The Vibration-reducing Control Algorithms And System Development Of Stay-cable Magnetorheological Damper

Harmful vibration caused by unknown external forces such as vehicle running and wind and rain has become an important issue affecting the safety of cable stayed bridges.As the main force-bearing member of a cable-stayed bridge,a cable is extremely susceptible to vibration under unknown loads.Therefore,reducing and suppressing the stay cable vibration and effectively controlling the cable vibration amplitude and time are of great practical significance for improving the stability and safety of the bridge structure.Stay cable vibration control is mainly focused on semi-active control of magnetorheological dampers.Scholars at home and abroad have made a series of research results on control algorithms,but there are still some problems with the stability,reliability and fast response of vibration damping controllers.In this paper,in order to suppress the vibration amplitude and time of the stay cable,and improve the stability and fast response of the vibration control system,the optimization design of the stay cable vibration control algorithm,the dynamic inverse modeling of the magnetorheological damper,and vibration reduction Conduct research on controller system development.Firstly,research on non-fragile vibration reduction control of stay cable-magnetorheological damper.Aiming at suppressing the amplitude of oscillation and shortening the convergence time,this paper proposes a non-fragile reduction under the constraints of multiple performance indicators for the external disturbance of the stay cable-magnetorheological damper coupling system and the parameters uncertainty of the vibration damping controller.Vibration control algorithm.Based on the linear matrix inequality(LMI)theory,the H∞performance index is used to suppress external disturbances,and the regional pole configuration is used to characterize the fastness and stability of vibration reduction control,and the variance is used to characterize small amplitude and vibration speed.The LMI toolbox in MATLAB is used to solve the convex optimization problem with multiple LMI constraints and linear objective functions,and a non-fragile vibration damping controller design form with multiple performance index constraints is given.The cable-stayed cables C22(long cable)and C13(short cable)of a cross-sea bridge in Zhejiang Province are taken as examples for simulation verification.The results show that the vibration damping controller designed by this method can reduce the amplitudes of the vibration states of different cables by 57.805% and 74.395%,and reduce the convergence time by 56.705% and 77.845%,respectively.Secondly,research on sliding mode non-fragile vibration reduction control of stay cable-magnetorheological damper.In order to improve the stability of the stayed cable-magnetorheological damper control system,based on the non-fragile vibration damping control algorithm,the performance guarantee index is used to ensure that the system is asymptotically s Tab with a certain upper bound index,and the non-fragile state is set.Combining feedback control and sliding mode control,a multi-index constraint sliding mode non-fragile vibration reduction control algorithm is proposed,and the cable stay cables C22(long cable)and C13(short cable)of a cross-sea bridge in Zhejiang Province are taken as examples.Perform simulation verification.The results show that the algorithm not only makes the vibration damping control system have a superior upper guarantee index J~*and a smaller anti-interference performance index γ,but also reduces the oscillation amplitude of different cables by up to 66.67% and the convergence time by up to 16.67%,which has good stability and vibration damping effect.Thirdly,research on the dynamic reverse modeling of magnetorheological damper under leakage conditions.Aiming at the working conditions of the damper magnetorheological fluid leakage and different vibration frequencies and currents,an inverse model of the magnetorheological damper dynamic model based on the RBF neural network is proposed.By constructing a magnetorheological damper test platform and combining the mechanical characteristics data under the working condition of leakage(the leakage amount accounts for about 3% of the total),the nonlinear damping hysteresis characteristics of the magnetorheological damper are analyzed.The BP neural network and RBF neural network were used to construct a dynamic inverse model of the magnetorheological damper.A training set and a test set were randomly generated from the sample set,and two running tests were performed.The error accuracy and fitting effect of the reverse model were compared and analyzed.The results show that under the test conditions of about 3%leakage of the magnetorheological fluid,ambient temperature of 20℃,sinusoidal excitation frequency of 1-2Hz,amplitude of 10mm,and current of 0-1A,the power of the magnetorheological damper established by the RBF neural network is used.Compared with the inverse model of the magnetorheological damper dynamic inverse model established by the BP neural network,the accuracy error is reduced by 73.44% on average,and it has good model fitting accuracy.Fourthly,based on the physical mechanism of vibration control of magneto-rheological damper,the software and hardware of the vibration control system are developed.In terms of hardware design,TMS320F28335 DSP chip with fast processing speed of complex signals is selected as the CPU,and the control board module is composed of extending JTAG communication interface,data acquisition and processing,control signal output and other functional components.Combined with the power control module,signal conversion module,signal amplification and conditioning module,the control output signal conversion and conditioning.In terms of software design,the simulation analysis of the vibration control algorithm was carried out,and the SIMULINK program of the semi-active control system was designed in combination with the dynamic reverse model of the magnetorheological damper under leakage condition.The C/C++language code was exported after the ring test(SIL),downloaded to the CCS5.5 software,and sent to the DSP chip through debugging tools.The response test of DSP vibration damping controller under load excitation was carried out on the vibration test platform of cable-stayed bridge in the laboratory.The results show that under the same magnitude of excitation,the acceleration of the MR damper under unresponsive conditions is reduced by 50%and the maximum displacement value is reduced by 73.33% compared with the undamped condition.The cable-magnetorheological damper is half The acceleration under active control conditions is reduced by a maximum of 86.67% and 75% compared to the undamped condition,the magneto-rheological damper is not responding to the conditions,and the maximum displacement value is reduced by 80% and 33.33%,respectively.The error of the vibration control algorithm is about 7.69%,which proves that it is effective and feasible to develop a vibration control system with DSP chip as the core.