Research On The Cooperative Control Method Of Vehicle Interconnected Air Suspension System

Air suspension systems are widely used in the vehicle industry field because of its outstanding vibration isolation performance and flexible body height adjustment.As a derived structure of the air suspension system,the interconnected air suspension system significantly improves ride comfort and handling stability with its outstanding vibration isolation and torsion elimination performances.Mostly,the traditional multi-structure coordinated control strategies of the interconnected air suspension are all trying to explore the optimal parameter combinations of each controllable structures under different working conditions,which are mainly based on the influence of the controllable structures' parameters on the vehicle performance.However,most of them stop at the coordinated control between damping and body height adjustment or between damping and interconnection states,and rarely come to the coordinated control between interconnection states and body height adjustment.Pointing at the shortcomings of the traditional control strategy in coordinated controlling the multi controllable structures of the interconnected air suspension system,this paper proposes a coordinated control strategy for interconnected air suspension based on the model predictive control strategy.Specifically,the optimal suspension force is calculated based on the objective of optimizing the overall performance of the whole vehicle.Moreover,the optimal suspension force is generated through the multi controllable structures of the suspension system.This strategy realizes the coordinated control of the multi controllable structures in the interconnected air suspension system and provides a theoretical basis and a implementation method for further improving the suspension performance and the coordination of multi controllable structures.Firstly,the working principle of the interconnected air suspension system was explained from the perspective of gas exchange.According to the variable mass system thermodynamics theory and the hydromechanics theory,the model of air spring,interconnection pipeline and gas tank were built.Taking the excitation coherence between left and right roads and the time delay between front and rear wheels into consideration,a road excitation model was built based on the white noise generation method.According to the 7 DOF vehicle model which is widely used in suspension-related researches,a new vehicle mathematical model with interconnected air suspension was established for simulation.Meanwhile,the linearization of air spring,interconnected pipeline and damper was performed to simplify the controller design.Based on the existing test vehicle,the whole vehicle information acquisition system was built with Arduino open-source software and hardware platform.The accuracy of the simulation model was verified by the real car test,which lays the foundation for the research on the cooperative control of the interconnected air suspension system.Secondly,an innovative cooperative control strategy based on model predictive control was proposed with the idea of optimal control.During the design of the model predictive controller,the cost function was constructed corresponding to the comprehensive performance of the whole vehicle;the state constraints were calculated based on the suspension forces that can be generated by each controllable structure;the weights of the objective function were optimized by using genetic algorithm.Moreover,an unscented Kalman filter and a H-infinity observer were designed,which provide the states required by the proposed model predictive controller.The simulation results showed that the designed observer can accurately estimate the state of the whole vehicle according to the measurable state.Meanwhile,using the proposed model predictive controller in the form of force feedback could comprehensively improve 7.6% ride comfort and 6.8% handling stability compared to the traditional control strategy,which fundamentally ensured the coordination of each controllable structures and the overall control performance.Thirdly,the multi-agent theory and game theory were introduced.Specifically,constraints,costs,and game relations of each agent,which are generated by the model predictive controller in completing the optimal suspension force task,were analyzed and discussed when taking each controllable structure as an agent.The shortcoming of solving the Nash equilibrium solution when using iterated best response and KKT-based approach in the current research was clarified.For the first time,a suspension force distribution method which aims at the optimal energy consumption was proposed.Meanwhile,the inherent contradiction between vehicle height and interconnection has been solved,and the responsibility scope between the interconnection agent and the vehicle body height adjustment agents has been divided.Specifically,it solves the task of discretizing the optimal suspension force,and then the generated task was assigned to the proper agent.Afterwards,the way of choosing the agent which could generate the current unit increment suspension force with the lowest energy consumption has been established.Finally,the results of the optimal suspension force distribution have been determined by the accumulative task values of each agent.The simulation results showed that the proposed suspension force distribution method could solve the optimal suspension force distribution task well,which ensured the consistency between the suspension force generated by each agent and the optimal suspension force based on resolving the fundamental contradiction between the interconnection and the vehicle body height adjustment agent.It achieved the cooperative control of multi controllable structures of the interconnection air suspension system and reduced the energy consumption.Overall,when combing this proposed method with the model predictive controller,it comprehensively reduced 73%?80% energy consumption while improved 4%?61% ride comfort and 6%?56% handling under different conditions compared to the traditional control strategy.Finally,based on the raspberry pi hardware platform and MATLAB / Simulink hardware in the loop simulation environment,a hardware in the loop simulation platform with raspberry pie 3B + was built.The corresponding hardware connection mode and software configuration were given for verifying the feasibility of the control strategy and the real-time performance of the controller,together with the configuration of the raspberry pi's system accordingly.The effects of different data types on the output data accuracy and the raspberry pi control effect were studied.The effects of different sampling time,prediction horizon and control horizon on the hardware in the loop simulation time consumption were analyzed.The hardware in the loop simulation results proved the feasibility of the proposed control strategy deployed on the raspberry pi hardware platform and the effectiveness of the raspberry pi controller.Based on the vehicle information acquisition system and the raspberry pi controller,a real car test of the interconnection and the vehicle body height adjustment cooperative control was carried out.The test results showed that the raspberry pi controller could achieve the cooperative control between the interconnection and the vehicle body height adjustment,which improved the performance of the interconnection air suspension system.The test results showed that the proposed strategy could improve 34% ride comfort under straight driving condition,reduce 20% vibration amplitude under bump condition,and reduce 21% pitch angle and 39% roll angle under up and down slope and turning conditions.