Research On Model Free Adaptive Control Method For Automotive Active Suspension System

The suspension system is a key component for improving vehicle ride comfort and handling stability.Compared to passive suspension and semi-active suspension,the active suspension system of automobiles has more obvious advantages in improving vehicle driving performance,and has become the main trend in the development of controllable suspension.The real vehicle suspension system is a high degree of freedom nonlinear system excited by random road surface,but the current research on active suspension mostly focuses on the two degree of freedom single wheel linear model,and among the many control algorithms proposed,there are few control methods suitable for high degree of freedom suspension,so it is difficult to ensure the vehicle to obtain good driving performance when dealing with the complex and changeable driving conditions of the vehicle.Therefore,this paper aims to explore effective and feasible control strategies for active suspension systems,taking into account the nonlinear characteristics of suspension springs,and establish a four degree of freedom semi vehicle active suspension system model and a seven degree of freedom vehicle active suspension system model that are closer to actual dynamics.At the same time,considering the complexity of the mathematical model and the difficulties in controller design,by utilizing the advantages of Model Free Adaptive Control(MFAC)theory,which does not rely on system mathematical models,the following three aspects of research work are emphasized:(1)Research on the Improved Model Free Adaptive Control(IMFAC)method based on the semi vehicle active suspension system.The method takes the vertical vibration acceleration and pitch angular acceleration of the vehicle body as feedback,zero acceleration as control expectation,and the actuating force of the front and rear suspension as controller output.Starting from the compact format dynamic linearization model of multi input multi output,a weighted improved control criterion function is introduced to derive the mathematical expression of the controller,and the stability of the multi input multi output IMFAC algorithm is theoretically analyzed.The control simulation results of different road surfaces and vehicle speeds show that this method can further reduce the body vibration acceleration,suspension dynamic travel,tire deformation,and body pitch angular acceleration compared with the traditional MFAC.(2)Research on Model Free Adaptive Forecasting Algorithm(MFAPC)method based on semi vehicle active suspension system.Considering the advantages of fast response speed and good control effect of predictive control,a method combining predictive control with MFAC is proposed.By predicting the output of the model(body vibration acceleration,body pitch angular acceleration)to optimize the input of the model(front and rear suspension actuation forces),the characteristics of the MFAC algorithm can make this prediction method do not need the accurate mathematical model of the system.Secondly,a theoretical analysis was conducted on the stability of the multi input multi output MFAPC algorithm.The simulation comparison results of different road surfaces and vehicle speeds show that MFAPC has better and more comprehensive control effects than traditional MFAC in semi vehicle active suspension,greatly improving the driving smoothness and handling stability of the vehicle.(3)Apply the above three control methods to the research of the vehicle active suspension system.According to the force characteristics of the active suspension system of the whole vehicle,the body vibration acceleration,pitch angular acceleration and roll angle acceleration are taken as the controller input,and the force or torque in the vertical,pitch and roll directions of the body are taken as the controller output.The four suspension subsystem forces are obtained through the pseudo inverse operation link of the body equation,and then the control of the active suspension system of the whole vehicle is completed.The simulation comparison results based on the vehicle active suspension control system show that the above three control methods can be effectively used to control the vehicle active suspension system,and MFAPC has a more significant control effect on improving various indicators of vehicle suspension compared to MFAC and IMFAC.The research results demonstrate that each model free adaptive algorithm has good dynamic decoupling and vibration reduction control capabilities in high degree of freedom suspension,providing useful control theory references for controllable suspension systems.