filter, moving average method, median, Savitzky-Golay filter, and extracting peak envelope method, to enhance raw vibration signals for further processing and classification. The results show significant variations in the impact of noise-cancelling and data-smoothing techniques on raw pavement vibration signals. According to the results, the high pass filter is a more accurate noise-cancelling and data smoothing technique on road pavement vibration data compared to other data filtering and data smoothing methods.

Response of acoustic emission and vibration monitoring data during rock block collapse in the tunnel: Small- and large-scale experiments study

Block collapse is a common disaster in the jointed rock mass of tunnels and usually involves two typical evolution stages: rock bridge breakage and rock block instability. The evolutionary characteristics of mutation induced by the gradual failure of rock mass are the mechanical basis for monitoring and assessing disasters. In this study, an acoustic emission (AE)-vibration joint monitoring method is proposed to fully investigate the two failure stages based on the disaster mechanism. It is critical to reveal the evolution law of AE and natural frequency information in block collapse. Small-scale shear tests on rock specimens are conducted to investigate the rock bridge breakage stage, while large-scale physical simulation tests are performed to investigate the block instability stage. During rock bridge breakage, AE exhibits a clearer signal response to the internal rock fracture than the natural frequency and captures brittle failure under low- and medium-stress conditions. The natural frequency exhibits a significant signal response under low-stress conditions during the block instability stage. Under medium- and high-stress conditions, the signal becomes less sensitive. Under low-stress conditions, the natural frequency of the block gradually decreases with increasing free surface, implying that the stability of the block declines. The natural frequency decreases with the decreasing stiffness of the block, which is a vital precursor to the early warning of block collapse disasters. Additionally, AE is the primary indicator in the rock bridge breakage stage, while natural frequency is the primary indicator in the rock block instability stage. From the perspective of technological innovation in underground tunneling, AE-Vibration joint monitoring prompts a new idea to evaluate the stability of key blocks in tunnel construction safety applications.

Damage identification of seismic-isolated structure based on CAE network using vibration monitoring data

Seismic-isolated systems have been investigated and practiced globally to prevent the destruction of existing structures, and vibration-based structural health monitoring is important for maintaining the functionality of the isolation systems. Given the recent boom in computational science and machine learning algorithms, this paper deployed a structural health monitoring system for a rubber bearing-isolated gymnasium in an area with high seismic fortification intensity and introduced an unsupervised deep learning network named convolutional autoencoder (CAE) to identify damage from vibrations of the isolation layer. The CAE network is first trained by operational monitoring vibrations and tested by new data. The vibration assessment results indicate that the CAE network can accurately reconstruct daily data and effectively detect unexpected ground motion. Thereafter, to further test the network's damage identification and localization performance, an analytical finite element model of the gymnasium structure is established, and the undamaged and damaged datasets are generated for network verification and updating. The verification results demonstrate that the CAE network leads to reliable feature extraction power in global and local damage detection of the isolation layer. The proposed vibration assessment method is beneficial to building administrators in making accurate and timely decisions with the assistance of the score result provided by the CAE network.

A novel efficient probabilistic prediction approach for train-induced ground vibrations based on transfer learning

To deal with the issues of high computational cost and prediction uncertainty of numerical models in train-induced ground-borne vibration prediction, a prediction method based on transfer learning is proposed in this study. In this method, the vehicle–track-coupled analytical model and three-dimensional finite element model are first used to calculate the train-induced ground vibration under various condition variables, and these data were used as training samples to pre-train the deep neural network models. Numerous train-induced ground vibration experiments were then conducted along the metro lines in Beijing, and those measured vibration data were used to fine-tune the pre-trained deep neural network model with the transfer learning strategy. A random variable obeying a Gaussian distribution is assumed over the predicted vibration acceleration levels to model the randomness of train-induced vibration, and the parameters of this distribution were determined by the statistical results of vibration monitoring data in the metro tunnels. The fully trained model could complete the prediction of train-induced ground vibration in seconds. Finally, a case study was carried out, by comparing the probabilistic prediction results with the statistical results of the field measurements, and the feasibility and the improvement of the proposed method were demonstrated.

Anomaly detection for construction vibration signals using unsupervised deep learning and cloud computing

In-operation construction vibration monitoring records inevitably contain various anomalies caused by sensor faults, system errors, or environmental influence. An accurate and efficient anomaly detection technique is essential for vibration impact assessment. Identifying anomalies using visualization tools is computationally expensive, time-consuming, and labor-intensive. In this study, an unsupervised approach for detecting anomalies in construction vibration monitoring data was proposed based on a temporal convolutional network and autoencoder. The anomalies were autonomously detected on the basis of the reconstruction errors between the original and reconstructed signals. Considering the false and missed detections caused by great variability in vibration signals, an adaptive threshold method was applied to achieve the best identification performance. This method used the log-likelihood of the reconstruction errors to search for an optimal coefficient for anomalies. A distributed training strategy was implemented on a cloud platform to speed up training and perform anomaly detection without significant time delay. Construction-induced accelerations measured by a real vibration monitoring system were used to evaluate the proposed method. Experimental results show that the proposed approach can successfully detect anomalies with high accuracy; and the distributed training can remarkably save training time, thereby realizing anomaly detection for online monitoring systems with accumulated massive data.

Construction Activity Classification Based on Vibration Monitoring Data: A Supervised Deep-Learning Approach with Time Series RandAugment

Construction Activity Classification Based on Vibration Monitoring Data: A Supervised Deep-Learning Approach with Time Series RandAugment

A combined deep learning model for damage size estimation of rolling bearing

The identification of damage sizes based on vibration monitoring data is significant in the study of fault evolution, prediction and diagnosis of rolling bearing. There are inherent restrictions in traditional identification models such as high dependence on prior knowledge and insufficient feature extraction, aiming at the limitation, a prediction method of rolling bearing damage size based on deep learning is proposed. In this paper, a combined model of deep convolutional long-short-term memory network is developed, which can sufficiently extract the multi-dimensional and time-series characteristics of bearing vibration signal, and realize the intelligent and efficient diagnosis of bearing fault. On the basis of theoretical analysis, the rolling bearing fault tests under various damage sizes and rotational velocities are carried out by using the accelerated fatigue testing machine for rolling bearings, and the traditional and novel methods are compared based on the test data. The results show that the prediction accuracy of the combined network can reach 99.94% and 98.67%, respectively under normal and noisy conditions, which is higher than the single deep convolution network, long-short-term memory network and other models. The comparison results amply demonstrate the superiority of the proposed method.

RANCANG BANGUN DATA LOGGER MONITORING VIBRASI PADA MOTOR LISTRIK 6,3 KV BERBASIS IOT SECARA REAL TIME DI PLTU JERANJANG

As a power plant industry, PLTU Jeranjang OMU (Operation and Maintenance Unit) owned oneof 6.3 kV nominal voltage induction motor that has critical characteristic that drives the waterpump into the boiler room which always has to operate. This motor has been equipped withtemperature sensor in real time that can monitor the temperature of bearing motor. However,manual control is still applied in vibration monitoring. It is done by taking the vibration data oncea week, and there is no deviation data found in real time. The result of this study shows that theprototype of IoT-based vibration monitoring data logger that is designed using the blynkapplication obtained the average of sensor deviation of 2%. Alarm notification of each sensorsdeliver "warning" if the vibration value indicates ? 4,5-70 mm/s and alarm notification delivers"emergency" if the vibration value indicates ? 7,1 mm/s in concordance with the ISO 10816-3vibration standard. Although, the performance of sensor data storage has successfully storingthe data with 8 GB microSD card, then will be fully charged in 4629 days or 154 months and 3days.