Study On Time Delay Ideal Predictive Compensation Based Optimal Control Strategy For Mr Semi-active Suspension

Compared with the passive suspension system’s inability to adjust the vibration damping performance parameters and the active suspension system’s complex structure,high cost and high energy consumption,the magnetorheological vibration damping based on the magnetorheological fluid(MRF)technology Magneto Rheological Damper(MRD)provides an effective solution for the improvement of vehicle riding comfort and operation stability due to its simple structure,low energy consumption,and wide adjustable range of damping.However,as the core component of the Magneto Rheological semi-active suspension system,MRD will inevitably have a non-negligible response due to the measurement of the system state,the calculation of the control force and the dynamic response of the damping force during the controlled process.This time delay phenomenon which has a great impact on the controlled accuracy of the suspension and system performance,and even leads to the instability of the suspension system in severe case.In order to effectively reduce the influence of time delay,this paper proposes a time delay ideal predictive compensation based optimal control strategy for Magneto Rheological semi-active suspension,and designs a time delay based on e~x function definition expansion-LQG(EFE-LQG)ideal prediction compensation optimal controller.The main work done in this paper is as follows:First,a test plan for the mechanical properties of the MRD using the INSTRON-8800 single-channel hydraulic servo vibration bench was designed.By analyzing the test results of the mechanical properties,an asymptotically saturated magic formula model with high fitting accuracy was selected.By fitting the model parameters of the magnetorheological damper progressive saturation magic formula,an accurate model of the mechanical characteristics of the MRD is obtained,which is used to establish a MR semi-active suspension model with a quarter of the vehicle two-degree-of-freedom.The ideal prediction compensation optimal control of the system model and the design of the MR semi-active suspension system for the time delay has laid a theoretical foundation.Secondly,a test method of response time delay of MRD is proposed.A buck circuit integrated driving current source composed of two symmetrical n-channel and p-channel MOS tubes is used to systematically study the effects of excitation current and excitation frequency on the rise and fall response time of damping force of MRD.Through the analysis and calculation of the test data,the parameters are fitted and the accurate model of the response time delay of MRD is established,which provides a reliable experimental basis for the subsequent design of the optimal control strategy of ideal predictive compensation for time delay of Magneto Rheological semi-active suspension.Secondly,a time-delay ideal predictive compensation optimal control strategy based on e~x function definition formula and equivalent substitution(EFE-LQG)is proposed.Firstly,a controllable predictive equation for accurately predicting the ideal control at the next time delay is constructed,so that the quadratic optimal control force of the time-delay system at the current time is equal to the ideal control force of the time-delay system without time delay;In order to meet the design conditions of the quadratic optimal LQG controller,the part of the square of the time-delay control force in the objective function is equivalent to the part of the square of the predictive control force,and the time-delay compensation effect of the Magneto Rheological semi-active suspension system and the tracking accuracy of the predictive control force are improved at the same time.Furthermore,Matlab numerical simulation models of passive suspension,ideal Magneto Rheological semi-active suspension and Magneto Rheological semi-active suspension based on EFE-LQG control are constructed to verify the proposed optimal control strategy of time delay ideal prediction compensation.Numerical simulation was carried out under different time-delay conditions,the body acceleration,suspension dynamic deflection,tire dynamic deformation and the comprehensive performance indicators of the suspension which affected the suspension performance were numerically compared and analyzed in the frequency domain and time domain.Among them,considering the practicability of the optimal control strategy of time-delay ideal prediction compensation,a solution method to define the number of expansion segments for the prediction equation of this strategy is proposed.The analysis of numerical simulation test data shows that in the time domain and frequency domain,the performance indicators of the magnetorheological semi-active suspension based on EFE-LQG control are very close to the ideal semi-active suspension.The overall performance index of the frame is greatly reduced.Finally,in order to verify the control effect of the EFE-LQG time-delay ideal prediction and compensation optimal controller in closer to the actual working conditions,a discretized EFE-LQG controller model and a MR semi-active suspension system model are built.Two YX-SPACE rapid prototyping controllers(Rapid Control Prototyping,RCP)are used to design the controller hardware-in-the-loop test scheme.Through the comparison and analysis with the numerical simulation test data in the time domain,it is shown that the MR semi-active suspension proposed in this paper is optimal control strategy of time-delay ideal prediction compensation of the frame has good validity and practicality.The research shows that the time delay ideal predictive compensation optimal control strategy based on the definition formula of e~x function and equivalent replacement(EFE-LQG)proposed in this paper has the advantages of effectiveness and applicability.Under the condition of meeting the number of expansion segments of the definition formula required by accuracy,the comprehensive performance index of MR semi-active suspension based on EFE-LQG control is significantly lower than that of passive suspension.The ideal semi-active suspension can meet the preset error requirement of1%;Under the condition of 7.2ms time delay,compared with the passive suspension,the root mean square error of body acceleration is reduced by 37.01%,and the ride comfort is greatly improved.The comparative analysis of the experimental results of the controller hardware in the loop and numerical simulation also shows that the variance error of the control force is only 2.67%under the condition of 30ms time delay,which verifies the reliability and validity of the strategy.