• Structural Monitoring and Maintenance
• /
• v.7 no.2
• /
• pp.109-124
• /
• 2020

Benefiting from the massive monitoring data collected by the Structural health monitoring (SHM) system, scholars can grasp the complex environmental effects and structural state during structure operation. However, the monitoring data is often missing due to sensor faults and other reasons. It is necessary to study the recovery method of missing monitoring data. Taking the structural temperature monitoring data of Nanjing Dashengguan Yangtze River Bridge as an example, the long short-term memory (LSTM) network-based recovery method for missing structural temperature data is proposed in this paper. Firstly, the prediction results of temperature data using LSTM network, support vector machine (SVM), and wavelet neural network (WNN) are compared to verify the accuracy advantage of LSTM network in predicting time series data (such as structural temperature). Secondly, the application of LSTM network in the recovery of missing structural temperature data is discussed in detail. The results show that: the LSTM network can effectively recover the missing structural temperature data; incorporating more intact sensor data as input will further improve the recovery effect of missing data; selecting the sensor data which has a higher correlation coefficient with the data we want to recover as the input can achieve higher accuracy.

• Journal of Biomedical Engineering Research
• /
• v.22 no.4
• /
• pp.349-358
• /
• 2001

A novel electro-mechanical implantable ventricular assist system is developed as a bridge to transplantation or recovery for patients with end-stage heart failure. The developed system is composed of an implanted blood pump, an external monitoring system which stores data, and a wearable system including a portable external driver and a portable power supply system. The blood pump is designed to be implanted into the left upper abdominal space and provides blood flow from the left ventricular apex to the aorta. The pulsatile blood flow is generated by a double cylindrical cam. There was mo excessive heat emission from the blood pump into the temperature-controlled chamber in the heat test and no stagnated flow within the blood sac by the observation in the flow visualization test. Animal experiments were performed using sheep and calves. The maximum assist flow rate reached 7.85L/min in the animal experiment. The evaluation results showed that the developed system was feasible for the implantable ventricular assist system. The long-term in vitro durability test and mid-term in vivo experiments are in progress and mow the modified next model is under development.

• Smart Structures and Systems
• /
• v.25 no.1
• /
• pp.23-35
• /
• 2020

It is important to take into account the thermal behavior in assessing the structural condition of bridges. An effective method of studying the temperature effect of long-span bridges is numerical simulation based on the solar radiation models. This study aims to develop a time-varying solar radiation model which can consider the real-time weather changes, such as a cloud cover. A statistical analysis of the long-term monitoring data is first performed, especially on the temperature data between the south and north anchors of the bridge, to confirm that temperature difference can be used to describe real-time weather changes. Second, a defect in the traditional solar radiation model is detected in the temperature field simulation, whereby the value of the turbidity coefficient t_u is subjective and cannot be used to describe the weather changes in real-time. Therefore, a new solar radiation model with modified turbidity coefficient γ is first established on the temperature difference between the south and north anchors. Third, the temperature data of several days are selected for model validation, with the results showing that the simulated temperature distribution is in good agreement with the measured temperature, while the calculated results by the traditional model had minor errors because the turbidity coefficient t_u is uncertainty. In addition, the vertical and transverse temperature gradient of a typical cross-section and the temperature distribution of the tower are also studied.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.77-91
• /
• 2022

Various monitoring systems have been implemented in civil infrastructure to ensure structural safety and integrity. In long-term monitoring, these systems generate a large amount of data, where anomalies are not unusual and can pose unique challenges for structural health monitoring applications, such as system identification and damage detection. Therefore, developing efficient techniques is quite essential to recognize the anomalies in monitoring data. In this study, several machine learning techniques are explored and implemented to detect and classify various types of data anomalies. A field dataset, which consists of one month long acceleration data obtained from a long-span cable-stayed bridge in China, is employed to examine the machine learning techniques for automated data anomaly detection. These techniques include the statistic-based pattern recognition network, spectrogram-based convolutional neural network, image-based time history convolutional neural network, image-based time-frequency hybrid convolution neural network (GoogLeNet), and proposed ensemble neural network model. The ensemble model deliberately combines different machine learning models to enhance anomaly classification performance. The results show that all these techniques can successfully detect and classify six types of data anomalies (i.e., missing, minor, outlier, square, trend, drift). Moreover, both image-based time history convolutional neural network and GoogLeNet are further investigated for the capability of autonomous online anomaly classification and found to effectively classify anomalies with decent performance. As seen in comparison with accuracy, the proposed ensemble neural network model outperforms the other three machine learning techniques. This study also evaluates the proposed ensemble neural network model to a blind test dataset. As found in the results, this ensemble model is effective for data anomaly detection and applicable for the signal characteristics changing over time.

• Structural Engineering and Mechanics
• /
• v.47 no.5
• /
• pp.623-641
• /
• 2013

The fiber-reinforced polymer (FRP) composite panel, with the benefits of light weight, high strength, good corrosion resistance, and long-term durability, has been considered as one of the prosperous alternatives for structural retrofits and replacements. Although with these advantages, a further application of FRPs in bridge engineering may be restricted, and that is partly due to some unsatisfied thermal performance observed in recent studies. In this regard, Kansas Department of Transportation (DOT) conducted a field monitoring program on a bridge with glass FRP (GFRP) honeycomb hollow section sandwich panels. The temperatures of the panel surfaces and ambient air were measured from December 2002 to July 2004. In this paper, the temperature distributing behaviors of the panels are firstly demonstrated and discussed based on the field measurements. Then, a numerical modeling procedure of temperature fields is developed and verified. This model is capable of predicting the temperature distributions with the local environmental conditions and material's thermal properties. Finally, a parametric study is employed to examine the sensitivities of several temperature influencing factors, including the hollow section configurations, environmental conditions, and material properties.

• Structural Engineering and Mechanics
• /
• v.56 no.6
• /
• pp.939-957
• /
• 2015

In this paper, a new approach based on the continuum model is proposed to estimate the main cable tension force of suspension bridges from measured natural frequencies. This approach considered the vertical vibration of a main cable hinged at both towers and supported by an elastic girder and hangers along its entire length. The equation reflected the relationship between vibration frequency and horizontal tension force of a main cable was derived. To avoid to generate the additional cable tension force by sag-extensibility, the analytical solution of characteristic equation for anti-symmetrical vibration mode of the main cable was calculated. Then, the estimation of main cable tension force was carried out by anti-symmetric characteristic frequency vector. The errors of estimation due to characteristic frequency deviations were investigated through numerical analysis of the main cable of Taizhou Bridge. A field experiment was conducted to verify the proposed approach. Through measuring and analyzing the responses of a main cable of Taizhou Bridge under ambient excitation, the horizontal tension force of the main cable was identified from the first three odd frequencies. It is shown that the estimated results agree well with the designed values. The proposed approach can be used to conduct the long-term health monitoring of suspension bridges.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.3A
• /
• pp.209-216
• /
• 2009

The structural stability as well as economical efficiency of the wind sensitive structures are strongly dependant on accurate evaluation of the design wind speed. Present study demonstrates a useful wind data obtained at the wind monitoring tower in the Kwangyang Suspension Bridge site. Moreover the Measure-Correlate-Predict (MCP) method has been applied to estimate the long-term wind data at the bridge site based on the wind data at the local weather station. The measured data indicate that the turbulent intensities and roughness exponents are strongly affected by the wind direction and surrounding topography. The new design wind speed based on MCP method is 20m/s lower than that at the original estimation, and the resulting design wind load is only 36% of the old prediction. The field measurement of wind data is recommended to ensure the economical and secure design of the wind sensitive structures because the measured wind data reveal much different from the estimated one due to local topography.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.28 no.3
• /
• pp.66-73
• /
• 2024

Stay-cable is one of the most important load carrying members in cable-stayed bridges. Monitoring structural integrity of stay-cables is crucial for evaluating the structural condition of the cable-stayed bridge. For stay-cables, tension and damping ratio are estimated based on modal properties as a measure of structural integrity. Since the monitoring system continuously measures the vibration for the long-term period, data acquisition systems should be stable and power-efficiency as the hardware system. In addition, massive signals from the data acquisition systems are continuously generated, so that automated analysis system should be indispensable. In order to fulfill these purpose simultaneously, this study presents an autonomous cable monitoring system based on domain-knowledge using IoT for continuous cable monitoring systems of cable-stayed bridges. An IoT system was developed to provide effective and power-efficient data acquisition and on-board processing capability for Edge-computing. Automated peak-picking algorithm using domain knowledge was embedded to the IoT system in order to analyze massive data from continuous monitoring automatically and reliably. To evaluate its operational performance in real fields, the developed autonomous monitoring system has been installed on a cable-stayed bridge in Korea. The operational performance are confirmed and validated by comparing with the existing system in terms of data transmission rates, accuracy and efficiency of tension estimation.

• Smart Structures and Systems
• /
• v.24 no.6
• /
• pp.733-744
• /
• 2019

Wind speed forecasting is critical for a variety of engineering tasks, such as wind energy harvesting, scheduling of a wind power system, and dynamic control of structures (e.g., wind turbine, bridge, and building). Wind speed, which has characteristics of random, nonlinear and uncertainty, is difficult to forecast. Nowadays, machine learning approaches (generalized regression neural network (GRNN), back propagation neural network (BPNN), and extreme learning machine (ELM)) are widely used for wind speed forecasting. In this study, two schemes are proposed to improve the forecasting performance of machine learning approaches. One is that optimization algorithms, i.e., cross validation (CV), genetic algorithm (GA), and particle swarm optimization (PSO), are used to automatically find the optimal model parameters. The other is that the combination of different machine learning methods is proposed by finite mixture (FM) method. Specifically, CV-GRNN, GA-BPNN, PSO-ELM belong to optimization algorithm-assisted machine learning approaches, and FM is a hybrid machine learning approach consisting of GRNN, BPNN, and ELM. The effectiveness of these machine learning methods in wind speed forecasting are fully investigated by one-year field monitoring data, and their performance is comprehensively compared.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.181-193
• /
• 2022

Data-driven structural health monitoring (SHM) of civil infrastructure can be used to continuously assess the state of a structure, allowing preemptive safety measures to be carried out. Long-term monitoring of large-scale civil infrastructure often involves data-collection using a network of numerous sensors of various types. Malfunctioning sensors in the network are common, which can disrupt the condition assessment and even lead to false-negative indications of damage. The overwhelming size of the data collected renders manual approaches to ensure data quality intractable. The task of detecting and classifying an anomaly in the raw data is non-trivial. We propose an approach to automate this task, improving upon the previously developed technique of image-based pre-processing on one-dimensional (1D) data by enriching the features of the neural network input data with multiple channels. In particular, feature engineering is employed to convert the measured time histories into a 3-channel image comprised of (i) the time history, (ii) the spectrogram, and (iii) the probability density function representation of the signal. To demonstrate this approach, a CNN model is designed and trained on a dataset consisting of acceleration records of sensors installed on a long-span bridge, with the goal of fault detection and classification. The effect of imbalance in anomaly patterns observed is studied to better account for unseen test cases. The proposed framework achieves high overall accuracy and recall even when tested on an unseen dataset that is much larger than the samples used for training, offering a viable solution for implementation on full-scale structures where limited labeled-training data is available.