Proc. SPIE. 10970, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2019
KEYWORDS: Mathematical modeling, Data modeling, Sensors, Earthquakes, Structural health monitoring, System identification, Performance modeling, Systems modeling, Global Positioning System, Model-based design
This paper presents the application of a probabilistic method to estimate the complete dynamic response and state of damage of a large-scale structural system subjected to strong base excitations. The structure considered consists of a sevenstory slice of a shear wall building tested at the George E. Brown Jr. Network for Earthquake Engineering Simulation site at the University of California at San Diego. The acceleration response measured at limited locations is combined with a nonlinear model to estimate the complete dynamic response at unmeasured degrees of freedom. To improve the predicting capability of the mathematical model a subset of the parameters is jointly estimated with the response, a strategy known as joint state-parameter estimation or augmented state estimation. The estimated response is used to compute a damage index as a quantitative measure of damage.
Strain measurements are essential in structural health monitoring. Traditional strain gages require physical
contact between the sensor and read-out device, perturb the surface being monitored, and allow measurement
only at the specific location and orientation axis of the sensor. We demonstrate a novel non-contact, multi-
point, multi-directional strain sensing approach that overcomes these limitations. In our method, the surface is
coated with a thin film of "smart skin" containing individualized single-walled carbon nanotubes in a polymeric
host. After curing, substrate strains are transmitted through the polymer film to embedded nanotubes. This
induces axial strains in the nanotubes, systematically shifting the wavelengths of their characteristic near-infrared
fluorescence peaks. To measure strain, a visible laser excites nanotubes at points of interest on the surface, and
the near-infrared emission is collected and spectrally analyzed. Observed spectral shifts reveal quantitative strain
values. Laboratory tests show sensitivity down to ~400µm, limited by mechanical properties of the polymeric
host film. We also vary excitation beam polarization to find the axis of substrate strain. Our method provides
spatial resolution down to its gage length of ~100µm. Because the entire substrate is coated with nanoscale
strain sensors, measurements can be made at arbitrary locations to construct a full strain map. We will describe
recent smart skin refinements involving selection of polymer host, nanotube surfactant, nanotube dispersion
method, and preparation protocol. Finally, we characterize the orientational distribution of nanotubes using a
In this paper, an approach based on a new damage index-Distributed Force Change(WDFC), for monitoring the structural health of risers used for production in deep-water floating platforms, is presented. Experiments of a scaled pipe are carried out to validate the vibration based damage identification method. The influences of multiple cracks in the WDFC damage index are studied. Futhermore, this paper demonstrates the effectiveness of wave propagation based structural health monitoring (SHM) strategies within the pipe model. This is realized based on the results of numerical investigation obtained by the use of Finite Element Method(FEM) together with application of Time-of-Flight(FoT) damage identification method in which the damage severity is indicated by Root Mean Square(RMS) of the damage-reflected wave. The influence of crack(s) in the riser/pipe on the wave propagation are studied. The results from the experiments and numerical analysis indicate that both the two damage identification methods can provide information about the estimated crack location(s) and the possible extent of crack. Hence the two methods are suitable for globally and locally monitoring the structural health of deepwater risers respectively.
This paper addresses tracking-control of hysteretic systems using a gain-scheduled (GS) controller. Hysteretic
system with variable stiffness and damping is represented as a quasi linear parameter varying (LPV) system.
Designed controller is scheduled on the measured/estimated stiffness and damping in real-time. GS controller
is constructed from the parameter dependent Lyapunov matrices, which are obtained as optimal solutions of
linear matrix inequalities (LMIs) that ensures the feasibility solution for closed loop system performance. The
proposed method is worked on semiactive independently variable stiffness (SAIVS) device. It is shown that the
gain-scheduled controller developed for the quasi-LPV system results in excellent tracking performance even in
the cases where robust-H∞ controller failed to function.
Structural damage will change the dynamic characteristics, including natural frequencies, modal shapes, damping ratios
and modal flexibility matrix of the structure. Modal flexibility matrix is a function of natural frequencies and mode
shapes and can be used for structural damage detection and health monitoring. In this paper, experimental modal
flexibility matrix is obtained from the first few lower measured natural frequencies and incomplete modal shapes. The
optimization problem is then constructed by minimizing Frobenius norm of the change of flexibility matrix. Gauss-
Newton method is used to solve the optimization problem, where the sensitivity of flexibility matrix with respect to
structural parameters is calculated iteratively by only using the first few lower modes. The optimal solution corresponds
to structural parameters which can be used to identify damage sites and extent. Numerical results show that flexibility-based
method can be successfully applied to identify the damage elements and is robust to measurement noise.
A novel sensor failure detection method is developed in this paper. Sensor failure considered in this paper can be any type of measurement error that is different from the true structural response. The sensors are divided into two groups, sensors that correctly measure the structural responses, are termed as reference sensors, and sensors that may fail to correctly measure the structural responses, are termed as uncertain sensors henceforth. A sensor error function, one for each uncertain sensor, is formulated to detect the corresponding uncertain sensor failure in real-time, using the measurements from reference sensors and the uncertain sensor being monitored. The sensor error function is derived using the indirect and direct approaches. In the indirect approach, the error function is obtained from the state space model in combination with the inverse model and interaction matrix formulation. The input term is eliminated from the error function by applying inverse model and the interaction matrix is applied to eliminate the state and all uncertain sensor measurement terms excepted the examined uncertain sensor from the error function. In the direct approach, the coe±cients of the error function can be directly calculated from the healthy measurement data from the examined uncertain sensor and all reference sensors without having to know the state-space model of the system. Thus the need to know the state-space model of the plant can be bypassed. The sensor failure detection formulations are investigated numerically using a four degree-of-freedom spring-mass-damper system and experimentally using a 4m long NASA 8-bay truss structure. It is shown by means of numerical and experimental results that the developed sensor failure formulations correctly detect the instants of sensor failure and can be implemented in real structural systems for sensor failure detection.
This paper presents a new and innovative semi-active variable stiffness tuned mass damper (SAIVS-TMD). The system has the distinct advantage of retuning in real time thus making the system robust to changes in building stiffness and damping, whereas the passive tuned mass damper (TMD) can only be tuned to a fixed frequency. The SAIVS-TMD is based on a novel semi-active variable stiffness control (SAIVS) device. SAIVS system requires nominal power for operation as compared to active tuned mass dampers. The SAIVS-TMD is retuned using a new control algorithm based on instantaneous frequency estimation using Hilbert transform and short-time Fourier transform (STFT). An analytical model of a three-story structure with SAIVS-TMD is developed. Numerical simulations are performed using the analytical model. The system is implemented in a 1:10 scale three-story scale model in real time using a digital signal processing system and controller. Shake table test results of the system with the SAIVS-TMD are presented. It is shown that the SAIVS-TMD is very effective in reducing the response and providing retuning capability when the building stiffness changes, whereas the TMD is mistuned and loses its effectiveness. Analytical modeling and comparisons between analytical and experimental results are also presented.