Optimal Control For Gearshift Of A Two-Speed AMT Electric Vehicle

With the development of multi-gear drive system equipped in pure electric vehicles,the control of gearshift process has become a research direction in automotive electronic control system development and design.For one hand,the external characteristic of motor meets the ideal operating characteristics of the car,for the other hand the motor can be faster and more flexible controlled compared with the traditional engine,thus the gearshift control of electric vehicle should fully take the advantage of motor's fast response speed and high accuracy.The coordinate control of motor and clutch is feasible and reasonable for the gearshift control of EVs.The concept of gearshift coordinate control makes it possible to achieve more economical and smooth shift process,meanwhile the increased control degrees of freedom leads to higher requirements of the control system.From the perspective of engineering applications,the huge computation burden of optimal problem solving makes the implementation difficult to realize.Under this research situation,in this paper,the gearshift control of a novel 2 speed AMT electric vehicle is taken as an example,and based on its structure characteristics and the gearshift control demand,a control method under the frame of model predictive control is proposed.And the explicit solution of the optimal control problem is obtained by combining the minimum principle and method of bisection.The effectiveness and real-time performance are verified by off line co-simulation and online tests.Firstly,the structure and gearshift process of the 2 speed electric vehicle are analyzed in detail,and then the simulation model of the electric vehicle is built in AMESim software.Is worth pointing out that because the clutch friction and thermal characteristic influence the gear shift process a lot,a detailed clutch model is built for this research.Then,simulation tests are conducted to test the rationality and functionality of the simulation model,the results indicates that the model is suitable for this research.Secondly,aiming at the signal interference in gearshift process and model mismatch problem in shift controller design,a control scheme is proposed under the frame of model predictive control.The control-oriented mathematical model of the gearshift inertia phase is built considering the clutch main thermal process and friction characteristics.Then the control problem is concluded as a fixed terminal time,multi control variable,constrained,multi objective optimal control problem.Thus the control horizon is diminishing with the gearshift time in this control method.The solution of the optimal control is obtained by combining the minimum principle and method of bisection,which can improve the real-time performance of the controller.Based on above control strategy and solving method,the controller model is built in Matlab.Then,the control performance of the proposed controller under different acceleration command,vehicle mass and road grade is tested through off line co-simulation.And the control performance without considering the thermal effect of the clutch is also given in this paper,which indicates that the effectiveness of the proposed control method.At last,the effectiveness of the controller is further verified in the d SPACE and x PC-Target real-time experimental platform.This paper provides some ideas for the gear shift controller designing,but there are also many works to be done.In the future we will match the simulation model with the actual drive-line and simplify the solving program.And the further experimental validation study of the control method in real vehicle should be considered.