Extended High-Order Disturbance Observer-Based Clutch Actuator Model Uncertainty Estimation of Ball-Ramp Dual-Clutch Transmission

Abstract

The ball-ramp dual-clutch transmission (BR-DCT) was proposed to overcome the disadvantages of conventional DCT. Specifically, BR-DCT is designed to reduce the amount of clutch actuator energy consumption and reverse torque during tie-up, in which two clutches are engaged simultaneously. In addition, since the self-energizing principle is used, the actuator model can be used without the friction coefficient, which is the main uncertainty of the clutch actuator. These features can greatly contribute to enhancing the shift control performance of the BR-DCT. However, the measurement error due to friction between actuator parts causes a small amount of uncertainty. In this article, a method using extended high-order disturbance observer (EH-DOB) is proposed to estimate this uncertainty. The EH-DOB used in the proposed method guarantees convergence even in a system in which the gain of the disturbance is time-varying. The convergence of the estimation error was verified through proof of the proposed algorithm. Experimental verification of the proposed algorithm was performed using a test bench that simulates a powertrain equipped with BR-DCT. As a result, it was confirmed that the uncertainty estimation result using EH-DOB showed high accuracy. As a result, it was verified that the uncertainty estimation method using the proposed algorithm has high accuracy and stability enough to perform disturbance rejection control.