Dai-Hua Wang

Magnetorheological fluid dampers: a review of parametric modelling

Due to the inherent nonlinear nature of magnetorheological (MR) dampers, one of the challenging aspects for developing and utilizing these devices to achieve high performance is the development of models that can accurately describe their unique characteristics. In this review, the characteristics of MR dampers are summarized according to the measured responses under different conditions. On these bases, the considerations and methods of the parametric dynamic modelling for MR dampers are given and the state-of-the-art parametric dynamic modelling, identification and validation techniques for MR dampers are reviewed. In the past two decades, the models for MR dampers have been focused on how to improve the modelling accuracy. Although the force–displacement behaviour is well represented by most of the proposed dynamic models for MR dampers, no simple parametric models with high accuracy for MR dampers can be found. In addition, the parametric dynamic models for MR dampers with on-line updating ability and the inverse parametric models for MR dampers are scarcely explored. Moreover, whether one dynamic model for MR dampers can portray the force–displacement and force–velocity behaviour is not only determined by the dynamic model itself but also determined by the identification method.

A magnetorheological valve with both annular and radial fluid flow resistance gaps

In order to increase the efficiency of magnetorheological (MR) valves, Ai et al (2006) proposed an MR valve simultaneously possessing annular and radial fluid flow resistance channels with the assumption that the magnetic flux densities at the annular and radial fluid flow gaps are identical. In this paper, an MR valve simultaneously possessing annular and radial fluid flow resistance channels is designed, fabricated, modeled and tested. A model for the developed MR valve is produced and its performances are theoretically predicted based on the average magnetic flux densities in the annular and radial fluid flow gaps through finite element analysis. The theoretical results for the developed MR valve are compared with the experimental results. In addition, the performances of the developed MR valve are theoretically and experimentally compared with those of the MR valve with only annular fluid flow gaps. It has been shown that the theoretical results match well with the experimental results. Mainly attributed to the radial fluid flow gaps, the pressure drops across the MR valve with both annular and radial fluid flow gaps are larger than those across the MR valve with only annular fluid flow gaps for varying valve parameters. The radial fluid flow gaps in the MR valve can reach a higher efficiency and larger controllable range than those by annular fluid flow gaps to some extent.

Health Monitoring and Diagnosis for Flexible Structures with PVDF Piezoelectric Film Sensor Array

How to access the health situation of civil infrastructures is a brand-new challenge that scientists and civil engineers faced up to. Therefore, the method of health monitoring and diagnosis for flexible structures is a very active area of both academic and industrial research and development. For the large flexible structures, the mini cracks, which are the premise to induce the large cracks and damages in the structures, exert a little influence on the resonant frequencies of the structures. So the health monitoring method based on the changes of the resonant frequencies need that the cracks are so large that it can influence the resonant frequencies. However the mini cracks would change the vibration amplitude determined through monitoring the strain near the cracks, which can be used as the information sources realizing the health monitoring and diagnosis. It is a pity that the random external disturbances would influence the vibration amplitude, which can influence the effectiveness of the method that based on the vibration amplitude, so the key problem using the health monitoring and diagnosis method based on the vibration amplitude is how to avoid the influence of the random external disturbance. In this paper, a health monitoring and diagnosis method for flexible structures, which based on the relative outputs between sensors among PVDF piezoelectric film sensor array, is put forward. A Functional Link Neural Network (FLNN) is used as the damage modes classification unit. The experimental results show that the health monitoring method proposed in this paper is effective for diagnosing the damage and its severity, although the damage modes are not too complicated. Because the FLNN realizes the classification through enhancing the input patterns, the architecture of the neural network used in this paper is simple, the learning algorithm is easy, and the learning speed is fast. During the training and validation of the FLNN, its input patterns are formed by the relative outputs between the PVDF piezoelectric film sensors, which are affixed on the surface of flexible structures and formed into the sensor array. So the external exciting amplitude is not needed to be fixed or to be confined to a fixed variation range for the health diagnosis, which is validated by the experiment on the flexible beams and guarantees the method proposed in this paper to be suitable for a general usage.

An integrated relative displacement self-sensing magnetorheological damper: prototyping and testing

In this paper, an integrated relative displacement self-sensing magnetorheological damper (IRDSMRD) and the corresponding electronic system to realize the integrated relative displacement sensing and controllable damping, including the relative displacement modulator/demodulator, circuit for superposing the carrier signal for the integrated relative displacement sensor (IRDS) on the exciting current from the controllable current driver for the controllable damping and controllable current driver are developed and tested. In the developed IRDSMRD, the exciting coil is energized by the current from the controllable current driver, on which the carrier signal for the IRDS is superposed by the superposition circuit. The amplitude modulation of the carrier signal for the IRDS by the relative displacement between the piston and cylinder of the IRDSMRD and the magnetization of the MR fluid are realized through the frequency division multiplexing of the exciting coil for both the IRDS and the MR damper and the relative displacement is accessed by demodulating the induced harmonic voltage from the induction coil of the IRDSMRD by the demodulator. The characteristics of the developed IRDSMRD, including the linearity, sensitivity and hysteresis error of the IRDS and the controllable damping force are tested on the established experimental setup based on the MTS 849 shock absorber test system and the real time simulation system. The testing results indicate that the developed IRDSMRD can not only achieve the integration of the relative displacement sensing capability but also possesses good performance of the relative displacement sensing of the IRDS and the large controllable damping force range. In addition, the performance of the IRDS will not be affected by the exciting current within a certain range and the damping force will not be degraded by the carrier signal for the IRDS. The realized principle and technology of the IRDSMRD lay a foundation for reducing the commercializing cost of MR dampers.

A magnetorheological damper with an integrated self-powered displacement sensor

In this paper, aiming at self-powering the integrated relative displacement sensor (IRDS) and the corresponding electronic system of an integrated relative displacement self-sensing magnetorheological (MR) damper (IRDSMRD) based semi-active system, the principle of an MR damper with an integrated self-powered displacement sensor is proposed and realized. The prototype of the MR damper with an integrated self-powered displacement sensor is designed and fabricated. In this MR damper, a coil evenly wound on the piston simultaneously acts as the exciting coil for the MR fluid and the IRDS, while a coil evenly wound on the cylinder simultaneously acts as the induction coil (i.e., pick-up coil) for the IRDS and the pick-up coil for the energy harvesting device. On one hand, both the MR fluid and the IRDS are simultaneously magnetized by a mixed signal, in which the carrier signal for the IRDS and the current for the MR fluid with different frequencies are superposed by a superposition circuit. That is, the exciting coil is frequency division multiplexed. On the other hand, when the exciting coil of the MR damper is energized by the carrier signal for the IRDS and the current for the MR fluid, the induced voltage in the pick-up coil not only can be harvested by the energy harvesting circuit to power the IRDS and the corresponding electronic system of the IRDSMRD, but also can be demodulated to obtain the relative displacement of the piston relative to the cylinder. That is, the induction coil for the IRDS and the pick-up coil for the energy harvesting device are functionally multiplexed. The characteristics of the fabricated MR damper with an integrated self-powered displacement sensor, including the energy harvested by the pick-up coil, the relative displacement sensed by the IRDS, and the controllable damping force, are modeled, analyzed, and tested. The feasibility and capability of the proposed principle are validated theoretically and experimentally.

A phenomenological model for pre-stressed piezoelectric ceramic stack actuators

In order to characterize the hysteretic characteristics between the output displacement and applied voltage of pre-stressed piezoelectric ceramic stack actuators (PCSAs), this paper considers that a linear force and a hysteretic force will be generated by a linear extension and a hysteretic extension, respectively, due to the applied voltage to a pre-stressed PCSA and the total force will result in the forced vibration of the single-degree-of-freedom (DOF) system composed of the mass of the pre-stressed PCSA and the equivalent spring and damper of the pre-stressed mechanism, which lets the PCSA be pre-stressed to endure enough tension. On this basis, the phenomenological model to characterize the hysteretic behavior of the pre-stressed PCSA is put forward by using the Bouc–Wen hysteresis operator to model the hysteretic extension. The parameter identification method in a least-squares sense is established by identifying the parameters for the linear and hysteretic components separately with the step and periodic responses of the pre-stressed PCSA, respectively. The performance of the proposed phenomenological model with the corresponding parameter identification method is experimentally verified by the established experimental set-up. The research results show that the phenomenological model for the pre-stressed PCSA with the corresponding parameter identification method can accurately portray the hysteretic characteristics of the pre-stressed PCSA. In addition, the phenomenological model for PCSAs can be deduced from the phenomenological model for pre-stressed PCSAs by removing the terms related to the pre-stressed mechanisms.

Linearization of Stack Piezoelectric Ceramic Actuators Based on Bouc—Wen Model

In this article, in order to linearize the hysteresis behavior of stack piezoelectric ceramic actuators (SPCAs), the feedforward linearization method based on the Bouc-Wen model and the hybrid linearization method combining the feedforward method and PI feedback loop are proposed and explored. The rapid control prototypes of the linearization controllers for the SPCA are established and tested. The research results show that both the feedforward and hybrid linearization methods can linearize the hysteresis behavior and the SPCAs with linearization controllers can be regarded as linear actuators. However, the linearization accuracy using the feedforward method is confined by the modeling error of the Bouc-Wen model. Although the proposed hybrid method combining the feedforward method and the PI feedback loop can reach higher linearization accuracy than that with the feedforward method, the hybrid method will result in high frequency components in the additional voltage, which will have a serious impact on the lifespan of the controlled SPCA. Utilizing the linearization methods proposed in this article, the open-loop and closed-loop controls for the tip displacement of a piezoelectric-driven microgripper are realized, which indicates that the proposed linearization methods can simplify the control of SPCAs and SPCA-based systems with high accuracy.

Pareto Optimization-Based Tradeoff Between the Damping Force and the Sensed Relative Displacement of a Self-sensing Magnetorheological Damper

In order to make the best compromise between the damping force and the linearity of the integrated relative displacement sensor (IRDS) of an integrated relative displacement self-sensing magnetorheological damper (IRDSMRD), a Pareto optimization-based method, which optimizes the key structural parameters by taking the damping force and the linearity of the IRDS of the IRDSMRD as the objective functions, is proposed and realized in this article. The Pareto front, representing the best tradeoff between the damping force and the linearity of the IRDS of the IRDSMRD, is obtained by considering that the maximum magnetomotive force applied to the exciting coil of the IRDSMRD is constant. Three IRDSMRDs with different piston modules, which are determined according to the Pareto optimal solutions, are developed and tested. The research results indicate that every point on the Pareto front is an optimal solution that compromises the damping force and the linearity of the IRDS of the IRDSMRD and the optimal damping force of the IRDSMRD with a certain linearity of the IRDS as well as the optimal linearity of the IRDS with a certain damping force can be determined according to the Pareto front for a specific application.

A High-voltage and High-power Amplifier for Driving Piezoelectric Stack Actuators

Aiming at the strong capacitive impedance of piezoelectric stack actuators, the principle to improve the dynamic performance of piezoelectric stack actuators through increasing the peak values of the output current and output power of power amplifiers are explored. Based on the error-amplified principle, the method that enlarges the output voltage of the dynamic power amplifiers through using a high-voltage operational amplifier in series with the power booster section, as well as the method that improves the peak values of the output current and output power through paralleling multiple power booster units utilizing the class AB quasi-complementary symmetry power amplifier circuits, are proposed and analyzed. Utilizing the proposed principle and method, a high-voltage and high-power amplifier for driving piezoelectric stack actuators is developed, simulated, and tested. The research results indicate that the developed power amplifier not only can break through the limit that each single power booster unit can only achieve the power output <125 W, but also can disperse the current and power averagely among the power booster units in parallel, which are beneficial to realizing the high power output with good static and dynamic performance and enhancing the reliability of the power amplifier.

Modeling and Testing of the Static Deflections of Circular Piezoelectric Unimorph Actuators

In this article, based on the classical laminated plate theory, a new static deflection model for CPUAs subjected to voltage is established and the bonding layer is taken into account as an individual layer. According to the established analytical model, the influences of the structural parameters and material properties of the CPUA on the transverse deflection are numerically simulated and the static and dynamic characteristics of the CPUA are experimentally tested. The research results show that the predicted static deflections of the CPUA by the established deflection model in this article agree well with the measured results, the established static deflection model considering the bonding layer is more accurate than the existing model neglecting the bonding layer, and the maximum relative error is reduced by 8.45%. The static deflections of the CPUA are apparently affected by the structural parameters and the material properties, which indicate that the performance of the CPUA can be optimized by the structural parameters and material properties. Because the hysteresis of the piezoelectric material is not considered when establishing the static deflection model, the apparent error exists when utilizing the static deflection model to predict the dynamic characteristics of the CPUA.