Research On Clutch Temperature Model And Gray Predication Control Of Dry DCT

DCT technology is the latest automatic transmission technology, which at moment attract much more attention from both automobile manufactures and end customers. Based on different clutch work principles, there are two type DCTs, one is called dry DCT, the other one is called wet DCT. Dry double clutch is the most important core components in dry DCT, the working behaviors of dry double clutch can influence the performance of dry DCT considerably.The design of dry double clutch is oriented from normal dry single clutch, but the structure is more complex, the arrangement is more compact, the install space is more limited. In order to realize the optimized shift schedule, to meet the demands of transmission control strategy, dry double clutch needs to work more frequent than normal dry single clutch. To improve shift comfort, and filter the vibration in drivetrain, dry double clutch has to slip much more. These factors make the working condition of dry double clutch much worse than normal dry single clutch, people need to do deeper research on dry double clutch to find out better solution on design and control, to meet the demands of dry DCT development.Dry clutch generated lots of friction power during slip, converted to heat to make the temperature of clutch components increase. The working performance of dry clutch depends on clutch temperature a lot; the friction coefficient, the friction area, the clamping force, and the wear are all influenced by temperature. The increase of clutch temperature may lead to big variation of clutch torque and big increase of wear, diminish the vehicle performance, and shorten the clutch lifetime. Higher temperature can also cause clutch failure, decline the robust and safety of vehicle. Therefore, monitor and control clutch temperature is the key technology of dry DCT development. Aim at the problems mentioned above, the thesis did deep research on clutch temperature model, clutch temperature prediction, clutch actuation system control, and so on.Since it is difficult to use sensor technology in dry DCT to obtain temperature information at clutch friction surface, and also intends to reduce system costs, currently, the feasible solution for the problem is, build clutch temperature mathematic model, and observe the real time clutch temperature according to vehicle running, then emend the clutch working parameters and control strategy. The research in chapter three solved the problem; the establishment of clutch temperature limitation also provides protection to avoid high clutch temperature.The simple clutch control based only on clutch temperature model still has disadvantages, the reasons are:1. The emendation of clutch control strategy based on current clutch status, lack of trend information of clutch temperature, therefore can not realize quantitative control for clutch, and may leads to risks of over control or inadequate control.2. On purpose of lowering the inputted friction power into clutch to reduce clutch temperature, it takes long time for clutch to reach new temperature balance with big lag, the clutch temperature may reach dangerous range.So, the prediction of clutch temperature is needed to solve the problem. Based on future trend of clutch temperature, amend clutch control strategy earlier and properly, to realize clutch temperature prior control, can compensate the weak points of simply clutch temperature model control. The clutch temperature gray prediction research in chapter four is corresponding solution.After developed effective clutch temperature control strategy, the control result still depends on the response of clutch actuation system. If clutch actuator react very slow and can not follow the control signals, or just work with very big overrun, the control target also will not be reached.Since clutch actuation system is nonlinear, time varying, uncertain, and has very big lag, the normal PID control can not fulfill the control requirements, new control methods to improve clutch actuator control performance are needed. The research on application of gray prediction PID control was done in chapter five.The thesis combine theoretical analysis, simulation, and test verification together, did deep research on above problems, found out solutions for clutch temperature problem, got applicable research results, details are introduced as followings.1. Research on DCT dynamic model and calculation of clutch friction powerThe dynamic model of DCT is the basis and key point for research on DCT control technology and clutch friction power calculation. The research analyzed the work process of DCT, based on former research of other people, create mathematic model of single parts firstly, and then divide DCT vehicle running into fixed gear running and shift situations, build separate system dynamic models. Research on inertia phase and torque phase during DCT shifts was also carried out.Based on DCT dynamic model, the thesis did simulation for several typical work conditions, such as acceleration, brake, and so on. The results show that, the shift time of DCT is very short, jerk during shift is also very small, and DCT can provide excellent performance in each condition. For example, the kick-down with shift on same input shaft is one of the worst, most complex work conditions, the research result shows that, DCT can still performs so good even in this work condition, and realized excellent acceleration performance in parallel to provide good shift comfort. The thesis also introduced different approaches on different work conditions for kick-down, which are very helpful for DCT development.In order to provide input for clutch temperature model, based on DCT dynamic model, the friction power as calculated, and the formula for minimum friction power calculation during clutch engagement was deducted, these can provide useful reference for clutch control improvements.2. Build clutch temperature modelThe influence of temperature on characteristics of dry clutch was introduced in details. But, the emendation of clutch parameters corresponding to current work condition, the control of clutch temperature, the control of clutch wear, need to get the clutch temperature information.Deep research on the generation and dissipation of clutch heat and the spread model was done on the basis of Fourier heat conduction law and Energy conservation law, clutch temperature of dry DCT was built, temperature on different work cross sections of clutch pressure plate and flywheel was calculated. The simulation and test with temperature model verified the correctness of clutch temperature model. Research based on two different typical conditions also proved the necessity of the study on clutch temperature model.To eliminate the influence of clutch temperature, corresponding improvement methods were raised on both hardware design and on control strategy. Increase first gear ratio of DCT or improve heat absorb and dissipate ability of clutch, can lower clutch temperature significantly. Lower engine speed and reduce shifts properly can reduce clutch heat through reduced friction power. The threshold settled in the thesis can thereafter prevent clutch from failures on the condition of too high temperature, to protect clutch, which can offer great significance on clutch robust.3. Gray prediction of clutch temperature To realize clutch temperature prediction, research on the prediction of clutch temperature was done on the basis of the theory of gray system. Theory of gray system focus on poor information and uncertain systems, predicts future behavior of the system through the calculation on gray model in stead of precise model of control object, can be used in widespread range.Based on calculated clutch temperature, the latest five data are chosen to build preliminary data sequences for GM(1,1) model, thereafter, the gray prediction model of clutchtemperature is built. The calculation on temperature model can provide multi steps clutch temperature prediction and get the trend of clutch temperature. The prior emendation on clutch control strategy based on predicted temperature realized prior control for clutch.Based on predicted clutch temperature trend, clutch control strategy is improved. With the prior emendation of clutch target torque, on condition to ensure clutch control performance, the temperature increase speed is lowered, the temperature trend is changed, the big adjustment on clutch torque is avoided, the engagement comfort is ensured, and the system performance is improved further.4. Gray prediction PID control on clutch actuation systemIt is difficult for traditional PID technology to meet the demands of clutch temperature control. The gray prediction PID control which based on theory of gray system, can builds gray model for uncertain factors in the system, can provide prior prediction and compensation, can 'whiten' certainly gray factors in the system, can improve the control quality and robust of PID, and also can solve the problems of overrun and oscillation of clutch control.The mathematic model of clutch actuation system as built, the functions of clutch torque and hydraulic pressure was deducted. Based on deducted equation of state of clutch actuation system, the uncertain factors in the system, include state relevant factors (parameters uncertain) and state independent interferences, were separated and merged. During the start up of PID controller, use gray estimator to estimate the model parameters of uncertain factors through GM(0, N) model, the control signal is compensated with estimated value; control quality can thereafter be improved significantly.The simulation and test result show that, gray prediction PID control can solve the lag and oscillation problems of clutch actuation systems, can enhance the response speed of clutch, can improve the robustness of the system, and thereby realized clutch temperature control finally.The research of the thesis, can provide complete solution to eliminate the influence of clutch temperature on system performance of dry DCT, by calculate and predict clutch temperature as well as correspondingly emended clutch control strategy, can improve control quality of DCT system significantly, thereby can provide valuable benefits for dry DCT development, and as well as for the development of AMT.