Research On Adaptive Neural Network Control Of Active Vibration Isolator Of Magneto-variable Elastomer

Magnetorheological material mainly includes magnetorheological fluid(MRF)and magnetorheological elastomer(MRE),whose stiffness and damping can be adjusted by the varied magnetic field strength in a certain range,and the property of the adjustable stiffness and damping can be applied to the semi-active control system,which has wider working frequencies range and the superior ability of vibration reduction than passive control systems,and lower cost on the fabrication and energy than active control systems.MRE shows a less significant damping shifting and a more apparent stiffness shifting than that of MRF under the varied applied magnetics density.Thus,in the application of isolator used for resonance elimination,MRE is more prospective than MRF.Currently,the MRE-based semi-active isolators have been successfully applied in constructions,bridges,vehicles and other engineering projects,while the research of MRE-based isolator is required to further explore,especially for the control strategy development of the semi-active isolator.The presented paper focuses on the control development of the MRE based semiactive isolator,aiming to propose a novel adaptive neural network control scheme to improve the control performance of isolator in multiple working conditions.The analysis and validation of the proposed control scheme were then carried out with the dynamics modeling.The main research works are summarized as the following:(1)The rotary magneto-rheometer was applied to conduct the test and characterization of MRE samples.Subsequently,based on the experimental data,the function of applied magnetics density with respect to MRE's stiffness and damping was obtained.(2)The dynamics modeling of the micro-vibrator was established.Subsequently,an adaptive neural network control scheme was proposed to improve the control performance of the isolator in the presence of multiple working conditions.Eventually,the comparison of the micro-vibration system using various control scheme was performed,and the results show that the proposed control can better mitigate the vibration transmission from the excitation than ON-OFF control and the passive control using different levels of constant magnetic density.Meanwhile,Lyapunov function method was adopted to approve the stability and convergence of the proposed control law.(3)The dynamics modeling of the vehicle seat suspension system equipped with the MRE isolator was established.Subsequently,the effectiveness of the proposed control scheme was verified by the comparison of the vehicle seat suspension system using various control scheme.