In this paper, a distributed structural damage detection approach is proposed for large size structures under limited input
and output measurements. A large size structure is decomposed into small size substructures based on its finite element
formulation. Interaction effect between adjacent substructures is considered as 'additional unknown inputs' to each
substructure. By sequentially utilizing the extended Kalman estimator for the extended state vector and the least squares
estimation for the unmeasured inputs, the approach can not only estimate the 'additional unknown inputs' based on their
formulations but also identify structural dynamic parameters, such as the stiffness and damping of each substructure.
Local structural damage in the large size structure can be detected by tracking the changes in the identified values of
structural dynamic parameters at element level, e.g., the degrading of stiffness parameters. Numerical example of
detecting structural local damages in a large-size plane truss bridge illustrates the efficiency of the proposed approach. A
new smart wireless sensor network is developed by the authors to combine with the proposed approach for autonomous
structural damage detection of large size structures. The distributed structural damage detection approach can be
embedded into the smart wireless sensor network based on its two-level cluster-tree topology architecture and the distributed computation capacity of each cluster head.
In recent years, some innovative wireless sensing systems have been proposed. However, more exploration and research
on wireless sensing systems are required before wireless systems can substitute for the traditional wire-based systems. In
this paper, a new type of intelligent wireless sensing network is proposed for the heath monitoring of large-size
structures. Hardware design of the new wireless sensing units is first studied. The wireless sensing unit mainly consists
of functional modules of: sensing interface, signal conditioning, signal digitization, computational core, wireless
communication and battery management. Then, software architecture of the unit is introduced. The sensing network has
a two-level cluster-tree architecture with Zigbee communication protocol. Important issues such as power saving and
fault tolerance are considered in the designs of the new wireless sensing units and sensing network. Each cluster head in
the network is characterized by its computational capabilities that can be used to implement the computational
methodologies of structural health monitoring; making the wireless sensing units and sensing network have "intelligent"
characteristics. Primary tests on the measurement data collected by the wireless system are performed. The distributed
computational capacity of the intelligent sensing network is also demonstrated. It is shown that the new type of
intelligent wireless sensing network provides an efficient tool for structural health monitoring of large-size structures.
In this paper, the technique recently proposed by the authors for the detection of structural local damage in large size
linear structures is extended to explore the detection of local damage in some complex nonlinear structures. The
technique is based on substructural approach in which a complex nonlinear structure is decomposed into substructures
Interaction effect between adjacent substructures is accounted by considering the interaction forces at substructural
interfaces as the 'unknown inputs' to the substructures. An algorithm utilizing the classical Kalman extended estimator
and the recursive least squares estimation for the unknown inputs is proposed to identify structural parameters at element
level and the 'unknown inputs' to the substructure. Two cases that measurements at the substructure interfaces are
available or not available are considered. Structural local damage is estimated from the change of structural parameters,
such as the degradation of the stiffness, at element level. The technique enables distributed identification of local damage
in complex nonlinear structures utilizing only a limited number of measured acceleration responses. Performance of
proposed technique is illustrated by a numerical example of detecting local damage in a multi-story hysteretic building. It
is shown that the proposed technique can be used for detecting structural local damage in some complex nonlinear
structures.
Damage detection is the core technique of structure health monitoring systems. Mostly, the detection is based on
comparison of initial signatures (frequency, mode shapes and so on) of intact structure with that of damaged structure.
The techniques based on the analysis of vibration data of structures have received great attention in recent years.
Generally, high-rise buildings have enough security under wind or some other natural conditions. Instances of damage
caused by routine work can be rarely found. But under earthquake, high-rise buildings damages may occur on some
weakness areas. In this paper, based on establishing the stiffness matrix of the columns and beams with joint damage, the
finite element model of the damaged frame structure is set up. Calculating the modal date by the finite element model
between the intact and damaged structure, simple and multi damages being imitated at the locations of the joints, the
curvature mode shape method is used to identify the damage. The numerical example shows that the structural damage
can be efficiency identified by using vibration characteristics of the building.
In practical structural health monitoring, it is essential to develop an efficient technique which can detect structural local
damage utilizing only a limited number of measured acceleration responses of structures subject to unknown
(unmeasured) excitations inputs. In this paper, a finite-element based time domain system identification method is
proposed for this purpose. Structure state vectors are treated as implicit functions of structural dynamic parameters and
excitations. The unknown structural dynamic parameters and excitation inputs are identified by an algorithm based on
recursive least squares estimation with unknown excitations (RLSE-UI). Structural damage at element level is detected
by the degrading of stiffness of damaged structural elements. Numerical simulation of a 3-story building demonstrates
the proposed method can identify structural element stiffness parameters with good accuracy and structural damage at
element level can be located from the degrading of element stiffness parameters.
KEYWORDS: Damage detection, Autoregressive models, Detection and tracking algorithms, Sensors, Sensing systems, Signal detection, Structural monitoring, Time series analysis, System identification, Finite element methods
In this paper, a damage detection and localization algorithm, which is suitable for the implementation of automated damage detection system based on the wireless structural monitoring sensing network, is presented. Vibration signals obtained from sensors are modeled as autoregressive moving average (ARMA) time series. Coefficients of the ARMA models are estimated by a two-stage linear identification process. Stable poles and residues of the ARMA models are extracted based on the stability tolerances on the change in frequency and damping ratios. Then, these stable poles and residues are transmitted to the centralized data server, where structural damage is detected from the change of the poles estimated from undamaged and damaged structural signals, damage locations are identified by the change ratio of the estimated mean values of first vibration mode shape of the undamaged and damaged structure. Implementation of the damage detection and localization algorithm in the wireless structural monitoring sensing system for automated damage detection is illustrated. To test the efficacy of the damage detection and localization methodologies, the algorithm is applied on the benchmark problem designed by the ASCE task group on health monitoring.
Wireless health monitoring schemes are innovative techniques, which effectively remove the disadvantages associated with current wire-based sensing systems, i.e., high installation and upkeep costs. However, recorded data sets may have relative time-delays due to the blockage of sensors or inherent internal clock errors. In this paper, two algorithms are proposed for the synchronization of the recorded asynchronous data measured from sensing units of a wireless monitoring system. In the first algorithm, the input signal to a structure is measured. Time-delay between an output measurement and the input is identified based on the minimization of errors of the ARX (auto-regressive model with exogenous input) models for the input-output pair recordings. The second algorithm is applicable when a structure is subject to ambient excitation and only output measurements are available. ARMAV (auto-regressive moving average vector) models are constructed from two output signals and the time-delay between them is evaluated based on the minimization of errors of the ARMAV models. The proposed algorithms are verified by simulation data and recorded seismic response data from multi-story buildings.
In this paper, we make a brief study of some of the important requirements of a structural monitoring system for civil infrastructures and explain the key issues that are faced in the design of a suitable wireless monitoring strategy. Two-tiered wireless sensor network architecture is proposed as a solution to these issues and the protocol used for the communication in this network is described. The power saving strategies at various levels, from the network architecture, to communication protocol, to the sensor unit architecture are explained. A detailed analysis of the network is done and the implementation of this network in a laboratory setting is described.
KEYWORDS: Matrices, Signal processing, Time metrology, Numerical simulations, System identification, Chemical elements, Solids, Frequency modulation, Civil engineering, Smart materials
Recently, the method of Hilbert transform has been used successfully by the authors to identify parameters of linear structures with real eigenvalues and eigenvectors, e.g., structures with proportional damping. Frequently, linear structures may not have proportional damping so that normal modes do not exist. In this case, all the eigenvalues, eigenvectors and modeshapes are complex. In this paper, the Hilbert transform and the method of Empirical Mode Decomposition are used to identify the parameters of structures with nonproportional damping using the impulse response data. Measured impulse response signals are first decomposed into Intrinsic Mode Functions using the method of Empirical Mode Decomposition with intermittency criteria. An Intrinsic Mode Function (IMF) contains only one characteristic time scale (frequency), which may involve the contribution of a complex conjugate pair of modes with a unique frequency and a damping ratio, referred to as the modal response. It is shown that all the modal responses can be obtained from IMFs. Then, each modal response is decomposed in the frequency-time domain to yield instantaneous phase angle and amplitude as functions of time using the Hilbert transform. Based on only a single measurement of the impulse response time history at one location, the complex eigenvalues of the linear structure can be identified using a simple analysis procedure. When the response time histories are measured at all locations, the proposed methodology is capable of identifying the complex modeshapes as well as the mass, damping and stiffness matrices of the structure. The effectiveness and accuracy of the methodology presented are demonstrated through numerical simulations. It is shown that complete dynamic characteristics of linear structures with nonproportional damping can be identified effectively using the Hilbert transform and the Empirical Mode Decomposition method.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.