Research On Active Suspension Control Strategy Based On Swarm Intelligence Algorithm Optimization

As a key component that affects the safety and comfort of the vehicle,the suspension plays a pivotal role in the car chassis.Passive suspension due to the immutability of its parameters,in the play of suspension performance is too limited,active suspension because of its unique actuator that can generate active force,can achieve the suspension performance with the road conditions adaptive adjustment,greatly improving the car driving smoothness and handling stability.The focus of research on active suspension systems is often on its controller,and the development of control algorithms is particularly important.In this paper,this paper aims at the simulation experiment of control strategy,a key link in the development process of active suspension,taking 1/4 suspension as the research object,based on PID control,Fuzzy control and neural network control,integrating new intelligent algorithms,and using MATLAB software as the simulation platform for research.This article mainly has the following work:Firstly,according to the common models of vehicle dynamics research,the selected model of this paper is determined,and a simulation model is built in the Simulink environment according to the 1/4 active and passive suspension dynamic equations;the pavement excitation is simulated on the theoretical basis of the stochastic pavement analysis method,the conversion of time spectrum and spatial spectrum,and the impact pavement modeling method is used as the theoretical basis,and the output response of C-class pavement and impact pavement in the time domain is obtained;and the comprehensive evaluation index of suspension system performance is given,laying the foundation for subsequent simulation work.Secondly,a 1/4 active suspension PID controller is designed according to PID control theory,and based on the standard artificial swarm algorithm（ABC）,an improved ABC algorithm is proposed based on the search method and selection strategy,and applied to the SELF tuning of PID parameters.Under the conditions of the same level of pavement,the simulation and comparative analysis results show that the improved ABC algorithm adjusts the PID parameters faster and the solution accuracy is higher,and the 1/4 active suspension body vertical acceleration,suspension dynamic stroke and tire dynamic displacement based on the improved ABC-PID control are optimized by 15.1%,26.3% and 15.4% respectively before the improvement,and the effect is better in the active suspension PID control.Then,based on The Fuzzy control theory,a 1/4 active suspension fuzzy PID controller is designed,based on the standard instructional optimization algorithm（TLBO）,based on the adaptive teaching factor,the self-learning process of "variation" operation,and K-means clustering,a TLBO improvement method is proposed,and the parameters of the membership function on the fuzzy rule are designed as optimization variables,and a 1/4 active suspension Fuzzy-PID controller optimized based on the improved K-means TLBO algorithm is designed.Under the conditions of the same level of pavement,the simulation and comparative analysis results show that compared with the four control algorithms of traditional PID control,fuzzy PID control and Fuzzy-PID based on TLBO optimization before and after improvement,the Fuzzy-PID control optimized based on TLBO algorithm is better than the general fuzzy PID and traditional PID control in suppressing vehicle vibration,And the vertical acceleration and suspension dynamic stroke of the active suspension controlled by Fuzzy-PID control optimized based on the improved TLBO algorithm were optimized by 19.5% and 38.5%,respectively;under the impact road input,the Fuzzy-PID control optimized based on the improved TLBO algorithm is 14.9% lower than the vertical acceleration of the active suspension body under the improved ABC-PID control,which verifies the superiority of the algorithm in controlling the vehicle smoothness and comfort,and the adaptability and robustness of the fuzzy PID control system are also stronger.Finally,the 1/4 active suspension PID control system optimized based on the improved TLBO algorithm is simulated,and the ANFIS controller design after training is carried out with the obtained data,and the simulation results show that the active suspension body vertical acceleration,suspension dynamic stroke and tire dynamic displacement based on TLBOANFIS control are optimized by 12.6%,20.8% and 19% respectively compared with the active suspension controlled by TLBO-PID.The active suspension system controlled by TLBOANFIS not only combines the high solution speed and high accuracy of the algorithm itself,but also combines the adaptive self-learning ability of the neuro-fuzzy system to effectively improve the suspension performance.