The paper presents a short overview of the 1997 Smart Structures and Integrated Systems Conference program from the perspective of an overview of the technologies contributing to this field, their interrelationships, and selected key areas in which development is needed to further their application. Four specific technology areas are addressed; integrated systems applications, vibration and acoustic control, actuation technologies for smart structures, and health monitoring, damage detection and damage mitigation. The states of the art in each of these four areas, as exemplified by contributions to this conference, are examined and evaluated with respect to technical maturity and application challenges. Conclusions are drawn from this examination which indicate several, interrelated technical areas in which future concentration of study and development would significantly benefit the evolution of smart structures and other integrated smart systems.

In this paper we develop an integrated model of the amplifier, actuator material, and structural dynamics of a smart material. Stacked electrostrictor actuators are driven by a PWM amplifier. These actuators are attached to a graphite epoxy composite base and they drive an alumina headmass. The whole device, designed from underwater acoustics, is potted. We model the actuators using nonlinear constitutive laws. These equations are integrated into a dynamic model of the amplifier. Then these equations are integrated with a dynamic model of the actuator. Finally, a finite element model is developed for the multilayered composite material that provides the dynamic response of the device. Using superelements, stiffness coefficients are extracted from the finite element model of the multilayered composite material.

Space missions require higher and higher performance such as high pointing accuracy and stability, and high shape precision. Passive damping means often cannot fulfill the requirements. Besides space applications at the same time numerous applications in machine design require higher accuracy. For a lot of applications the passive measures come up against limits. Active technologies have to be considered more often. Active mechanical components are more and more used as a necessary step towards adaptive structures. Active mechanical interfaces are simpler systems having very useful applications and can be used as kind of test benches in order to master the most exacting technologies. The main advantages of such an active interface are the following: (1) state-of- the-art sensors and actuators can be used, (2) the mechanical design of the interface is conventional, (3) the passive behavior of the system is not deteriorated, (4) the design is compact and rather easy to integrate, (5) easy repair (replacement) of the active mechanical part, (6) standardization of the interfaces results in cost reductions. An important property in such intermediate step is that no major redesign of the conventionally designed mechanical structure should be needed. The design, numerical analysis, manufacturing and test of a fully integrated active tubular interface (ATI) is presented. The design of the ATI includes the optimal laminate stacking sequence with respect to maximum deformation efficiency. The results of an active damping application, an antenna support beam, including the controller layout are discussed.

Satellite programs are moving in the direction of smaller and lighter structures. Technological advances have permitted more sophisticated equipment to be consolidated into compact spaces. Micro-satellites, between 10 and 100 kg, will incorporate micro-electric devices into the lay-up of the satellite structure. These structures will be designed to carry load, provide thermal control, enhance damping, and include integrated passive electronics. These multifunctional structures offer lighter weight, reduced volume, and a 'smarter' overall package for incorporation of sensors, electronics, fiber optics, powered appendages or active components. McDonnell Douglas Corporation (MDC) has applied technology from the synthesis and processing of intelligent cost effective structures (SPICES) and independent research and development (IRAD) programs to the modular instrument support system (MISS) for multifunctional space structures and micro-satellites. The SPICES program was funded by the Defense Advanced Research Projects Agency (DARPA) to develop affordable manufacturing processes for smart materials to be used in vibration control, and the MISS program was funded by NASA-Langley. The MISS program was conceived to develop concepts and techniques to make connections between different multifunctional structures. MDA fabricated a trapezoidal carbon composite structure out of IM7/977-3 tape prepreg. Flex circuits, thermal and optical conduits were embedded to realize a utility modular connector. These provide electrical, thermal, optical and mechanical connections between micro- satellite components. A quick disconnect mount was also developed to accommodate a variety of devices such as solar arrays, power sources, thermal transfer and vibration control modules.

This paper considers the use of non-intrusive piezoceramic (PZT-5H) strain sensors coupled with wavelet analysis to characterize the vibration signature from a planetary geartrain under simulated fault conditions. This planetary geartrain assembly consists of an electric motor, coupled with a planetary geartrain drive (reduction ratio: 4 to 1) and a belt drive (reduction ratio: 1.67 to 1). The piezoelectric sensors are placed at strategic locations on the geartrain ring gear to measure loads created by the meshing of the sun, planet and ring gears. Strain time histories are obtained at 500 rpm under no-aerodynamic loading. In addition, output rpm and motor temperature are monitored. Each vibration signature is passed into a harmonic wavelet transform algorithm and the mean square wavelet map is computed. This map is used to classify three types of planet gear faults: missing tooth, spalled gear and cracked gear. Results from the application of the harmonic wavelet transform indicate that the different simulated fault cases have unique characteristics which appear in the mean square wavelet map. Visual inspection of the wavelet map provides a wealth of information regarding the frequency of occurrence of the fault under steady rpm and the associated intensity or power changes at various harmonics. When combined in the mean square wavelet map, these two pieces of information allow for better detection and classification of the fault.

An active non-linear vibration absorber for flexible structures is developed. The absorber exploits the inherent quadratic nonlinearity of the actuator material Terfenol-D to produce a two-to-one autoparametric resonance between the forced vibrations of a structure and a second-order analog controller circuit. Nonlinear resonance of this type exhibits the well-known saturation phenomenon. When the structure is forced near resonance, its response saturates to a small value. This type of control has been demonstrated by previous researchers using linear actuators where nonlinearities were introduced via the analog circuit. In contrast, we use the natural nonlinearity of the Terfenol-D material to achieve the same results. We develop the theory and present experimental results for the control of the first and second modes of a cantilever beam. We also consider the application of the strategy experimentally when the forcing is due to a rotating imbalance. In this case, the excitation source is nonideal. Our results indicate that the saturation based control technique implemented with a Terfenol-D actuator constitutes an effective nonlinear vibration absorber.

This article presents an approach to vibration suppression that does not use a viscous damper. The actuator used behaves effectively as a spring with an adjustable unstressed length. By adjusting this unstressed length on-line from feedback, it is possible to obtain fast settling times and superior shock isolation characteristics. At the instants when energy in the actuator is maximized, we discard this energy as heat in the working fluid by opening a low power solenoid valve for a short time interval. This sets the unstressed length of the actuator, or effective spring, to be zero. The total energy in the vibrating structure is therefore decreased at these instants of unstressed length adjustment. Analytical results and simulations are given that demonstrate the performance of our approach.

To apply smart structures to buildings, a new vibration control strategy is proposed for flexural-shear type frame structures with smart structures using piezoelectric actuators. Actuators are incorporated in columns, and the response of the entire structure is reduced by controlling the bending moment and axial force of columns. Combined application of these two control is performed by arrangement of actuators. Two control strategies were tested. One is model matching method and author is H(infinity) control theory. Excitation tests were conducted with a 4-story fame structure of total weight 2,000 kg and height 3.7 m having H-steels incorporating actuators as columns. Effectiveness of combined use of bending moment control and axial force control of columns with smart structures was confirmed.

A cylindrical test-bed has been designed and modeled to aid in the study and control of interior acoustics in aircraft. The test-bed accounts for local as well as global structural- acoustic dynamics encountered in typical aircraft. The design is based on several existing aircraft and models used to study aircraft dynamics. The test-bed incorporates the basic geometry and materials common to a majority of aircraft, including an aluminum skin shrouding a framed structure composed of ribs and stringers. The design is approximately a one-third scale representation with a cylindrical geometry measuring 0.91 m diameter and 1.98 m long. The test-bed has been modeled using finite element method and Rayleigh-Ritz assumed modes analysis. The models were used to refine the design of the test-bed as well as to model the coupled structural-acoustic dynamics. The models predict that the test-bed will have a modal structure commensurate with experimental identifications on existing aircraft.

A reconfigurable actuator-sensor array is demonstrated for the active control of sound and vibration. The concept is motivated by applications in which shaped polyvinylidene fluoride films are used as error sensors for feedforward and feedback noise control. The advantage of the present concept, as compared to a fixed-shape sensor, is that the reconfigurable array can be adapted on line to account for uncertain structural dynamics, making it a more effective error sensor for applications in which the structural dynamics are not accurately known a priori. The array developed for this work consists of twenty-two polyvinylidene fluoride film sensors and four multilayer piezoceramic actuators connected to a set of reprogrammable electronics. The reprogrammable electronics consist of two 16 by 1 digitally programmable gain boards and a digital signal processor with two analog inputs and outputs. The gains on the sensor array are set via the digital I/O of the digital signal processor. An off-line, frequency domain technique is used to design modal filters with the array of film sensors. Feedback and feedforward control algorithms are then implemented to demonstrate the use of the actuator-sensor array for noise and vibration suppression. The results indicate that the reconfigurable array technique has merit for noise and vibration control, although a fully adaptive, on-line algorithm has yet to be implemented.

Extensions of our previous analytical and experimental investigations into active control of bandlimited disturbances in a three-dimensional enclosure are presented. The frequency range of the bandlimited disturbance is chosen to encompass structural and/or enclosure resonances. Distributed piezoceramic actuators and distributed polyvinylidene fluoride (PVDF) sensors are mounted on one of the enclosure boundaries. Acoustic sensors are used inside and outside the enclosure. An analytical formulation based on a state-space model is developed for local vibration control on an enclosure boundary and/or local noise control inside the enclosure for bandlimited disturbances. The error sensor signals are minimized by using digital feedforward control schemes based on filtered-U gradient descent algorithms. Analytical predictions from the current work are compared with the results obtained in our previous experimental investigations and discussed.

The present study addresses the optimization of the locations of the error sensors and the control actuators in an active noise control system. This system is implemented on a double- panel partition in order to improve its sound insulation characteristics in the low frequency range. A model of the active control system, allowing both structural and acoustical control, is developed and combined with a finite element model of the double-panel partition. This yields a discrete description of the actively controlled structure, resulting in a combinatorial optimization problem. As the optimal solution for this problem can only be found by enumeration, a genetic algorithm is generally used to generate a near-optimum solution, requiring however a lot of useless function evaluations. In order to overcome this severe drawback of the genetic algorithm a number of non-linear optimization methods are also tested. As these methods can only handle continuous problems, interpolating functions are used to approximate the dynamic behavior of the structure in between the finite element nodes. Finally the performances of the methods are compared in terms of quality of the obtained solution, computation time and algorithm complexity. This comparison shows that the best gradient method converges faster and yields a better final solution than the best genetic algorithm.

Finite-element piezoelectric modeling theory is merged with electronics design and implemented to study of elastic plate structures exhibiting control action via multiple piezoelectric attached devices. The theoretical results are validated utilizing FEA modeling of NASA Langley Acoustics Division Transmission Loss Facility circular elastic plate testbed. This testbed is modeled using advanced meshing capability and processing that includes a full electrical- mechanical modeling capability for piezoelectric elements. The model is validated using the simple test case of integral feedback, together with a fully coupled piezo-element feature. The detailed development of the multimodal piezo-coupled (continuum) equations of motion together with the FEA feedback coupled modeling for an elastic plate structure incorporating piezoelectric devices as sensors and actuators is presented.

The primary purpose of this paper is to present a conceptual design for an adaptive materials driven actuator, to be used for rotorcraft retreating blade stall control. The displacement of a single induced strain actuator is hydraulically amplified and the resulting output displacement is used to actuate a split-flap. Such an actuator is designed to produce high displacements through high amplification factors. The actuator design procedure is presented for a sample case. The required actuator force, displacement and amplification are computed. The effects of the flap deployment on the performance of the helicopter are not presented.

Discrete trailing edge servo-flap actuator designs for use in rotor control applications are considered. A survey was conducted comparing the pros and cons of a number of feasible actuator designs. The major conclusions from this survey indicate that any successful actuator design will utilize a high bandwidth active material, produce large amplification of the active material stroke, and incorporate a simple compressive pre-stress mechanism while remaining efficient in a mass normalized sense. The mass efficiency, defined as the ratio of the specific work performed to the specific energy available, was used as a metric to rate the actuators considered in this survey. Thus, unnecessarily heavy actuators are penalized, which is appropriate when designing components operating under high centrifugal forces. The most feasible discrete actuators are those where the active material reacts against an inert support frame housing. An upper bound on the mass efficiency of this type of actuator is shown to be a function of the ratio of active material to frame specific moduli. A new high efficiency discrete actuator called the x- frame actuator, developed at MIT, and designed in accordance with the lessons learned from the actuator survey, is described. A prototype of this actuator, 150% of model scale, was built and tested on the bench top to confirm the predicted performance. The prototype demonstrates an output energy density of 14.6 ft-lb/slug. It also has a bandwidth of 543 Hz upon driving a nearly impedance matched load. This performance is shown to correspond to a mass efficiency between 18 and 31%.

This paper presents the testing of a Froude scaled rotor featuring piezoceramic bimorph actuated trailing-edge flaps for active vibration suppression. An analytical model for the coupled bimorph actuator trailing-edge flap dynamic response in the rotating environment was developed. The analysis takes into account the aerodynamic, centrifugal, inertial and frictional loads acting on the actuator-flap system. The linkage arm length associated with the mechanical amplification mechanism was selected in order to maximize flap performance in the rotating environment. The bimorph clamping was improved to prevent actuator slippage under high centrifugal loads. The analytical model was validated by carrying out a series of bench tests, vacuum chamber tests and hover tests. In hover, flap deflections of plus or minus 6 degrees at 4/rev flap excitation were achieved at 900 rpm, thus demonstrating the potential of the piezoceramic bender as a lightweight and compact actuation system for individual blade control purposes. This paper also includes a feasibility study for piezo-bimorph actuation of a trailing-edge flap for a Mach scaled rotor model.

The theory, construction and experimentation of a shape memory alloy composites plate is presented and discussed. The modeling of shape memory effects due to the temperature- induced phase transformation is briefly reviewed. Constitutive relations that relate the stress, strain and temperature of the shape memory alloy are briefly presented. The general tension-bending theory of a composite plate incorporating shape memory alloy wires are derived. Substitution of the shape memory alloy constitutive relations and of the strain compatibility conditions yield a linear system in two variables, the midplane strain, (epsilon) ₀, and curvature, (kappa) . However, the coefficients of this linear system are functions of the stress and temperature in the shape memory alloy part of the active composite material. Hence, an iterative process is used to obtain the solution. Construction and experimentation of a 300 mm by 50 mm by 5 mm shape memory active composite plate is described. The composite plate was fabricated from polymeric resin and shape memory wires embedded in the top and bottom surfaces. The experimental set up, instrumentation, test procedure and data collection are presented. Processing of the experimental data, and the resulting displacement-temperature variation are shown and discussed. The shape memory active composite plate was shown to be capable of 35 mm tip displacement when activated.

The objective of this research is to develop a smart rotor with active control of blade twist using embedded piezoceramic elements as sensors and actuators to minimize rotor vibrations. A 1/8-th Froude-scale (dynamically-scaled) bearingless helicopter rotor model was built with banks of torsional actuators capable of manipulating blade twist at frequencies from 5 to 100 Hz. To assess the effectiveness of the torsional actuators and vibration suppression capabilities, systematic wind tunnel testing was conducted in the Glenn L. Martin Wind Tunnel. Using accelerometers embedded in the blade tip, the oscillatory blade twist response was measured. The changes in rotor vibratory loads due to piezo- induced twist were determined using a rotating hub balance located at the rotor hub. Experimental test results show that tip twist amplitudes on the order of 0.5 deg are attainable by the current actuator configurations in forward flight. Although these amplitudes were less than the target value (1 - 2 deg for complete vibration suppression control), test results show that partial vibration reduction is possible. Using open-loop phase shift control of blade twist at the first four rotor harmonics, changes in rotor thrust of up to 9% of the steady-state values were measured, resulting in up to 3 and 8% reductions in rotor pitching and rolling moments, respectively. It is expected that the hub load control authority of the smart rotor can be improved in future models with refined actuator configurations and implementation of closed-loop feedback controls.

The design principles, analytical models, construction methods and test results for a new type of solid state adaptive rotor (SSAR) are presented. A pair of directionally attached piezoelectric (DAP) torque-plates were fabricated and attached to the root of a 23.5' diameter helicopter rotor assembly. The DAP torque-plate tips were joined to a pair of graphite-epoxy servopaddles which were moved in pitch by the action of the torque-plates. The torque-plates were constructed from a single aluminum substrate and PZT-5H DAP elements mounted symmetrically at 45 degrees. Electrical signals were carried to the DAP torque-plates via a shielded brush and rotating contact assembly. A series of non-rotating static tests were conducted on the rotor, demonstrating servopaddle pitch deflections up to plus or minus 5.8 degrees and good correlation with classical laminated plate theory. Non rotating dynamic testing showed a system natural frequency in excess of 2.5/rev and good correlation with inertial models. Because the servopaddles were aeroelastically tailored to balance out propeller moments, deflection degradation with increasing rotor speed was barely noticeable up to plus or minus 1 degree pitch levels. However, as rotor speed increased, total servopaddle deflections in the rotating frame at 1600 rpm (full speed) were degraded, but still operated up to plus or minus 2.7 degrees in pitch. To conclude the study, the rotor was attached to a converted Kyosho Hyperfly electric helicopter. Flight tests demonstrated fundamental controllability. A system-level comparison showed that the SSAR Hyperfly experienced a 40% drop in flight control system weight, an 8% cut in total gross weight, a 26% decrease in parasite drag and a part count reduction from 94 components to 5.

Modeling of magnetostrictive Terfenol-D transducer performance requires reliable data on functional trends of the magnetostrictive element's material properties under various operating conditions. A statistical study was designed to experimentally evaluate material properties of 50 Terfenol-D samples under varied mechanical loads, varied ac drive levels, magnetic bias and mechanical prestress typical of transducer applications. This approach is based on a low-signal, linear- magnetomechanical model of the test transducer, and electroacoustics theory. Statistical analysis is provided for the material properties: Young's modulus at constant applied magnetic field, linear coupling coefficient (or axial strain coefficient), magnetomechanical coupling factor, and magnetic permeability at constant strain. Functional relations between material properties and ac drive levels and loads are developed, and corresponding confidence intervals are assessed. The trends show the sensitivity of Terfenol-D material properties to the operating variables and highlight the importance of properly understanding the effects of operating conditions on transducer performance.

This paper examines the dynamic response of electrostrictive rod actuators. A non-linear constitutive model for electrostrictors is used to obtain periodic solutions for the actuator's displacement and polarization. The method accounts for both inertial forces and a non-homogeneous stress distribution in the device. Results of the analysis predict displacement distortion and power requirements for the actuators as a function of excitation frequency. Resonance behavior of the actuator and the effect of electro-mechanical coupling are investigated. Voltage control parameters for the actuators are studied to determine the optimum control scheme for minimizing distortion of the displacement output.

A coupling coefficient is a measure of the effectiveness with which a shape-changing material (or a device employing such a material) converts the energy in an imposed signal to useful mechanical energy. There are different kinds of material and device coupling coefficients, corresponding to different modes of excitation and response. Device coupling coefficients are properties of the device and, although related to the material coupling coefficients, are generally different from them. It is commonly held that a device coupling coefficient cannot be greater than some corresponding coupling coefficient of the active material used in the device. A class of devices was recently identified in which the apparent coupling coefficient can, in principle, approach 1.0, corresponding to perfect electromechanical energy conversion. The key feature of this class of devices is the use of destabilizing mechanical pre- loads to counter inherent stiffness. The approach is illustrated for a piezoelectric bimorph device: theory predicts a smooth increase of the apparent coupling coefficient with pre-load, approaching 1.0 at the buckling load. An experiment verified the trend of increasing coupling with pre-load. This approach provides a way to simultaneously increase both displacement and force, distinguishing it from alternatives such as motion amplification, and may allow transducer designers to achieve substantial performance gains for some actuator and sensor devices.

The performance of magnetostrictive Terfenol-D is highly dependent on the state of the material and in particular on the mechanical prestress. This paper presents an experimental investigation of the effect of prestress on the dynamic performance of a Terfenol-D transducer. The effects of both prestress and magnetic bias on the near dc transducer performance are also presented. Experimental results demonstrate the sensitivity of the transducer performance in terms of strain, strain rate with applied field, and material properties to relatively small changes in initial mechanical prestress. Trends in material properties, Young's Modulus, magnetomechanical coupling factor, permeability, dynamic strain coefficient, and mechanical quality factor with prestress and drive level are developed. In addition, the effect of magnetic bias and frequency of operation on the strain at different applied fields are examined and shown to significantly influence transducer output at a given prestress level. For the transducer as operated in this study, including the appropriate magnetic bias, both the magnetomechanical coupling and the strain coefficient are optimized with a prestress of 1.0 to 1.25 ksi.

Self-sensing approaches for controlling piezoelectric materials thus far have been composed of slight variations on a theme; the theme being a linear bridge technique for extracting the self-sensing signal. Linear bridge-based self- sensing however has always been plagued with problems of impedance mismatching, drift, and instability, all due to the nonlinearities inherent in piezomaterials. Additionally, all past self-sensing research has been limited to strictly ac sensing applications, hence the abundance of vibration suppression papers. We propose a novel self-sensing technique which shows promise in reducing most if not all of these problems. The approach dubbed OBSS (observer based self- sensing) is akin to state estimation techniques in that is uses electric field and electric displacement information coupled with the linear piezoelectric constitutive relations to estimate the material's stress and strain states. Additionally, since most piezomaterials exhibit significant hysteresis, a real-time compensation method using the generalized Maxwell slip hysteresis model must be employed to minimize this nonlinear effect. Theoretical development and experimental validation of the OBSS technique is presented.

In this paper we present a linear motor according to the inch- worm-principle as an application for a hybrid actuator. A hybrid actuator consists of magnetostrictive and piezoelectric transducers oscillating in their electrical resonance. The reactive energy is exchanged between the different transducer types which have to be supplied externally just to compensate for the energy transferred into mechanical work and inner losses. So a higher efficiency compared to applications with new actuators realized up to now can be achieved. The motor consists of a fixed magnetostrictive rod for the feed-forward and two piezo-stacks for the clamping. Research revealed remarkable influence of the mechanical load and the non- linearities and hysteresis of the transducers on their electrical behavior. This led to a new switching drive allowing a variable operating frequency and a miniaturized design without magnetic bias. Implementing a model of the electro-mechanical behavior of the transducers, precision positioning under the step-width can be performed just by evaluating the applied voltage and current. Furthermore, information about the applied mechanical load can be achieved (smart actuator).

A two-sided piezoelectric traveling-wave rotary ultrasonic motor is being developed for a high-torque, low-speed application. In the symmetric two-sided configuration, the elliptical motion of teeth on both sides of the vibrating element is coupled to two respective interfaces, inherently increasing the motor's torque capability at reduced rotating speeds. Greater torque densities are attainable than with present configurations which operate with a single interface, enabling reduced size and mass and expanding the potential for direct-drive applications. The design of a two-sided motor prototype was initiated with a previously developed nonlinear physical dynamics model for the purpose of predicting, a priori, steady-state motor performance in a timely, automated manner. To assess the model's merit, the performance of a similar commercially available one-sided motor was simulated and compared to experimentally obtained measurements. Then, using the knowledge gained from correlating the two, a two- sided motor was designed to output significantly greater torque. A prototype was realized, and its speed-torque and efficiency-torque characteristics were obtained and compared to the predicted curves. Impressive stall torque and torque density have been achieved, yet lower than expected performance raises questions regarding the contact profile. Careful redesign of the passive interface washers to improve performance constitutes ongoing work, and improvements to the model are suggested which would alleviate present limitations and make it a more powerful design tool.

Piezomotors as driving components have been designed more or less successfully for more than 50 years. They are coming to the stage of commercial precision drive systems at last. At the Institute for Production Engineering and Machine Tools (IFW) in Hannover, Germany, the finite element method was used to simulate their vibration friction mechanism which is generating linear (or rotational) motion. The mechanical principle and numerical results are outlined in this paper. Since it is difficult to measure in the micrometer and microsecond range at the same time, the finite element model and its transient analysis is a support for further development and optimization in the field of ultrasonic motors. The model's mechanical and piezoelectric dynamic vibration behavior can now be observed and analyzed in detail. An animated motion analysis is available giving further insights into the drive system behavior.

A spacecraft-like optical interferometry system is investigated experimentally over several different optimized controlled structures configurations. Configurations represent common and not-so-common approaches to mitigating pathlength errors induced by disturbances of two different spectra. Results show that an optimized controlled structure for low frequency broadband disturbances achieves modest performance gains over a mass equivalent regular structure, while an optimized structure for high frequency narrow band disturbances is four times better in terms of rms pathlength. These results are predictable given the nature of the physical system and the optimization design variables. Fundamental limits on controlled performance are discussed based on average system transfer function gains and system phase loss.

The design of large, high performance structures has led to problems with structural control. A specific example of such a problem occurs in the simultaneous position and vibration control of active surface structures, such as large radio telescopes. This type of structure features a precision surface which has significant mass, but which does not contribute stiffness to its supporting backstructure. The surface is connected to the backstructure via a large number of positioning actuators, and these actuators serve to maintain the precise surface figure in the presence of disturbances to the structure. Previous work has addressed the problem of simultaneous positioning of the active surface elements and suppressing structural vibrations using independent modal space control. However, the previous work has been limited to surfaces which employ force actuators. While it is possible to implement control for such a structure, it has the disadvantage that the decoupling of the system requires knowledge of the support structure in absence of the active surface, which is difficult to obtain experimentally. This work investigates the problem of simultaneously controlling backstructure vibration and active surface positioning using displacement actuators. In this type of system it is shown that it is still possible to decouple the problem modally, allowing for a straightforward controller. Additionally, it is shown that this problem decouple using the model of the complete system, and is thus suitable for practical applications in which experimental measurement data is used to verify the theoretical model of the system. Because the controller design is based upon the eigenvalues and eigenvectors of the model, this allows for a more accurate controller design. A method for decoupling the system is presented, and the effects of the control on other modes is investigated.

In this computational study, we use thin piezoelectric films as sensors to detect vibration of circular cylindrical shells. Rather than using complicated sensor shapes for modal filtering, we use simple rectangular shapes and leave the measurement filtering to full order modal observers. Simulations indicate that decreasing the magnitude of the real part of the observer gain leads to slower convergence but to the desired mode. Larger values lead to faster convergence but to the actual sensor measurements, indicating that little signal filtering had time to occur. Increasing the number of sensors also improves observer performance provided that they are placed intelligently on the shell. If a sensor cannot adequately detect the mode which it was intended to sense, sensor output cancellation may affect the behavior of all modal observers. The contribution of different mode types within different frequency bands is also investigated, as well as the determination of an adequate number of vibrational modes to include in the simulations.

The development of an intelligent sensor/electronics/actuator system is presented in this paper. The system is autonomous since it only needs some approximate knowledge of the frequency range where each vibration mode may exist. Given this information, the device senses and identifies each of the principal vibration modes, tracks them (should they drift in frequency), and generate a counter-action which suppresses each of the vibration modes to a fraction of their corresponding uncompensated magnitudes. This control action is robust since it concentrates its energy only at those frequencies where periodic disturbances are detected. The main novelty of this system is the use of phase-lock-loops and switch capacitor bandpass filters for the on-line identification and tracking of the vibration modes, and the use of feedforward and feedback signals allowing to automatically adjust the optimal gain and phase compensations for each vibration mode. This approach has many advantages: it is practical and conceptually simple; it is modular and can be easily expanded to accommodate multiple vibration modes; it can be implemented with inexpensive low-power CMOS electronics; it is versatile since it does not rely on a model of the structure for developing its controller, and it only requires some approximate notion of the frequency ranges where periodic vibration is expected. The effectiveness of this new approach is experimentally evaluated using a test unit consisting of a simple structure, accelerometers and Terfenol- D actuators. The structure is excited by driving a shaker actuator with sinusoidal voltages. The resulting periodic vibration is measured using accelerometers. Feedforward and feedback acceleration signals are processed by the electronics producing an actuator command signal. The actuator generates a counter vibration that reduces the uncompensated vibration. Attenuation levels as high as 36 dB were obtained in an actual test unit.

This paper presents a novel active vibration control technique using embedded or bonded Piezo-electric actuators and multi- channel fiber-optics sensors. The main principle of the technique is the application of actuator pulses at the calculated temporal and spatial locations on the structures based on the on-line estimated (or off-lined calculated) model parameters. The applied pulses introduce active vibration suppression. The simulation and experimental results demonstrated that the developed time-delay technique has superior performance over other conventional control techniques (e.g. PD and impedance control) in terms of overshoot, elapsed time for vibration suppression, actuator saturation limits, real-time computational aspects.

The appearance of cracks in composite rotorcraft flexbeams can lead to degradation in flapwise and lagwise performance of the rotor blade. In addition, cracks in composite rotorcraft flexbeams under cyclic loading can result in rapid fatigue failure of these elements. A novel detection strategy is evaluated in this work which attempts to use the dereverberated response to locate damage in the form of a crack. Previous approaches have focused on modal response and damage detection methods. These techniques are incapable of detecting small defects or flaws in structures. However, a continuum mechanics description of the local structural dynamics provides greater resolution in the ability to detect small flaws. This modeling is employed here to determine the location and depth of a crack by examining the frequency domain properties of the scattering coefficient. The phase of this parameter contains properties that capture the size and location of the crack. A least squares optimization is performed to determine both crack depth and location. Analytical predictions are confirmed on an experimental [0/90]s graphite epoxy flexbeam of approximately 16 inches in length. Analytical and experimental results indicate that this approach is able to locate the crack and approximate its depth under centrifugal loading. The model predicts the location within 3% and approximates the depth to within 20%.

A method to characterize the energy dissipation in an overlapped bolted beam structure was derived in the first part of this paper. As a continuation of this work, a comparison of the results obtained by this methodology and the experiments used to corroborate them, are presented here. The gain of the structure (ratio of the response over the forcing function) is numerically calculated, and then experimentally verified for a high frequency content (up to 25 kHz). A linear model for the bolted joint is first assumed, and the resonant frequencies for the first twenty flexural modes are investigated. The energy flow across the jointed section is then obtained by mathematical manipulation of the power flux of the propagating waves at the connection. Thereafter, an extended nonlinear model, accounting for the nonlinear behavior of the joint, is studied in order to increase the accuracy of the method at a higher frequency range. Similar analysis of the wave interaction at the joint is also performed.

The purpose of this paper is to provide a concise introduction to the developments and recent findings of a BRITE-EURAM program of work (BRE2.CT94-0990 <SISCO>, structurally integrated system for the comprehensive evaluation of composites). The aim of the program has been to develop an acoustic/ultrasonic based structural monitoring system for composite structures using material compatible sensors. Since plate-like structures have been investigated, it has been a requirement to utilize the propagation of ultrasonic Lamb waves through the sample materials. Preliminary investigations utilized conventional piezo-electric sources coupled to the sample via perspex wedges. The Lamb waves generated by these sources were monitored using either a fully embedded or surface mounted optical fiber sensors. The system was tested with a variety of different carbon and glass fiber reinforced panels, and the interaction of the lamb waves with different defects in these materials was monitored. Conventional signal processing allowed the location of defects such as impact damage sites, delaminations and holes. Subsequent investigations have endeavored to refine the system. This paper reports the development of advanced wavelet based signal processing techniques to enhance defect visibility, the optical connectorization of composite panels, and the development of flexible low profile acoustic sources for efficient Lamb wave generation.

Induced-strain materials can produce very large forces and, hence, large energy density, but small actual displacements. A new concept for obtaining large-amplitude rotary displacements from small linear displacements generated by induced-strain material stacks is proposed. The concept utilizes the theory of twist-warping coupling in thin-wall open tubes. The theory of the proposed solid-state axial-to- rotary converter-amplifier, together with the appropriate bibliographical references, is given. A simple formula is generated for estimating the axial-to-rotary conversion- amplification coefficient from the geometrical length, L, and enclosed area, A, of the open tube. A large-displacement induced-strain rotary (LARIS) actuator proof-of-concept demonstrator was built and tested to verify and validate the theoretical developments. The LARIS actuator consisted of a 28 mm diameter, 1.2 m length open tube and a 120 micrometer, -1000 V PZT translator. The experimental set-up and the excitation and measuring equipment are fully described in the paper. A maximum rotary displacement of 8 degrees was measured, and the linear relationship between the rotation coefficient, the tube length, L, and the inverse of the enclosed area, A, was verified. An improved theoretical model, that accounts for the experimentally observed zero off-set, is also given. The theoretical developments and experimental tests presented in this paper show that the proposed LARIS actuator, based on a novel solid-state axial-to-rotary converter-amplifier utilizing the warping-torsion coupling of an open tube, is a viable design option, of great constructive simplicity and very low parts count. This concept can be successfully used in a series of aerospace and mechanical engineering applications, as for example in the actuation of adaptive control surfaces for aircraft wings and helicopter blades. The 8 degree rotary displacement capabilities measured on the proof-of-concept demonstrator can be easily scaled to other values to meet the operation requirements of specific applications. For very large angles (40 - 50 degrees), conventional electromechanical actuators (e.g., stepper-motors and ball-screws assemblies) can used.

Piezoelectric actuators have significant potential for use in smart systems like vibration suppression and acoustic noise canceling devices. In this work, a novel piezoelectric bending actuator CRESCENT was developed. CRESCENT is a stress-biased ceramic-metal composite actuator. The technology involves the use of the difference in thermal contraction between the ceramic and the metal plates bonded together at a high temperature by a polymeric agent to produce a stress-biased curved structure. An extensive experimental investigation of this device in the cantilever configuration was carried out. The tip displacement, blocking force and electrical admittance and were chosen to characterize the performance of the actuator under quasistatic conditions. The device fabricated at optimum temperature exhibits large tip displacement and blocking force and possesses superior electromechanical characteristics to conventional unimorph actuators.

A new class of bimorph actuators, called C-block actuators for their curved shape, have recently been proposed to provide improved performance characteristics over conventional straight bimorph actuators. Existing mathematical models of these actuators are based on classical curved beam theory which neglects transverse shear effects. The paper presents an improved mathematical model for C-block actuators of arbitrary thickness. The new model accounts for through-the-thickness transverse shear deformations using refined displacement fields. The results from this theory are compared to the existing classical model and experimental data. Both free and forced end conditions are investigated. An exact elasticity solution is also developed which provides a framework for comparison of the developed theory and other approximate theories.

There is an increasing need in the field of smart materials and structures for high-force, high-displacement actuators. C- blocks, created to meet this need, are curved piezoelectric laminated beams poled in the radial direction which flex when voltage is applied. C-blocks can be made from a number of different materials with any number of layers, both active and inactive. An important portion of the characterization of an individual C-block actuator is the identification of natural frequencies and the description of the amplitude response across the frequency spectrum. This paper introduces a simple model for the frequency-amplitude behavior of the C-block. The model is derived using Hamilton's principle to formulate the equations of motion for a general composite piezoelectric C- block, and solving the equations using the appropriate boundary conditions and piezoelectric forcing terms. To verify the model, piezoelectric polymeric C-block prototypes were fabricated and displacement amplitudes were experimentally determined across a range of frequencies. The response of the prototypes matched well the behavior predicted by the simple analytical model.

Two typical characteristics of direct piezoelectric actuators are displacements of ten micrometers and high stiffnesses. Recently, multilayer actuators have been improved, and they now display strains of approximately 1200 ppm at low excitation levels (less than two hundred volts). Thus, they are well suited to perform precise positioning of optical devices. But for industrial needs, this performance is still insufficient for positioning devices with larger displacements (in the range of several hundred micrometers). Numerous designs of mechanical amplifier devices based on the use of flexural hinges have been proposed. Due to their low stiffness, these devices cannot be used in space applications because they would not survive during takeoff. The amplified piezoelectric actuator which we designed and tested, eliminated the low stiffness drawback and ensures good force transmission. Due to the stiffness of the amplifier, the efficiency of the electromechanical transduction is significantly higher than those of conventional amplifier mechanisms. To design this actuator, we performed a numerical finite element simulation that included the piezoelectric effect. Among other things, this model shows the displacement as a function of the excitation and the electrical admittance. The static and the dynamic behaviors were determined. The main features of the actuator are a no-load displacement of 180 micrometers and stiffness of 5 N/m. These characteristics were experimentally verified using an electromechanical test bench including a laser Doppler interferometer, thus confirming the design method. Technological aspects, like the compressive force applied to the piezoelectric material, were considered. Many applications for this amplified actuator already exist. For example, an active mechanism using this actuator can be used to tilt a mirror. Another application of the amplified actuator is in the field of active damping of structures. In this case, the actuator is connected to a resistive shunt so that electrical damping is obtained through the direct piezoelectric effect. The experimental results show that the actuator is interesting because of its high electromechanical coupling, and, consequently, its ability to perform active damping.

This paper addresses vibration control of flexible structures using controllable ER (electro-rheological) dampers. A clamped-clamped flexible structure system supported by two short columns is considered. An ER damper which is operated in shear mode is designed on the basis of Bingham model of the ER fluid, and attached to the flexible structure system. The governing equation of motion and associated boundary conditions Are obtained from Hamilton's principle. A sliding mode control algorithm is then formulated to effectively suppress the vibration of the structure caused by external excitations. In the formulation of the controller, parameter variations such as frequency deviation are treated to take into account the robustness of control system. The effectiveness of the proposed control system is confirmed by both simulation and experimental results.

A high speed traversing mechanism which utilizes two electro- rheological clutches is described. The traversing mechanism can be used to wind filaments onto bobbins. The traverse speed is 5 m/s, the required turn round period is 10 milliseconds, the traverse length is 250 mm and the turn round position must be electronically controllable and repeatable within plus or minus 1 mm. These combined criteria of high speed and controllability makes the use of electro-rheological fluids an attractive proposition. The dynamic model used to predict the performance of the mechanism is outlined and theoretical performance predictions made for a variety of electro- rheological fluid characteristics. Supporting experimental data is used to illustrate the validity of the dynamic modeling. The effect of various fundamental electro- rheological fluid characteristics, such as electro-shear stress, time delays and viscosity are considered in relation to the requirements for the operation of the high speed mechanism. This study indicates important areas for consideration in the future development of electro-rheological fluids.

This paper provides dynamic modeling and shape control of a flexible plate containing an electro-rheological (ER) fluid between two elastic face layers. Firstly, a rotary oscillation test is conducted to obtain complex shear modulus of the employed starch-based ER fluid itself as a function of applied electric fields and excitation frequencies as well. A dynamic model of the plate structure is then developed by incorporating the measured modulus. Field-dependent vibration characteristics for a rectangular ER fluid-based plate with two simply-supported edges are evaluated through a finite element approach. Simulation results indicate that the proposed dynamic model can be employed to predict vibration characteristics of ER fluid-based structures and that the ER fluids imbedded in the distributed parameter-systems of plate- like structures have a great deal of authority in tuning vibration characteristics of natural frequencies, mode shapes, and damping properties.

This paper presents research into developing an adaptive nonlinear neural network control algorithm that can be used with smart structure actuators and sensors to control the vibrations of rotor blades. The dynamic equations of motion for a blade have the same form as a multilink manipulator (robot arm) and adaptive nonlinear control algorithms have proven successful in active control of these manipulators. The recent development of neural network control algorithms has provided the ability to adaptively learn in real time a set of parameters that will approximate external forces operating on the blades. The controller combines these two control techniques enabling the controller to adapt its parameters in response to changes in blade properties such as its mass or stiffness and to also learn the parameters necessary to account for the unknown but bounded, periodic disturbance forces such as those caused by the unsteady, periodic aerodynamic forces in the rotor system. Current efforts have been directed at testing the control algorithm on real beams with piezoceramic actuators and sensors. The initial test results have shown that vibration reduction and desired beam motion tracking can be achieved even under the influences of periodic disturbances.

Active aeroelastic panels have been proposed for transonic drag reduction and panel flutter control. These active panels use actuators to create deformation of the panel surface, but restrict the actuators to the inner panel surface. One such active panel configuration consists of long, thin actuator/stiffeners attached to the panel surface to create an active grid. The purpose of this study is to examine interactions between panel out-of-plane deformation, actuator geometry and placement, including how many are used. A Rayleigh-Ritz model is used to describe panel response to actuator input. The accuracy of this efficient model is demonstrated by comparison to more elaborate finite element analysis. The study includes static deflection and flutter investigations. One study conclusion is that gridwork actuator stiffeners can produce more panel deflection than a single actuator spread over the surface, but more material is needed. For small numbers of stiffeners, those with a rectangular cross section create more deflection than those with a square cross section. However, this trend does not hold for large numbers of stiffeners. Placing stiffeners in the airflow direction is the best configuration for increasing panel flutter dynamic pressure. Rectangular stiffener cross sections increase the dynamic pressure more than square stiffener cross sections.

This is the second part of a companion paper entitled 'Unified Approach of Active Panel Vibration and Flutter Control using Piezoelectric Actuation and Iterative Root Locus Method.' This paper reports the wind-tunnel test results of panel flutter suppression which is based on the methodology described in the companion paper. Composite panels with one-sided surface- mounted piezoelectric actuators and strain sensors are manufactured. Because the piezoelectric actuators are mounted on one side of the panels only, in plane forces and bending moments are induced simultaneously when the actuators are excited with input voltages. Strain and strain-rate signals picked up by strain gages and electronic differentiators are the system outputs for feedback. Three different configurations are tested and their results are compared, which are: one set of piezo-actuators at leading edge, one set of piezo-actuators at trailing edge, and two independent sets of piezo-actuators at leading and trailing edges respectively. The nonlinear finite element dynamic models for various configurations are introduced. While the experimental setup allows a systematic way of tuning the gains during the test, the theoretical control designs are introduced which comprises the iterative root locus method and the output feedback optimum algorithm. Reasonable details of test procedures are described and major findings of this research are in order. Firstly, strain level at leading edge is about 25% of that at trailing edge. Secondly, actuators at leading edge are more effective than those at trailing edge. Finally, flutter dynamic pressure can be raised by 20 to approximately 30% using active control.

Structural and aerodynamic models for a 10 degree half-angle, articulated, conical, barrel-launched adaptive munition (BLAM) are presented. The forward half of the test BLAM was gimballed and actuated by two orthogonal pairs of piezoceramic tendons. Laminated plate theory models showed that piezoceramic actuators could survive hard-launch conditions if they were properly precompressed by using an elevated temperature cure cycle and a high, favorable mismatch in coefficient of thermal expansion between the substrate and the piezoceramic. Aerodynamic models based on supersonic pressure distributions over cones were used to predict free-flight trim angles and normal force range. A series of bench tests on a 13.4 cm (5.28 in) long BLAM showed plus or minus 0.12 degree steady articulation angles and a first natural frequency of 198 Hz with close agreement between theory and experiment. A series of wind tunnel tests at Mach 3.26 demonstrated a steady deflection increase with field, producing untrimmed normal force coefficient changes of plus or minus 0.0019. Extrapolation of the data to a 105 mm (4.13 in) caliber PGU-28 shaped round fired at 610 m/s (2,000 ft/s) showed that effective range increases by more than a factor of 15 over the conventional PGU-28.

Conventional (i.e. passive) radar absorbing materials (RAM) have been in use now for over 50 years, but it is only with recent advances in conducting polymer composite materials that active RAM has become practicable. Techniques for utilizing these new materials in reconfigurable electromagnetic, or 'smart,' surfaces are reviewed, with due emphasis being given to the problem of system integration. The discussion is complemented by modelled and measured performance data on several smart surface configurations.

This paper analyzes various design considerations of a 'smart' rolling element bearing with self-diagnostic capabilities. The self-diagnostic capability was added to the bearing by integrating a condition monitoring module into the raceway. The advantages of this design over a conventional condition monitoring system are discussed. A structural analysis was done to evaluate the effect of various modifications to the bearing raceway. It was found that the raceway of a conventional bearing could accept a miniaturized sensor module without a significant impact to the bearing. A miniaturized acoustic emission sensor was designed to fit into the sensor module. The sensor was designed to respond to the bearing's characteristic frequencies.

A sensitivity based modal damage detection and characterization approach is applied to a rotating composite beam in vacuum. This damage detection methodology is analytically shown to be capable of detecting changes of both mass and stiffness in the rotating beam. The algorithm developed in this work is based on a finite element model of the rotating beam in the flap and lag directions. Elemental sensitivities are derived including an enhanced stiffness sensitivity arising from centrifugal loading of the composite beam. Damage in the structure is detected by performing a constrained minimization on the modal residuals of the structural response.

As control systems become integrated with structures in such applications as vibration suppression and structural shape modification, a damage tolerance/mitigation concern must be addressed. The fundamental issue of this concern is whether the controller is robust enough to remain stable after a step change in the structural parameters as a result of impact damage. A secondary issue is whether the controller can continue to perform its intended function (even if at some reduced level of effectiveness) following impact damage. Due to the complexity of the structural system such a controller must be lower-order than the structure that is controlled. This paper builds on previous research by Waters and Balas to demonstrate the effectiveness and limitations of a reduced- order direct model reference adaptive control (DMRAC) algorithm for damage mitigation. The reduced-order model matching conditions for implementing this algorithm and the theoretical stability results are presented. Also presented are simulation results demonstrating the effectiveness of the DMRAC algorithm in forcing a higher-order damaged plant to track a lower-ordered reference model. These simulation results look at matching a second-order plant to a first-order reference model, and a fourth-order plant to both a first- and second-order reference model. The results of these simulations indicate that a reduced-order DMRAC algorithm can, in some cases, effectively stabilize the system following severe damage to the structure, even if the system become inherently unstable. Additionally, the reduced-order DMRAC algorithm still allows for steady state tracking by the damaged smart structure of a reference input.

The detection of delamination damage is important in ensuring the safety of composite structures. In this paper, a novel approach for delamination detection is described. The new approach involves (1) the application of a moving load on the composite member, and (2) measurement of total extension in embedded optical fiber(s). A theoretical study indicates that with the new approach, a single loading test can provide sufficient information to deduce the delamination location (in both the length and thickness directions) as well as the delamination size. Multiple delaminations can also be easily detected if there is no significant interaction between them. While further analysis and experimental work are necessary, the feasibility and potential of the proposed delamination detection technique has been established.

Modal domain distributed fiber optic sensors by using a birefringent single mode fiber can be applied for measurement of mode coupling distribution along a fiber attached to a structure. The mode couplings are caused by strain, stress, temperature, etc. In the present paper, some fundamental experiments were conducted to understand relations between the mode coupling and the external force. The relations between mode couplings in x and y polarized modes and the stress distribution at the cross section of a fiber subjected to a transverse load were derived theoretically by using a coupled mode equation. The stress distribution was calculated by two dimensional finite element method (FEM). Experiments were conducted to measure the mode couplings of a coated or uncoated fiber subjected to a transverse load at one point. Experimental results showed relation between a mode coupling and an external load was almost linear on a logarithmic graph. It was found that an uncoated fiber measured load more quantitatively than a coated fiber. A fiber optic load sensor module for a serial multiplexed sensor system was developed for future application of the present study. The above load identification technique can be combined with the OCDR (optical coherence domain reflectometry) technique developed by K. Hotate et al. to quantify the applied loads along a fiber on structures.