Modeling Of Magnetorheological Damper And Fuzzy Control Of Vehicle Semi-active Suspensions

Suspension is an important part of vehicle and plays a key role in improving ride comfort and stability.In terms of vibration attenuation performance,semi-active suspensions are better than passive ones,and close to active ones.In addition,semi-active suspensions have broad application prospects due to small energy consumption,simple structure and low cost.Aiming at the strong nonlinearity,time-delay effect and uncertainty of magnetorheological damper(MRD),a hybrid MRD modeling method based on genetic algorithm(GA)and adaptive neuro-fuzzy inference system(ANFIS)is proposed in this thesis.In addition,considering the strong robustness and easy implementation of fuzzy control,a fuzzy controller is designed for the semi-active suspension system.On this basis,GA is used to optimize the fuzzy rules and membership functions simultaneously to improve the performance of the suspension system.The main work of this thesis is as follows:(1)A hybrid modeling method based on GA-ANFIS is proposed for the MRD.In this method,fuzzy reasoning and neural network learning ability of ANFIS are considered.Subtractive clustering technique is introduced to construct the rule base of ANFIS model.GA is used to optimize the clustering parameters to obtain the optimal number and structure of fuzzy rules for the MRD behavior's description.Simulation results show that the proposed method can effectively improve the forward and inverse modeling accuracy of MRD and better describe its hysteresis characteristics.(2)Considering the strong robustness and easy implementation of fuzzy control,a fuzzy controller is designed for the semi-active suspension,including the determination of input and output variables,the fuzzification of input and output variables and the selection of fuzzy control rules.Simulation results show that the fuzzy control method can improve the ride comfort of the vehicle and has strong robustness against uncertain vehicle driving conditions(such as changes in vehicle speed and sprung mass).(3)For the traditional fuzzy control,selection of fuzzy rules and membership functions is quite subjective.The GA is used to optimize fuzzy rule and membership function parameters,as well as to guarantee hard constraints of the semi-active suspension system.Simulation results verify that the GA-based fuzzy control method can effectively improve the ride comfort of the vehicle and make the suspension system meet the hard constraints.In addition,the proposed method also has good robustness against uncertain vehicle speed and sprung mass.