Dynamic Modeling Of Magnetorheological Damper And Its Application In Building Shock Absorption

In recent years,the engineering application of semi-active control is gradually mature.The most representative one is the use of magnetorheological damper in civil infrastructure,which provides a more effective anti-seismic strategy for the protection of sensitive structures,and the reliability and robustness of semi-active control depend on the accuracy and precision of the structural parameters of the building.Therefore,aimed at the dynamic model of magnetorheological damper and the uncertainty in the housing shock absorption system,this pape uses the LFT(linear fractional transformation)to decouple the uncertain part of the system from the parameters of the system The H infinite robust controller is designed to make the nominal parameters and disturbance consistent considering the performance of the system.At the same time.in order to reduce the uncertainty,the inverse model of magnetorheological damper based on BP(Back Propagation)neural network is established and analyzed,and the accuracy of the inverse model and the effectiveness of the algorithm are verified.The main work is as follows:First,the forward modeling of magnetorheological dampers is carried out.Taking RD-8040-1 magnetorheological damper supplied by Lord as an experimental object,the characteristics of its model,especially some strong nonlinear factors such as hysteresis,are analyzed.Three forward models,including hyperbolic tangent model,extended hyperbolic tangent model and phenomenological model,are established respectively,and corresponding global optimization is solved by genetic algorithm.Then the three models are analyzed.Second,reverse dynamics modeling of MR damper is proposed.The inverse model is the control foundation of the damper.In order to find the most suitable inverse model in this system,three control methods for magnetorheological dampers,including the optimal control method,the simplified hyperbolic tangent model method and the intelligent control method of the BP neural network,are compared and analyzed.Finally,the intelligent control method of the BP neural network is chosen to be the best optimal among them,and its effectiveness is validated.Third,this paper design an H infinite robust controller based on LFT.The three layer steel structure of the magnetorheological damping system is taken as the control object,the uncertain parameters in the seismic wave and the three layer steel structure are set as disturbance,and the uncertain parameters in the three layer steel structure are separated by LFT linear fractional transformation.Then the H infinite robust controller is designed to suppress the interference and the seismic waves after the separation.It makes the external disturbance finally converge to a very small area near zero,and proves the stability of the controller at the same time.Finally,based on the H infinity control strategy of LFT,the damping effect of three stores steel structure is simulated through Matlab/Simulink.Comparing the outcomes of the non-controlled state and robust H infinite active control,the effectiveness of the algorithm is proved.At the same time,the effectiveness of semi-active control and active control are compared,and the effectiveness of the semi-active control is proved.Finally,the robustness analysis is carried out for the model uncertainty,and the control effect of the deterministic model is carried out.The innovation of this paper is to take into account the inaccuracy of the model and separate the uncertainty of the model through LFT,so that the effect of semi-active control is more accurate and robust.And the inverse model based on BP neural network is found.Through the inverse model based on BP neural network,the controller of the damper is more accurate.It not only exerts the continuous adjustable advantages of the magnetorheological damper,but also can accurately predict the control current.