Shinae Jang

Corrosion Estimation of a Historic Truss Bridge Using Model Updating

Bridge structures are valuable national assets for transportation and economy that should be maintained properly for continuous stable operation. Corrosion is common in steel bridges; severe corrosion may result in significant economic impact and long downtime for retrofit. To date, various corrosion evaluation technologies have been developed such as nondestructive evaluation techniques and analytical model updating strategies. Among them, nondestructive evaluation is costly and time consuming for scanning entire bridges. For model updating, few examples on full-scale bridges with severe corrosion have been reported. In this paper, the corrosion level in a historic steel-truss bridge is estimated using model updating. Dynamic characteristics of the bridge are identified via a series of short-term full-scale experiments. An initial finite-element model of this bridge is then updated to match the field corrosion estimation results. The corrosion levels predicted by the proposed approach were consistent with the results of a visual inspection of this bridge. The results of the model updating routine could be used to monitor the overall corrosion levels in the structure with periodic inspection over time.

Feasibility study of wind generator for smart wireless sensor node in cable-stayed bridge

Long-term structural health monitoring (SHM) systems using wireless smart sensors for civil infrastructures such as cable-stayed bridges has been researched due to its cost-effectiveness and ease of installation. Wireless smart sensors are usually powered by high capacity batteries because they consume low power. However, theses batteries require regular replacements for long-term continuous and stable operation. To overcome this limitation of wireless smart sensor-based SHM, considerable attention has been recently paid to alternative power sources such as solar power and vibration-based energy harvesting. Another promising alternative ambient energy source might be a wind-generated power; in particular, it can be very useful for structures in windy area such as coastal and mountainous area. In this study, the feasibility of the wind-powered generation for wireless smart senor nodes is investigated by through experimental and analytical approaches, and the possibility of practical application to actual SHM system of a cable-stayed bridge is discussed.

Structural health monitoring system of a cable-stayed bridge using a dense array of scalable smart sensor network

This paper presents a structural health monitoring (SHM) system using a dense array of scalable smart wireless sensor network on a cable-stayed bridge (Jindo Bridge) in Korea. The hardware and software for the SHM system and its components are developed for low-cost, efficient, and autonomous monitoring of the bridge. 70 sensors and two base station computers have been deployed to monitor the bridge using an autonomous SHM application with consideration of harsh outdoor surroundings. The performance of the system has been evaluated in terms of hardware durability, software reliability, and power consumption. 3-D modal properties were extracted from the measured 3-axis vibration data using output-only modal identification methods. Tension forces of 4 different lengths of stay-cables were derived from the ambient vibration data on the cables. For the integrity assessment of the structure, multi-scale subspace system identification method is now under development using a neural network technique based on the local mode shapes and the cable tensions.

Full-Scale Experimental Validation of High-Fidelity Wireless Measurement on a Historic Truss Bridge

To meet the growing demands to monitor our aging infrastructure, wireless smart sensor networks (WSSN) have been the subject of intense interest due to their versatility and low cost. However, the performance of commercially available sensors is not sufficient to realize the high-fidelity data required for SHM. In particular, synchronization among the wireless sensor nodes has been found to be inadequate for data intensive applications such as SHM. To this end, the Illinois Structural Health Monitoring Project (ISHMP) Services Toolsuite has been developed as an open-source framework to enable reliable SHM application. One of the services, RemoteSensing, was specifically designed to enable monitoring of civil infrastructure through accurate time synchronization and reliable communication. In this paper, the performance of the Imote2 wireless sensor platform using a commercially available low cost sensor board is validated directly against a wired sensor system, in the context of validation of its capability for high-fidelity data measurement for bridge health monitoring. A series of vibration tests have been conducted using human jumping to excite the subject of this study, a 76-m historic steel truss bridge in Mahomet, Illinois. The dynamic properties of the bridge have been obtained by the peak picking method for both wireless and wired systems and compare well; thus, demonstrating the efficacy of the wireless sensor system.

A decentralized receptance-based damage detection strategy for wireless smart sensors

Various structural health monitoring strategies have been proposed recently that can be implemented in the decentralized computing environment intrinsic to wireless smart sensor networks (WSSN). Many are based on changes in the experimentally determined flexibility matrix for the structure under consideration. However, the flexibility matrix contains only static information; much richer information is available by considering the dynamic flexibility, or receptance, of the structure. Recently, the stochastic dynamic damage locating vector (SDDLV) method was proposed based on changes of dynamic flexibility matrices employing centrally collected output-only measurements. This paper investigates the potential of the SDDLV method for implementation on a network of wireless smart sensors, where a decentralized, hierarchical, in-network processing approach is used to address issues of scalability of the SDDLV algorithm. Two approaches to aggregate results are proposed that provide robust estimates of damage locations. The efficacy of the developed strategy is first verified using wired sensors emulating a wireless sensor network. Subsequently, the decentralized damage detection strategy is implemented on MEMSIC’s Imote2 smart sensor platform and validated experimentally on a laboratory scale truss bridge.

Decentralized Bridge Health Monitoring using Wireless Smart Sensors

Wireless Smart Sensor Networks (WSSN) facilitates a new paradigm to structural health monitoring (SHM) for civil infrastructure. Conventionally, SHM systems employing wired sensors and central data acquisition have been used to characterize the state of a structure; however, wide-spread implementation has been limited due to difficulties in cabling, high equipment cost, and long setup time. WSSNs offer a unique opportunity to overcome such difficulties. Recent advances in sensor technology have realized low-cost, smart sensors with on-board computation and wireless communication capabilities, making deployment of a dense array of sensors on large civil structures both feasible and economical. Wireless smart sensors have shown their tremendous potential for SHM in recent full-scale bridge monitoring examples. However, structural damage identification in WSSNs, a primary objective of SHM, has yet to reach its full potential. This paper presents an implementation of the stochastic dynamic damage locating vector (SDDLV) method on the Imote2 sensor platform and experimental validation in a laboratory environment. The WSSN application is developed based on the Illinois SHM Project (ISHMP) Services Toolsuite (http://shm.cs.uiuc.edu), combining decentralized data aggregation, system identification, receptance-based damage detection, and global damage assessment. The laboratory experiment uses a three-dimensional truss structure with a network of Imote2 sensors for decentralized damage identification. Future efforts to deploy a long-term structural health monitoring system for a fullscale steel truss bridge are also described.

Structural damage detection for in-service highway bridge under operational and environmental variability

Structural health monitoring has drawn significant attention in the past decades with numerous methodologies and applications for civil structural systems. Although many researchers have developed analytical and experimental damage detection algorithms through vibration-based methods, these methods are not widely accepted for practical structural systems because of their sensitivity to uncertain environmental and operational conditions. The primary environmental factor that influences the structural modal properties is temperature. The goal of this article is to analyze the natural frequency-temperature relationships and detect structural damage in the presence of operational and environmental variations using modal-based method. For this purpose, correlations between natural frequency and temperature are analyzed to select proper independent variables and inputs for the multiple linear regression model and neural network model. In order to capture the changes of natural frequency, confidence intervals to detect the damages for both models are generated. A long-term structural health monitoring system was installed on an in-service highway bridge located in Meriden, Connecticut to obtain vibration and environmental data. Experimental testing results show that the variability of measured natural frequencies due to temperature is captured, and the temperature-induced changes in natural frequencies have been considered prior to the establishment of the threshold in the damage warning system. This novel approach is applicable for structural health monitoring system and helpful to assess the performance of the structure for bridge management and maintenance.

Structural Damage Detection Using Extended Kalman Filter Combined with Statistical Process Control in Nonlinear Systems

The goal of structural health monitoring is to determine the status of the structure and identify the structural damage. Extended Kalman filter (EKF) has shown effective capability to track the structural parameters for civil structures. When structural damage occurs, the estimations of parameters from EKF will deviate from their constant values, and the changes can be observed visually. However, in view of the environmental and operational effects, structural parameters may fluctuate within a normal range, which may result false alarm problems and cause difficulties to observe the structural damage in real time. In this paper, EKF is combined with Statistical Process Control (SPC) to detect the structural damage in real time. Adaptive SPC control limits are derived based on parameter estimation from EKF and updated dynamically in each time step. When structural damage occurs, the estimation of parameters will deviate outside of the control ranges, thus can be captured by the SPC control limits. This approach is tested on a two-story nonlinear hysteretic structure. The numerical testing results demonstrate that the adaptive SPC-based Kalman filter method is capable to identify and track the general changes of structural parameters and detect damage online with high confidence for nonlinear structural dynamic systems. doi: 10.12783/SHM2015/303

Implementation of a piezoelectric energy harvester in railway health monitoring

With development of wireless sensor technology, wireless sensor network has shown a great potential for railway health monitoring. However, how to supply continuous power to the wireless sensor nodes is one of the critical issues in long-term full-scale deployment of the wireless smart sensors. Some energy harvesting methodologies have been available including solar, vibration, wind, etc; among them, vibration-based energy harvester using piezoelectric material showed the potential for converting ambient vibration energy to electric energy in railway health monitoring even for underground subway systems. However, the piezoelectric energy harvester has two major problems including that it could only generate small amount of energy, and that it should match the exact narrow band natural frequency with the excitation frequency. To overcome these problems, a wide band piezoelectric energy harvester, which could generate more power on various frequencies regions, has been designed and validated with experimental test. Then it was applied to a full-scale field test using actual railway train. The power generation of the wide band piezoelectric array has been compared to a narrow-band, resonant-based, piezoelectric energy harvester.

Optimization of piezoelectric energy harvester for wireless smart sensors in railway health monitoring

Wireless sensor network is one of the prospective methods for railway monitoring due to the long-term operation and low-maintenance performances. How to supply power to the wireless sensor nodes has drawn much attention recently. In railway monitoring, the idea of converting ambient vibration energy from vibration of railway track induced by passing trains to electric energy has made it a potential way for powering the wireless sensor nodes. Nowadays, most of vibration based energy harvesters are designed at resonance. However, as railway vibration frequency is a wide band range, how to design an energy harvester working at that range is critical. In this paper, the energy consumption of the wireless smart sensor platform, Imote2, at different working states were investigated. Based on the energy consumption, a design of a bimorph cantilever piezoelectric energy harvester has been optimized to generate maximum average power between a wide-band frequency range. Significant power and current outputs have been increased after optimal design. Finally, the rechargeable battery life for supplying the Imote2 for railway monitoring is predicted by using the optimized piezoelectric energy harvesting system.