Jiaju Zheng

Hysteresis compensation for giant magnetostrictive actuators using dynamic recurrent neural network

According to the hysteresis characteristics of the giant magnetostrictive actuator (MA), a dynamic recurrent neural network (DRNN) is constructed as the inverse hysteresis model of the MA, and an on-line hysteresis compensation control strategy combining the DRNN inverse compensator and a proportional derivative (PD) controller is used for precision position tracking of the MA. Simulation results validate the excellent performances of the proposed strategy

Modeling dynamic hysteresis for giant magnetostrictive actuator using hybrid genetic algorithm

This paper establishes a simple and novel dynamic hysteresis model for giant magnetostrictive actuator by considering the eddy current loss, anomalous loss and structural dynamic mechanical behavior of the actuator. To obtain parameters of the model, a hybrid genetic algorithm is proposed. Comparisons between the experimental and calculated results show the validity and practicability of the model

Dynamic Characteristics of Galfenol Cantilever Energy Harvester

Based on the Armstrong model and the established electromechanical model of a Galfenol cantilever energy harvester, a nonlinear dynamic model is proposed to describe the mechanical-magneto-electro coupled characteristics of the device. Comparisons between the experimental and calculated results show that the proposed model can provide a reasonable qualitative indication of data trends, and can predict the effects of the acceleration excitation, the load resistance, and the bias magnetic field on the output performance of the device, thus has very strong practicability.

An Equivalent Circuit Model and Energy Extraction Technique of a Magnetostrictive Energy Harvester

This paper proposes a distributed parameter equivalent circuit model of a magnetostrictive energy harvester, where the mechanical parameters and the mechanical-electro coupled terms are represented by electronic components. Based on the model, the harvester system with any energy extraction circuit (EEC) can be modeled in a circuit simulator for system-level evaluation and design. Comparisons between the measured and calculated results show that the proposed model can provide reasonable data of the harvester at low acceleration levels, and support the physical understanding of the dynamic behaviors for the harvester. In addition, the harvester system with a self-powered EEC is constructed, and its whole performances are effectively predicted. These prove that the proposed model can analyze and design the harvester system with both complicated mechanical structure and practical EEC, thus has very strong practicability.

A micro-position system of a giant magnetostrictive actuator

In this paper, a micro-position system is proposed and developed. This newly developed micro-position system contains a giant magnetostriction actuator (GMA), a digital control constant current source, a TMS320C31-based digital signal processor (DSP) position control board and a capacitive sensor. The digital control constant current source can be adjusted continuously from 0 A to 1.551 A and its resolution is up to 0.3788 mA. This micro-position system has good stability and high anti-noise ability, and can achieve 40 nm position resolution because of using the high precision capacitive sensor, the newly developed digital control constant current source and the TMS320C31-based DSP position control board.

Mechanical-magnetic-electric coupled behaviors for stress-driven Terfenol-D energy harvester

The stress-driven Terfernol-D energy harvester exhibits the nonlinear mechanical-magnetic-electric coupled (MMEC) behaviors and the eddy current effects. To analyze and design the device, it is necessary to establish an accurate model of the device. Based on the effective magnetic field expression, the constitutive equations with eddy currents and variable coefficients, and the dynamic equations, a nonlinear dynamic MMEC model for the device is founded. Comparisons between the measured and calculated results show that the model can describe the nonlinear coupled curves of magnetization versus stress and strain versus stress under different bias fields, and can provide the reasonable data trends of piezomagnetic coefficients, Young’s modulus and relative permeability for Terfenol-D. Moreover, the calculated power results show that the model can determine the optimal bias conditions, optimal resistance, suitable proof mass, suitable slices for the maximum energy extraction of the device under broad stress amplitude and broad frequency.The stress-driven Terfernol-D energy harvester exhibits the nonlinear mechanical-magnetic-electric coupled (MMEC) behaviors and the eddy current effects. To analyze and design the device, it is necessary to establish an accurate model of the device. Based on the effective magnetic field expression, the constitutive equations with eddy currents and variable coefficients, and the dynamic equations, a nonlinear dynamic MMEC model for the device is founded. Comparisons between the measured and calculated results show that the model can describe the nonlinear coupled curves of magnetization versus stress and strain versus stress under different bias fields, and can provide the reasonable data trends of piezomagnetic coefficients, Young’s modulus and relative permeability for Terfenol-D. Moreover, the calculated power results show that the model can determine the optimal bias conditions, optimal resistance, suitable proof mass, suitable slices for the maximum energy extraction of the device under broad stress a...

Dynamic nonlinear model with eddy current effect for stress-driven Galfenol energy

The energy harvester using magnetostrictive materials (e.g., Metglas, Terfenol-D and Galfenol) is receiving increased attention. However, prediction of performance for the magnetostrictive harvester is complicated because the device exhibits nonlinear magnetomechanical coupling characteristics and the stress-induced eddy current loss, which is the leading limiting factor of the device. Different models are presented, including the basic linear ones and the more accurate nonlinear ones. The linear models based on the linear magnetomechanical coupling equations of magnetostrictive materials can not describe the nonlinear effects of the pre-stress and bias magnetic field on the output voltage and power of the device. A FEM eddy current model combined a nonlinear magnetomechanical coupling model of Terfenol-D can better describe some nonlinear dynamic characteristics of the stress-driven Terfenol-D harvester. But this model can not provide the direct relationship among the input stress, the output voltage and power for the device. Moreover, the model also can not describe the nonlinear characteristics of the Galfenol harvester because the nonlinear behaviors of Galfenol and Terfenol-D are different. Galfenol has outstanding features, such as small saturated magnetic field, low brittleness and high tensile strength, thus is more suitable for harvesting the vibration energy than Terfenol-D and piezoceramics. In this paper, a dynamic nonlinear model with eddy current effect of a stress-driven Galfenol energy harvester is founded, and the effects of the varied operating conditions on the performance of the device are calculated and analyzed.

Dynamic Nonlinear Model With Eddy Current Effect for Stress-Driven Galfenol Energy Harvester

A dynamic nonlinear model with eddy current effect has been developed to predict the dynamic responses of the output current, the induced voltage, the electrical power, and the displacement of a Galfenol energy harvester subjected simultaneously to stress and bias magnetic field. This model is obtained by the constitutive equations with eddy current effect, the established electromechanical model and the Armstrong model. The calculated magnetic induction curves of a polycrystalline Galfenol are compared with the measured curves. The effects of the varied operating conditions on the performance of the Galfenol energy harvester are predicated and analyzed.

Modeling and analysis of Galfenol cantilever vibration energy harvester with nonlinear magnetic force

Galfenol traditional cantilever energy harvesters (TCEHs) have bigger electrical output only at resonance and exhibit nonlinear mechanical-magnetic-electric coupled (NMMEC) behaviors. To increase low-frequency broadband performances of a TCEH, an improved CEH (ICEH) with magnetic repulsive force is studied. Based on the magnetic dipole model, the nonlinear model of material, the Faraday law and the dynamic principle, a lumped parameter NMMEC model of the devices is established. Comparisons between the calculated and measured results show that the proposed model can provide reasonable data trends of TCEH under acceleration, bias field and different loads. Simulated results show that ICEH exhibits low-frequency resonant, hard spring and bistable behaviors, thus can harvest more low-frequency broadband vibration energy than TCEH, and can elicit snap-through and generate higher voltage even under weak noise. The proposed structure and model are useful for improving performances of the devices.

Dynamic hysteretic sensing model of bending-mode Galfenol transducer

A dynamic hysteretic sensing model has been developed to predict the dynamic responses of the magnetic induction, the stress, and the output voltage for a bending-mode Galfenol unimorph transducer subjected simultaneously to acceleration and bias magnetic field. This model is obtained by coupling the hysteretic Armstrong model and the structural dynamic model of the Galfenol unimorph beam. The structural dynamic model of the beam is founded based on the Euler-Bernouli beam theory, the nonlinear constitutive equations, and the Faraday law of electromagnetic induction. Comparisons between the calculated and measured results show the model can describe dynamic nonlinear voltage characteristics of the device, and can predict hysteretic behaviors between the magnetic induction and the stress. Moreover, the model can effectively analyze the effects of the bias magnetic field, the acceleration amplitude, and frequency on the root mean square voltage of the device.