An Automated Model-Order Reduction Method for Automatic Transmissions

Abstract

New generation of torque converter automatic transmissions (AT) includes a large number of gears for improved fuel economy and vehicle performance, which leads to exponentially increasing number of shift types and shift events. In order to facilitate various numerical/simulation analyses of AT dynamics, shift control optimization, and control strategy design, a full-order AT model is usually reduced by eliminating state variables related to locked clutches in particular gears or shifts. The paper proposes an automated model-order reduction method for an arbitrary, user-specified clutch state, and demonstrates its application on an example of ten-speed AT. The method is based on determining the locked-clutch torque variables and their substitution into the full-order state-space model input vector, as well as finding a linear relation between the reduced-order and full-order model state-space variables.