Atsuhiko Senba

Modal Identification of Truss Structures by Changing Stiffness Using Piezoelectric Actuator

This study proposes a modal identification technique for truss structures by changing the stiffness using a piezoelectric actuator. The technique is formulated on the basis of the vibration characteristics of truss structures with a metal-sealed-type multilayer piezoelectric actuator installed in a specific truss member. The natural frequency of the truss is changed by the stiffness control of a truss member to identify the natural frequency of the truss structure; the axial stiffness of the truss member can be semi-actively controlled using piezoelectric materials and an electrical circuit. The frequency response amplitude is changed by the change of stiffness of the truss, then the change ratio of the frequency response amplitude can be used to identify the natural frequency of the truss, and then the excitation data are not required to be measured. The method is appropriate for the identification of space structures in orbit in the case when excitation data are not fully available because of unknown excitation forces and disturbances. Experimental demonstrations of the identification of the first bending modal frequency of a 10-bay truss with a piezoelectric actuator are presented. It is found that the estimated natural frequency is in good agreement with the exact frequency, even though excitation force data are not provided. Furthermore, the factors causing the identification error are discussed through theoretical sensitivity analyses. The results show that the identification error can be reduced when the frequency response amplitude of a specific excitation frequency derived from the natural frequency is used for the identification.

Variable Matrices Method for Self-identication of Adaptive Gyroscopic Structure Systems

Variable matrices method is applied for adaptive gyroscopic structure systems to investigate the capability to change their inertia and damping characteristics in this paper. The method realizes the concept of selfidentication that means adaptive structures identify their unknown structural properties by using dynamical responses to the variable and controllable parameters. A cantilevered beam model with an adaptive gyroscopic damper is demonstrated as a exible space structure in the numerical analyses. The identication algorithms based on the variable matrices method is derived to identify the bending stiness and the mass of unit length of the beam by using the variation of the moment of inertia and the variation of the angular momentum of the rotor. The numerical examples showed that the identication algorithms had no singular characteristics with the large error. The results also showed that the large angular momentum of the rotor aected the accuracy of the identication with proposed algorithms. It was shown that both the variable moment of inertia and the variable angular momentum could be used for the self-identication with the algorithms.

Transient Response Analysis with Extended Kalman Filter for Recursive Self-Identification Using Semi-Active Devices

A possibility of recursive self-identication method that combines the concept of selfidentication and the extended Kalman lter is investigated for adaptive structures with semi-active devices such as piezoelectric transducers. It is assumed that the stiness of the structure can be controlled semi-actively so that structures have two dieren t states, that is, high-stiness and low-stiness states. An extended state vector consisting of state vector, constant stiness, and constant damping coecien ts are estimated recursively, where the piezoelectric constant is semi-actively controlled so that the convergence rate is increased. Numerical experiments show that the dierence of the convergence of recursive estimation and its accuracy for high-stiness or low-stiness states. Also, the eect of the change of stiness during the estimation is investigated. As a result, changing the stiness can be used to improve the convergence rate and accuracy of the recursive estimation of unknown parameters. Furthermore, the eect of the control error of the semi-active device on the estimation accuracy and the convergence is discussed.

Frequency Domain Self-Identification with Variable Structural Parameters Using Impulse Response

This paper presents a frequency domain self-identification method with variable structural parameters reducing measurement noise. The variable angular momentum of the adaptive gyroscopic damper system is used as the variable structural parameter to identify the equivalent modal parameters of the mathematical model. The impulse response data of the system, which contains measurement random and sinusoidal noises, is measured to obtain the Fourier spectrum. To reduce the measurement noise, we propose a noise reduction approach by using the difference between the two Fourier spectrums, where each spectrum is corresponding to the different variable angular momentum. Because the Fourier spectrum corresponding to the measurement noise can be assumed to be constant, the difference between the two spectrums has no spectrum of the measurement noise. As a result, multiplying the difference between the Fourier spectrums to the two spectrums reduces the measurement noise of them, and the equivalent modal parameters are identified accordingly. Numerical analyses indicate that the measurement noise are effectively reduced by the difference between the two Fourier spectrums changed by the variable angular momentum, and the accurate identification of the modal parameters is realized.

Wind Tunnel Test of Morphing Flap Driven by Shape Memory Alloy Wires

Morphing wings that transform their configuration seamlessly are expected to improve maneuverability and reduce noise and energy consumption of airplanes compared to conventional wings deformation including gaps and kinks. In addition, solid actuators such as electromagnetic motors and shape memory alloys (SMAs) reduce weight and maintenance cost compared to conventional hydraulic actuators. In this paper, a morphing flap model with a 500mm span × 642mm chord is developed, one-third of the trailing part of which deforms elastically. The actuator system is comprised of thin SMA wires and air blowers to improve the response of the SMA wires. Wind tunnel tests of the morphing flap model are carried out to check its controllability and examine its performance under aerodynamic loads.

Memory effects performance of polyurethane shape memory polymer composites (SMPC) in the variation of fiber volume fractions

The utilization of chopped glass fibers as reinforcement in shape memory polymer composites (SMPC) has been introduced due to their functions and high specific mechanical properties. The mechanical properties of proposed combining polymer composite that consists of a polyurethane matrix and chopped glass fibers were determined while influences of fibers volume fraction (%) was evaluated where composites with volumetric amounts of fiber up to 30 % were fabricated. The cyclic thermomechanical test was carried out to determine the efficiency of shape fixation and shape recovery of the material. The data gathered was also used to examine the performance characteristics of the polymer composite material. The obtained results showed that the strength of the composite tends to increase with the amount of chopped glass fiber, which indicates effective stress transfer between the glass fiber and polyurethane polymer. The performance of SMPCs also depicted inverse effects with increase of fiber volume fraction. Hence, the percentages of shape fixation and shape recovery decreased by the increment of chopped glass fiber. When higher volume fiber fraction of 30 % was used, it showed the minimum value for almost all the fixations and recovery modes. It was observed that the effects of reinforcing polyurethane matrix with chopped glass fibers caused the SMPC to be more inflexible to deformation due to high stiffness.

Shape-retainment control using an antagonistic shape memory alloy system

Since shape memory alloy (SMA) actuators can generate large force per unit weight, they are expected as one of the next generation actuators for aircraft. To keep a position of conventional control surfaces or morphing wings with SMA actuators, the SMA actuators must keep being heated, and the heating energy is not small. To save the energy, a new control method proposed for piezoelectric actuators utilizing hysteresis in deformation [Ikeda and Takahashi, Proc. SPIE 8689 (2013), 86890C] is applied to an antagonistic SMA system. By using the control method any position can be an equilibrium point within hysteresis of stress-strain diagrams. To confirm a feasibility of the control method, a fundamental experiment is performed. The SMA wires are heated by applying electric current to the wires. When a pulsed current is applied to the two SMA wires alternately, the equilibrium position changes between two positions alternately, and when a series of pulse whose amplitude increases gradually is applied to one SMA wire, the equilibrium position changes like a staircase. However, just after the pulse the position returns slightly, that is, overshoot takes place. To investigate such a behavior of the system, numerical simulation is also performed. The one-dimensional phase transformation model [Ikeda, Proc. SPIE 5757 (2005), 344-352] is used for a constitutive model of the SMA wires. The simulated result agrees with the experiment qualitatively, including the overshoot. By examining volume fraction of each phase, it is found that the overshoot is caused by that austenite phase transforms into stress-induced martensite phase during the cooling process after the pulse.

Suppression method of overshoot on non/less-energy shape-retainment control utilizing hysteretic behavior of piezoelectric actuators

To keep a shape of a smart structure with piezoelectric actuators bonded on it, electric voltage must be applied continuously. To reduce the amount of electricity usage, a new control method was proposed and its feasibility was examined in the previous studies [Proc. of SPIE 8689 86890C, Proc. of 29th ISTS 2013-c-40]. In this method hysteretic behavior of piezoelectric actuators in strain-electric field relationship was utilized effectively, which behavior is that some amount of strain remains even at zero voltage once a large voltage is applied. The results showed that displacement of a smart beam with a piezoelectric ceramic actuator bonded remained without applying voltage to the actuators after applying a pulsed voltage. However, the displacement overshot a final position while applying the pulsed voltage. That is generally undesirable. In this paper a suppression method of this overshoot was proposed. To this end another piezoelectric actuator was bonded on the beam opposing the original actuator. The original actuator was a soft type while a hard type piezoelectric actuator was used as the opposing actuator. With help from the two types of actuators, the overshoot could be suppressed while applying the pulsed voltage by controlling the voltage for the opposing actuator adequately, and a desired displacement could be obtained at zero voltage after the pulse.

Self-Identification Experiments Using Variable Inertia Systems for Flexible Beam Structures

The concept of self-identification and its feasibility are experimentally investigated. The modal parameters changed multiple times by the control input of variable inertia parameter are used to obtain linear equations about unknown structural parameters to overcome the lack of modes. We derive the controllability of the modal parameters as the requested conditions for implementing self-identification using sensitivity analyses of the modal parameters with respect to the control input. Also, a criterion for the self-identification is proposed to measure the controllability. Then, the self-identification experiments are performed to examine the present method using a flexible cantilevered beam structure with variable inertia systems, which include controllable additional mass attached to the beam. As a result, the bending stiffness and mass per unit length of the test beam are accurately identified when the controllable and observable lower modes are appropriately excited by input force and their changes due to the variable inertia are accurately estimated from the combinations of a few strain gage sensors output and a cubic spline interpolation technique. The results also indicate that the identification accuracy of higher modes is affected by the accuracy of the estimated controllable mode shape, which is sensitive to the locations of sensors. As the proposed criterion is larger, the identification accuracy becomes more insensitive to the estimation error of the mode shapes.Copyright

Noise Separation Experiment for Flexible Structures Using Variable Mass

This paper presents a technique for separating unknown measurement noise from vibration data. The technique is based on the spectral subtraction using variable structural parameters of adaptive structures. The difference between two periodograms obtained from measured vibration data is calculated, where each periodogram is corresponding to the different value of the variable structural parameter, and resulting difference is normalized with its maximum value. Then, it is multiplied to each periodogram to reduce the unknown measurement noise included in vibration data, and thus present technique needs no prior knowledge of properties of noise. Simple numerical examples using a spring-mass system with a variable parameter are demonstrated. Also, experimental results are presented to show the feasibility of the proposed technique using the appendage mass attached to a flexible cantilevered beam structure. The lower modal frequencies of the test beam are shifted by the variation of the appendage mass, and the present technique is implemented for separating measurement noise including random and periodic noise. The results are also compared with that obtained by a conventional averaging technique such as Welch’s method, and consequently it is indicated that the present technique is suitable for extracting the system’s dominant mode in the presence of periodic noise or colored noise, which is difficult to be reduced by the averaging method.