Suppression Control On Torsional Vibration Of Powertrain For Electric Vehicle

The growing environmental pollution and the energy crisis urge the development of electric vehicles(EVs)for the emission reduction and the improvement of energy efficiency.Centralizeddriven pure electric vehicles(PEVs)have become the steadfast trend in academia and industry due to the good inheritance of mature power structure from traditional fuel vehicle,which makes PEVs easy to be controlled.As one of the most important parts of PEV,the powertrain system affects the control performance of the vehicle directly.The electrified powertrain trends to make the powertrain fewer damping parts and more simple mechanical structure.Thus,the electrified powertrain is more sensitive to the oscillations.In this thesis,the vibration problem of electrified powertrain is studied with the coupling mechanical-electric-network effects considered comprehensively,and a compositive controller is designed to suppress the system oscillation.Based on the analysis of drive motor,drivetrain and communication network,where the factors may cause system vibration,torque ripple in driving motor,non-linear backlash and driveshaft flexibility in the drivetrain and network-induced time-varying delay are all taking into consideration.An integrated time-discrete model is built to cover all these factors.In order to deal with the coupling effect between the event-driven controller nodes and the network-induced delay,polytopic inclusion approach is used to describe the delay,and a delay-free discrete model is established via system augmentation technique.Owning to the nonlinear backlash,the system is divided into two operating modes named common mode(co)and backlash mode(bl).Due to the advantage of robust control method on dealing with the system uncertainty,this thesis uses robust control to handle the system uncertainty caused by network-induced delay.Taking speed tracking and torsional vibration suppression as control objectives,a robust controller based on peak to energy(E2P)is designed,whose stability was proved via Lyapunov theory with the aid of Schur complement theory.The bilinear terms in the robust matrix are decoupled by means of congruence transformation,and the robust control gain is obtained by solving a set of linear matrix inequalities(LMIS).When the system is in the bl mode,the motor side and load side are decoupled,which is considered as a nonlinear state,and the sliding mode control(SMC)is suitable for it.Thus,a sliding mode compensator is further designed to suppress the nonlinear oscillation.The stability of the electrified powertrain system is insured via Lyapunov theory.The simulation test is carried out by using MATLAB/Simulink,and the controller area network(CAN)is built via Sim Event toolbox.By loading a large number of background signals to the CAN bus,which causes the network congestion,and the time-varying delay is created.The influence of the coupling effect of mechanical-electric-network on the torsional vibration of the system is verified by simulation,and the robustness of the proposed controller is verified by comparing with traditional PI controller.The simulation model is consistent with the theoretical dynamics in the Simulink.And in order to further verify the control effect of the proposed controller,the drivetrain and the load is further established via AMESim,where the gear transmission and the load are considered more in detail to meet the reality.Through the co-simulation with AMESim and Simulink,it is further verified that the designed controller can effectively suppress the torsional vibration of the driving system of pure electric vehicle,and has sufficient applicability and effectiveness.