Truck Differential and Rear Axle Modeling

Abstract: Physical modeling of the driveline is crucial in many areas within the commercial vehicle industries. To have an accurate model helps the understanding of physical phenomena and is important when performing computational tests and when designing, for example, traction controllers. The differential gear is often modeled as simple as possible, neglecting more complex behaviour. The main goal with this thesis is to create a model of the differential that takes into account some of these neglected properties. To be able to test and observe the behaviour of the differential, a complete driveline is modeled, extending from the engine to the wheels. The driveline model is then validated using existing measurement data. The results from the tests performed on the model show that there are minor differences on the wheel velocities if the differential is modeled using a more physical approach. Especially the differential behaviour have been shown to accurately describe some of the important features, improving model usability. Implementation of a more complex differential model depends on the area of usage of the model. In addition to driveline modeling, a novel traction controller has been developed and implemented, using the model developed. The controller has shown interesting features such as constraints and prediction, however, further investigations are required to achieve desired performance.