Realtime modeling of driveline dynamics as estimator for control of AMT-gearbox in heavy-duty road vehicle

Abstract

With Automated Manual Transmissions(AMT), gearshifting is controlled through algorithms controlling both engine and actuators inside the gearbox. The objective is to optimise shift in terms of comfort, energy efficiency and to minimise the time without propulsion. There is relatively low stiffness and high inertia in the driveline of a truck which affects torques and speeds of shafts in the gearbox. Torque is complicated and expensive to measure in production gearboxes. Therefore there is a value in developing algorithms for estimation and prediction. The algorithms modeling the drivetrain needs to be computationally efficient to be run in real time but also include the first order of resonance for the complete driveline. It is shown that a simplified mathematical model of differential equations can replicate the behaviour of the driveline and predict torques and speeds of shafts. Efforts has been made to reduce the number of states in order to reduce the computational load. Comparison of model with data from a test truck shows that accuracy can be achieved for a simplified model with the use of known physical properties for a specific vehicle. The model is also tested with forward Euler iteration method to determine necessary step size to produce a stable prediction.