| To improve fuel economy, reduce emissions, and enhance driving performance, thereare currently several types of transmissions and the associated technologies that oferdiferent performance priorities when fit into a vehicle, such as Automated Manual Trans-mission AMT, Dual Clutch Transmission DCT, Automatic Transmission AT and Con-tinuously Variable Transmission CVT, etc. Each of these technologies has their ownpenetration levels. However, originally marketed by Volkswagen as the DSG and Audi asthe S-Tronic, DCT with its refinement, efciency, low cost and driving satisfaction is theprobable winner. DCTs are now being ofered by several automakers, including Nissan,Mitsubishi, BMW and Porsche, etc.In DCTs, gearshifts can be regarded as so-called clutch-to-clutch shift consisting oftwo distinctive parts, a torque phase and an inertia phase. During the inertia phase, notonly the torque but also rotational speeds change intensively as well, hence the inertiaphase is closely related to shift quality, and therefore, it is necessary to carry out closed-loop control of clutch slip during this phase. However, shift quality is a function of manyaspects including shift time, friction losses during the slipping phase and shift smoothness,and these objectives are diferent and sometimes in conflict. Furthermore, because ofdiferences in the design compared with ATs (such as the absence of one-way clutches andtorque converter on DCTs), the ability of DCTs to achieve good shift quality in variousoperating conditions has always been a control challenge.Much attention has been paid to clutch slip control, for example, calibration experi-ments and optimization-based algorithms are potential solutions for this problem. How-ever, because the duration of gearshift operations during driving is very short and thehighly nonlinear powertrain dynamics, clutch speed-tracking control scheme is adoptedto realize clutch slip control in this paper. Although it doesn’t consider the multiplecontrol objectives directly during the shift operations, the required shift time and shiftcomfort can be reached by selecting a proper reference trajectory, the friction losses canalso be reduced by suitable engine-torque coordination during shift process. Furthermore,in this control scheme, the only control objective is to make clutch speed track the ref- erence trajectory, thus it’s relatively easy to realize. Then, in order to achieve clutchspeed-tracking and improve shift quality, model-based control is introduced. Since dy-namics of the gearshifts during the inertia phase can be described in the form of strictfeedback, backstepping technique is used to design the clutch slip controller with the sys-tem nonlinearities taken into account. Here complex nonlinear characteristics from enginesystems are represented by lookup tables. And these tables with their original form willbe used in the proposed nonlinear controller. Furthermore, model uncertainties includ-ing steady-state errors and unmodeled dynamics are considered as additive disturbanceinputs, and the controller is designed such that the dynamics of the speed-tracking erroris input-to-state stable (ISS). Finally, the proposed clutch slip controller is programmedusing MATLAB/Simulink and combined with the powertrain simulation model, whichis established by AMESim, through co-simulation. Comparisons with a PID clutch slipcontroller verify the potential benfts of the proposed nonlinear controller in achieving lesstracking error. And it can be seen that the transmission output torque does not show anyabrupt changes at the point of the transition from the torque to the inertia phase, thusverifying the smooth control of the engine torque transfer by the clutch slip controller.Finally, it should be noted that, this paper focuses on the shift transient control andthe determination of optimal shift point is not considered. And testing the controller inHiL (hardware-in-the-loop) platform deserves the further attention, which has not beencarried out until now. |
PDF Full Text Request