University of Twente Student Theses

Login

Torque based anti wheel lock control at NEVS

Barels, E.G. (2017) Torque based anti wheel lock control at NEVS.

[img] PDF
10MB
Abstract:In order for the driver or autonomous driver to be able to maintain control over a vehicle when braking it is advantageous to have not too many slip between the tire of the wheels and the surface of the road. This is controlled with an Anti-lock braking system (ABS) which regulates the oil pressure in the hydraulic pipelines to the brake calipers, in this way the amount of friction on the friction disks can be controlled, leading to more or less braking effort. This system works great, but for vehicles equipped with an electric motor, the electric motor itself can also be used for braking. Especially for vehicles which have a separate electric motor for each wheel, it would be advantageous if the electric motor also could contain ABS to get rid of the mechanical ABS. Furthermore it is highly suitable for torque vectoring applications. As with electric motors the current can be measured to estimated the actual torque, it is investigated if it is possible to make an ABS controller which only needs this torque to be functional. This report contains two subjects, one is about the identification of a system and the other is about a control algorithm. First some effort was made to make an identification of the electric motor from NEVS which was available for testing at a test rig. There was already some dynamic model created, but it has to be sough out if this model was accurate enough since for using an electric motor for ABS more insight in the dynamic behavior might be very useful. Another company which makes in-wheel motors has provided some measurement data of demanded torques and resulted torques, to see if the identification is also applicable to their setup. This model is used in a quarter car model, which simulates one wheel and electric motor of a vehicle. After the model of the electric motor was validated, an algorithm is created that uses the torque measurements to determine if the wheel is starting to lock up, leading to slip between the tire and the road surface. The algorithm consist of three phases, but is called a two phase algorithm, as it changes phases between two phases, a phase in which the torque needs to be increased and a phase in which the torqued needs to be decreased. The remaining phase is an initial phase, which is only active in the beginning of a braking event. As the title of this report suggest the decision making of the phases is based on the actual torque between the wheel and the road surface. When the braking torque increases, there will be a point at which the tire starts to slip, when this happens the actual torque will be reduced hence the tire is slipping, In oder to prevent the tire from slipping the demanded torque should be reduced, however there is an optimal slipping point at which maximum torque transfer between the wheel and the road surface will be obtained, this optimum depend on many parameters and can be different for each wheel during a braking event or even change during a braking event. The challenge was to make an algorithm that was able to cope with that. Furthermore the ABS algorithm is combined with an existing Electronic Stability Controller (ESC), this is simulated in IPG Carmaker. A first order model for the system is used in the simulations regarding to the two-phase algorithm. Four simulations are performed, the difference between the test cases for the simulations are the road surface conditions and the activation of the ESC. More specific a split-mu case with and without ESC, and a changing mu case with and without ESC. Also a simulation with a build-in ABS controller is carried out for validation of the performance of the ABS using the two-phase algorithm.
Item Type:Internship Report (Master)
Faculty:ET: Engineering Technology
Programme:Mechanical Engineering MSc (60439)
Keywords:ABS, ESC, torque, fuzzy logic
Link to this item:http://purl.utwente.nl/essays/84453
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page