Design of a model-based robust controller for vehicle with wet type dual clutch transmission

Abstract

This research propose a model-based robust control strategy for the driveline system of a vehicle equipped with a wet Dual Clutch Transmission(wDCT). Since DCT transmits power to the vehicle by using two clutches without a torque converter, good shifting performance (quick and smooth shifting) can be achieved only when precise engagement control is achieved. Such shift control requires high control performance for a short period of time ending within about 2 to 3 seconds. In order to satisfy the required control performance, a model-based feed-forward control is performed. Since feed-forward control requires accurate model information, a model that accurately describes the phenomenon is required. The wDCT differs from the dry DCT in that it uses a hydraulic actuator and the torque characteristics of a wet clutch using lubricating oil. Nonlinearities exist in these hydraulic actuators and wet clutch torques. In this research, a novel model of hysteresis and clutch filling phase for hydraulic actuator, and a modified wet clutch torque model are proposed based on physical phenomena. Compared with the existing model, the proposed model is simplified through appropriate assumptions based on the physical phenomenon of the target system, increasing the applicability and practicality of the model to be controlled. The hydraulic actuator model is utilized to construct a pressure controller that acts as a lower controller of the driveline system. In addition, the wet clutch torque model is utilized to construct the clutch slip controller. The proposed new models have been simplified so that they can be used for control in consideration of practicality, and a feed-forward controller suitable for each situation is constructed based on the proposed model. In pressure control of hydraulic actuator, high tracking performance can be obtained by compensating for hysteresis and clutch filling phase using corresponding nonlinear models. The wet clutch torque model is applied to the clutch slip controller to predict the model behavior to increase the control performance. However, the feed-forward control can greatly contribute to the performance improvement, but cannot cope with the performance degradation due to model uncertainties and external disturbances. Therefore, the feed-forward control should be used together with the feedback control to ensure robustness. In this research, in order to solve the control problem of the driveline system, a modeling of the wDCT driveline system and a model-based 2-Degree of Freedom (2-DOF) robust controller that guarantees control performance and robustness are proposed. The proposed 2-DOF robust controller consists of model-based feed-forward and robust feedback control. System modeling considering non-linearity maximizes feed-forward performance, and feedback control consider more weight on robustness, excluding consideration of expected performance of feed-forward, so that optimal control performance can be obtained. Therefore, the 2-DOF robust controller proposed in this research shows a greater effect compared to the existing robust control methods in systems with limited feedback performance such as driveline systems. Therefore, in this dissertation, system modeling and controller design reflecting the characteristics of wDCT is performed to design a model-based 2-DOF robust controller that can maximize the control performance of the driveline system. By analyzing the uncertainties and its bounds of proposed models, a 2-DOF robust controller that enhance a robustness is constructed. Finally, the performance and effectiveness of the pressure controller and clutch slip controller of the vehicle's driveline are experimentally verified in various situations in production type vehicles.