Gear Shifting with Retained Power Transfer
The most essential property of a powershifting transmission is that power can be transferred even during a gear shift. Automatic transmissions in vehicles are a common application. The gearbox is shifted by simultaneous engaging and disengaging of clutches, connected to a gear train. Analytic simulation models are useful tools when optimizing the control strategy for these clutches.
This dissertation presents the following models for analysis of powershift operations in vehicle drivelines:
* Model for dimensionless equations: A driveline model without too many details generates equations, which are made dimensionless in a lucid way. Through these, a good overview of the principles of powershifting is found. Also, some phenomena of interest are strictly defined (tie-up, flare, torque and inertia phase, etc.).
* Two-clutch model: This driveline model has a gearbox, which shifts between two gears by controlling two clutches, one for the lower gear and one for the higher gear. The simulation results are verified by tests. Thus the level of detail in the driveline model is found to be appropriate.
* Multi-clutch model: In this model, the gearbox is equipped with as many clutches as the real gearbox has. (In the presented example, this means five clutches.) Therefore, all gear shifts for a certain gearbox can be analyzed with the same model. Also, it is possible to analyze other shift operations than those where exactly one clutch is engaging and one is disengaging.
* Model including surrounding systems: The driveline model is connected to models of the engine block and vehicle. Also, a simple model of a passenger is present, making it possible to study the passenger comfort more properly.
The clutches are often oil-immersed, and so they cannot be properly described with dry friction characteristics. A model of an engaging multi-disc clutch, including squeeze and viscous friction in the oil film, is therefore developed. The influence of the oil film on the gear shift quality is studied.
The modelling is carried out systematically. Therefore, the work also gives some general advice as to how dynamic transmission systems should be modelled. An algorithm is developed to find the appropriate state variables for a constrained dynamic transmission system.