Research On ECAS Multi-Agent Fuzzy Control Based On Fish Population Dynamic Optimization

Electrically controlled air suspension(ECAS)has the function of actively adjusting suspension stiffness or damping,which can greatly improve the comprehensive performance of vehicles.At the same time,it has a broad market prospect.At present,the key to breaking through the technical bottleneck of ECAS is how to decompose the front,rear,left and right suspension characteristics from the comprehensive performance requirements of the vehicle in complex driving conditions and make quick response.Therefore,based on the idea of Multi-Agent,not only this paper regards each suspension as an independent Agent,but also it introduces the fish swarm algorithm to simplify the connection relationship between the vehicle and the Agent,and it applies the fuzzy rules to solve the tasks with complex logical relations within the Agent to reduce the interaction between Agents,which improves the parallel processing capability of the system.In order to test the effectiveness of this method,the performance of a certain vehicle model under the conditions of constant speed,straight line,deceleration and steering was simulated and analyzed.The main research contents and work details are as follows:1.Establishment and verification of mathematical model of ECAS vehicle system.Firstly,a nine degree of freedom ECAS vehicle system model is established,including four tire lateral mechanical models,longitudinal mechanical models,four-wheel random excitation pavement model,etc.Secondly,Simulink is used to build ECAS vehicle system simulation model.Finally,the vehicle performances under uniform speed,straight line,reduction and turning conditions are simulated,and the simulation results are compared with TruckSim software simulation results.The results show that the two results are consistent.2.Research on ECAS fuzzy control based on fish swarm dynamic optimization.ECAS Multi-Agent hybrid structure is proposed,which is composed of system Agent,four suspension Agents and four tire Agents.Then the functional attributes of each Agent are assigned and the control mode of the system is determined.The system Agent dynamic optimization suspension Agent method based on fish swarm algorithm is proposed.Finally,a suspension Agent control method based on fuzzy rules is proposed.3.Simulation analysis.In order to verify the effectiveness of the proposed method,the correlation simulation analysis is carried out.The results show that: under the uniform linear condition,compared with the passive suspension,the vehicle body acceleration,roll Angle acceleration and pitch Angle acceleration of the suspension Agent are reduced by 20.03%,14.64% and 10.55% respectively.On the basis of suspension Agent,adding system Agent can further reduce the above parameters by 16.55%,14.38% and 12.49% respectively.In the braking condition,compared with PID control,the vehicle body pitch Angle RMS of suspension Agent and system Agent are reduced by 18.94% and 56.06% respectively.Under steering conditions,the RMS of vehicle body roll Angle of suspension Agent and system Agent were reduced by 20.81% and 61.07% respectively compared with PID control.Under the condition of mutual switching,it can be seen from the comparison curve of each performance parameter of the vehicle that,compared with PID control,ECAS Multi-Agent fuzzy control based on fish swarm dynamic optimization realizes real-time optimal control of the vehicle in a variety of conditions and mutual switching.