Table of contents

In this paper we propose a special type of colocated feedback controller for smart structures. The controller is a parallel combination of high-Q resonant circuits. Each of the resonant circuits is tuned to a pole (or the resonant frequency) of the smart structure. It is proven that the parallel combination of resonant controllers is stable with an infinite gain margin. Only one set of actuator–sensor can damp multiple resonant modes with the resonant controllers. Experimental results are presented to show the robustness of the proposed controller in damping multimode resonances.

The control of embedded shape memory alloy (SMA) actuators has recently become a topic of interest in the field of smart structures. The inherent difficulties associated with SMA actuators has resulted in a variety of approaches. Homogenization provides a simplified, yet mathematically rigorous, method of determining average stress and strain fields in a composite. A modified constitutive model is presented based on experimental results demonstrating the inability of most simple phenomenological models to capture the effective behavior of SMAs during thermal activation. A feedforward controller is presented for a SMA composite based on the homogenization of a modified phenomenological model for SMAs in a linear matrix.

Much of the current rotorcraft research is focused on improving performance by reducing unwanted helicopter noise and vibration. One of the most promising active rotorcraft vibration control systems is an active trailing edge flap. In this paper, an induced-shear piezoelectric tube actuator is used in conjunction with a simple lever–cusp hinge amplification device to generate a useful combination of trailing edge flap deflections and hinge moments. A finite-element model of the actuator tube and trailing edge flap (including aerodynamic and inertial loading) was used to guide the design of the actuator–flap system. A full-scale induced shear tube actuator flap system was fabricated and bench top testing was conducted to validate the analysis. Hinge moments corresponding to various rotor speeds were applied to the actuator using mechanical springs. The testing demonstrated that for an applied electric field of 3 kV cm⁻¹, the tube actuator deflected a representative full-scale 12 inch flap ±2.8° at 0 rpm and ±1.4° for a hinge moment simulating a 400 rpm condition. The per cent error between the predicted and experimental full-scale flap deflections ranged from 4% (low rpm) to 12.5% (large rpm). Increasing the electric field to 4 kV cm⁻¹ results in ±2.5° flap deflection at a rotation speed of 400 rpm, according to the design analysis. A trade study was conducted to compare the performance of the piezoelectric tube actuator to the state of the art in trailing edge flap actuators and indicated that the induced-shear tube actuator shows promise as a trailing edge flap actuator.

In this paper, various methods for the real-time estimation of multi-modal frequencies are realized in real time and compared through numerical and experimental tests. These parameter-based frequency estimation methods can be applied to various engineering fields such as communications, radar and adaptive vibration and noise control. Well-known frequency estimation methods are introduced and explained. The Bairstow method is introduced to find the roots of a characteristic equation for estimations of multi-modal frequencies, and the computational efficiency of the Bairstow method is shown quantitatively. For a simple numerical test, we consider two sinusoids of the same amplitudes mixed with various amounts of white noise. The test results show that the auto regressive (AR) and auto regressive and moving average (ARMA) methods are unsuitable in noisy environments. The other methods apart from the AR method have fast tracking capability. From the point of view of computational efficiency, the results reveal that the ARMA method is inefficient, while the cascade notch filter method is very effective. The linearized adaptive notch filter and recursive maximum likelihood methods have average performances. Experimental tests are devised to confirm the feasibility of real-time computations and to impose the severe conditions of drastically different amplitudes and of considerable changes of natural frequencies. We have performed experiments to extract the natural frequencies from the vibration signal of wing-like composite plates in real time. The natural frequencies of the specimen are changed by added masses. Especially, the AR method exhibits a remarkable performance in spite of the severe conditions. This study will be helpful to anyone who needs a frequency estimation algorithm for real-time applications.

This paper presents the results of longitudinal wave propagation in piezoelectric coupled rod structures. The deduction of non-dispersive or dispersive characteristics of the structures is based on a classical rod model and the Mindlin–Herrmann rod model. In the classical model, correction factors for piezoelectric effects are introduced to provide remedy for the discontinuity of the normal stress at the interface of the host rod and the piezoelectric layer. The non-dispersive characteristics of the structures, in which the piezoelectric effects are modelled are thus obtained by this classical model. The two-mode characteristics for the piezoelectric coupled rod structures are obtained by the Mindlin–Herrman model in which the lateral inertia is considered. This model is more accurate in predicting the dispersive characteristics of the structures. The results of the long- and short-wavelength limits are obtained as by-products. The results of phase velocity by the two models are agreeable at low wavenumber by adjusting the correction factor defined in the classical model. The results of this paper can serve as a reference for future study on wave propagation in coupled structures as well as in the design of smart structures incorporating piezoelectric materials.

In this paper we present a dynamic analytical model for the torsional vibration of an anisotropic piezoelectric laminate induced by the extension-twisting coupling effect. In the present approach, we use the Hamilton principle and a reduced bending stiffness method for the derivation of equations of motion. As a result, the in-plane displacements are not involved and the out-of-plane displacement of the laminate is the only quantity to be calculated. Therefore, the proposed method turns the twisting of a laminate with structural coupling into a simplified problem without losing its features. We give analytical solutions of the present model with harmonic excitation. A parametric study is performed to demonstrate the present approach.

We have used quasi-static equations of piezoelectricity to derive a finite element formulation capable of modelling two different kinds of piezoelastically induced actuation in an adaptive composite sandwich beam. This formulation is made to couple certain piezoelectric constants to a transverse electric field to develop extension-bending actuation and shear-induced actuation. As an illustration, we present a sandwich model of three sublaminates: face/core/face. We develop a control scheme based on the linear quadratic regulator/independent modal space control (LQR/IMSC) method and use this to estimate the active stiffness and the active damping introduced by shear and extension-bending actuators. To assess the performance of each type of actuator, a dynamic response study is carried out in the modal domain. We observe that the shear actuator is more efficient in actively controlling the vibration than the extension-bending actuator for the same control effort.

We present an inverse procedure for the detection of a three-dimensional crack in plates and shells using a genetic searching algorithm. The integral strain, which can be measured by optical fibers, is utilized as the input for the inverse procedure. Two parameters, each with three components, are used to describe the location and size of the crack. A shell element model is created with a re-mesh operation in the numerical simulation of the integral strains. The squared difference of the integral strains obtained from a plate with an actual crack and from a plate with a trial crack is defined as the objective function. The best crack parameters are found by minimizing the objective function using a searching technique of the micro genetic algorithm (μGA). We have presented numerical examples to demonstrate the procedures and effectiveness of the method.

This paper presents a methodology for identifying variable-structure nonlinear models of magneto-rheological dampers (MRD) and similar devices. Its peculiarity with respect to the mainstream literature is to be especially conceived for obtaining models that are structurally simple, easy to estimate and well suited for model-based control. This goal is pursued by adopting linear-in-the-parameters NARX models, for which an identification method is developed based on the minimization of the simulation error. This method is capable of selecting the model structure together with the parameters, thus it does not require a priori structural information. A set of validation tests is reported, with the aim of demonstrating the technique's efficiency by comparing it to a widely accepted MRD modelling approach.

A three-dimensional finite-element closed-loop model has been developed to predict the effects of active–passive damping on a vibrating structure. The Golla–Hughes–McTavish method is employed to capture the viscoelastic material behavior in a time domain analysis. The parametric study includes the different control gains as well as geometric parameters related to the active constrained layer damping (ACLD) treatment. Comparisons are made among several ACLD models, the passive constrained model and the active damping model. The results obtained here reiterate that ACLD is somewhat better for vibration suppression than either the purely passive or the active system and provides higher structural damping with less control gain when compared to the purely active system. Since the ACLD performance can be reduced by the viscoelastic layer, the design of the ACLD model must be given a careful consideration in order to optimize the effect of passive damping.

In this paper on finite element linear quadratic regulator (LQR) vibration control of smart piezoelectric composite plates, we propose the use of the total weighted energy method to select the weighting matrices. By constructing the optimal performance function as a relative measure of the total kinetic energy, strain energy and input energy of the system, only three design variables need to be considered to achieve a balance between the desired higher damping effect and lower input cost. Modal control analysis is used to interpret the effects of three energy weight factors on the damping ratios and modal voltages and it is shown that the modal damping effect will increase with the kinetic energy weight factor, approaching √2/2 as the strain energy weight factor increases and decrease with the input energy weight factor. Numerical results agree well with those from the modal control analysis. Since the control problem is simplified to three design variables only, the computational cost will be greatly reduced and a more accurate structural control analysis becomes more attractive for large systems.

In this paper we report on the application of an in situ health monitoring system, comprising an array of piezoceramic wafer elements, to the detection of fatigue degradation in metallic specimens exposed to cyclic loading. Lamb waves, transmitted through a beam test coupon, are sensed using small surface-mounted piezotransducer elements, and the signals are then autonomously analysed for indications relating to the onset of structural degradation. The experimental results confirm the efficacy of the approach and provide a demonstration of good robustness under realistic loading conditions, emphasizing the great potential for developing an automated in situ structural health monitoring system for application to fatigue-prone operational structures, such as aircraft.

The use of neural networks for identification and control of smart structures is investigated experimentally. Piezoelectric actuators are employed to suppress the vibrations of a cantilevered plate subject to impulse, sine wave and band-limited white noise disturbances. The neural networks used are multilayer perceptrons trained with error backpropagation. Validation studies show that the identifier predicts the system dynamics accurately. The controller is trained adaptively with the help of the neural identifier. Experimental results demonstrate excellent closed-loop performance and robustness of the neurocontroller.

We have used the plane strain theory of transversely isotropic bodies to study a piezoelectric cantilever. In order to find the general solution of a density functionally gradient piezoelectric cantilever, we have used the inverse method (i.e. the Airy stress function method). We have obtained the stress and induction functions in the form of polynomials as well as the general solution of the beam. Based on this general solution, we have deduced the solutions of the cantilever under different loading conditions. Furthermore, as applications of this general solution in engineering, we have studied the tip deflection and blocking force of a piezoelectric cantilever actuator. Finally, we have addressed a method to determine the density distribution profile for a given piezoelectric material.

Shape memory alloys (SMAs) are being embedded in or externally attached to smart structures because of the large amount of actuation deformation and force that these materials are capable of producing when they are heated. Previous investigations have focused primarily on using single or opposing SMA wires exhibiting the two-way shape memory effect (SME) because of the simplicity with which the repeatable actuation behavior of the structure can be predicted. This repeatable actuation behavior is achieved at the expense of reduced levels of recoverable deformation. Alternatively, many potential smart structure applications will employ multiple SMA wires exhibiting a permanent one-way SME to simplify fabrication and increase the recoverable strains in the structure. To employ the one-way wires, it is necessary to investigate how they affect the recovery of large strains when they are embedded in a structure. In this investigation, the large strain recovery of a one-way SMA wire embedded in a flexible polyurethane is characterized using the novel deformation measurement technique known as digital image correlation. These results are compared with a simple actuation model and a three-dimensional finite element analysis of the structure using the Brinson model for describing the thermomechanical behavior of the SMA. Results indicate that the level of actuation strain in the structure is substantially reduced by the inelastic behavior of the one-way SMA wires, and there are significant differences between the deformations of the matrix material adjacent to the SMA wires and in the region surrounding it. The transformation behavior of the SMA wires was also determined to be volume preserving, which had a significant effect on the transverse strain fields.

The effects of dispersed phase saturation magnetization and applied magnetic fields on the rheological properties of magnetorheological (MR) fluids are described. MR fluids based on two different grades of carbonyl iron powder with different average particle size, 7–9 μm (grade A) and 2 μm (grade B), were prepared. Vibrating sample magnetometer measurements showed that the saturation magnetization values were 2.03 and 1.89 T for grades A and B, respectively. Rheological measurements were conducted for 33 and 40 vol% grade A and grade B based MR fluids with a specially built double Couette strain rate controlled rheometer at flux densities ranging from 0.2 to ∼0.8 T. The yield stresses of 33 and 40 vol% grade A were 100 ± 3 and 124 ± 3 kPa, respectively at 0.8 ± 0.1 T. The yield stress values of MR fluids based on finer particles (grade B) were consistently smaller. For example, the yield stresses for 33 and 40 vol% grade B based MR fluid were 80 ± 8 and 102 ± 2 kPa, respectively at 0.8 ± 0.1 T. The yield stresses at the flux density approaching magnetic saturation in particles (B ∼ 0.8T) were found to increase quadratically with the saturation magnetization (μ₀M_s) of the dispersed magnetic phase. This is in good agreement with the analytical models of uniformly saturated particle chains developed by Ginder and co-workers. The results presented here show that the decrease in yield stress for finer particle based MR fluids is due to the relatively smaller magnetization of the finer particles.

In this paper we present experimental investigations of the vibration testing of an inflated, thin-film torus using smart materials. Lightweight, inflatable structures are very attractive in satellite applications. However, the lightweight, flexible and highly damped nature of inflated structures poses difficulties in ground vibration testing. In this study, we show that polyvinylidene fluoride (PVDF) patches and recently developed macro-fiber composite actuators may be used as sensors and actuators in identifying modal parameters. Both smart materials can be integrated unobtrusively into the skin of a torus or space device forming an attractive testing arrangement. The addition of actuators and PVDF sensors to the torus does not significantly interfere with the suspension modes of a free–free boundary condition, and can be considered an integral part of the inflated structure. The results indicate the potential of using smart materials to measure and control the dynamic response of inflated structures.

The purpose of this paper is to demonstrate that vibrations of a truss structure can be suppressed nicely by a magneto-rheological (MR) fluid variable damper for semiactive vibration suppression. A variable MR fluid damper was designed and fabricated for this study. The principal characteristics of an MR damper were measured in dynamic tests, and a mathematical model of the damper was proposed. To investigate if the variable damper effectively suppresses the vibration of actual truss structures, semiactive vibration suppression experiments were performed using a cantilevered ten-bay truss beam. The experimental result has shown that the vibration was suppressed nicely by the variable MR damper, and that was compared with that of an electro-rheological (ER) damper investigated in previous research. The MR damper showed a higher performance than that of the ER damper.

In this paper we are concerned with the design, manufacture and performance test of a lightweight piezo-composite curved actuator (called LIPCA) using a top carbon fiber composite layer with near-zero coefficient of thermal expansion (CTE), a middle PZT ceramic wafer, and a bottom glass/epoxy layer with a high CTE. The main point of the design for LIPCA is to replace the heavy metal layers of THUNDER™ by lightweight fiber reinforced plastic layers without losing the capabilities for generating high force and large displacement. It is possible to save up to about 40% of the weight if we replace the metallic backing material by the light fiber composite layer. We can also have design flexibility by selecting the fiber direction and the size of prepreg layers. In addition to the lightweight advantage and design flexibility, the proposed device can be manufactured without adhesive layers when we use an epoxy resin prepreg system. Glass/epoxy prepregs, a ceramic wafer with electrode surfaces, and a carbon prepreg were simply stacked and cured at an elevated temperature (177 °C) after following an autoclave bagging process. We found that the manufactured composite laminate device had a sufficient curvature after being detached from a flat mould. An analysis method using the classical lamination theory is presented to predict the curvature of LIPCA after curing at an elevated temperature. The predicted curvatures are in quite good agreement with the experimental values. In order to investigate the merits of LIPCA, performance tests of both LIPCA and THUNDER™ have been conducted under the same boundary conditions. From the experimental actuation tests, it was observed that the developed actuator could generate larger actuation displacement than THUNDER™.

In this technical note we survey the area of smart collision information processing sensors. We review the existing technologies to detect collision or overlap between fast moving physical objects or objects in virtual environments, physical environments or a combination of physical and virtual objects. We report developments in the collision detection of fast moving objects at discrete time steps such as two consecutive time frames, as well as continuous time intervals such as in an interframe collision detection system. Our discussion of computational techniques in this paper is limited to convex objects. Techniques exist however to efficiently decompose non-convex objects into convex objects. We also discuss the tracking technologies for objects from the standpoint of collision detection or avoidance.