This thesis is concerned with electronic stability of an all-wheel drive electric vehicle with independent motors mounted in each wheel. The additional controllability and speed permitted using independent motors can be exploited to improve the handling and stability of electric vehicles. In this thesis, these improvements arise from employing a direct yaw moment control (DYC) system that seeks to adapt the understeer gradient of the vehicle and achieve neutral steer by employing a supervisory controller and simultaneously tracking an ideal yaw rate and ideal sideslip angle. DYC enhances vehicle stability by generating a corrective yaw moment realized by a torque vectoring controller which generates an optimal torque distribution among the four wheels. The torque allocation at each instant is computed by finding a solution to an optimization problem using gradient descent, a well-known algorithm that seeks the minimum cost employing the gradient of the cost function. A cost function seeking to minimize excessive wheel slip is proposed as the basis of the optimization problem, while the constraints come from the physical limitations of the motors and friction limits between the tires and road. The DYC system requires information about the tire forces in real-time, so this study presents a framework for estimating the tire force in all three coordinate directions. The sideslip angle is also a crucial quantity that must be measured or estimated but is outside the scope of this study. A comparative analysis of three different formulations of sliding mode control used for computation of the corrective yaw moment and an evaluation of how successfully they achieve neutral steer is presented. IPG Automotive’s CarMaker software, a high-fidelity vehicle simulator, was used as the plant model. A custom electric powertrain model was developed to enable any CarMaker vehicle to be reconfigured for independent control of the motors. This custom powertrain, called TVC_OpenXWD uses the torque/speed map of a Protean Pd18 implemented with lookup tables for each of the four motors. The TVC_OpenXWD powertrain model and controller were designed in MATLAB and Simulink and exported as C code to run them as plug-ins in CarMaker. Simulations of some common maneuvers, including the J-turn, sinusoidal steer, skid pad, and mu-split, indicate that employing DYC can achieve neutral steer. Additionally, it simultaneously tracks the ideal yaw rate and sideslip angle, while maximizing the traction on each tire[CB1] . The control system performance is evaluated based on its ability to achieve neutral steer by means of tracking the reference yaw rate, stabilizing the vehicle by means of reducing the sideslip angle, and to reduce chattering. A comparative analysis of sliding mode control employing a conventional switching function (CSMC), modified switching function (MSMC), and PID control (HSMC) demonstrates that the MSMC outperforms the other two methods in addition to the open loop system.
| Identifer | oai:union.ndltd.org:CALPOLY/oai:digitalcommons.calpoly.edu:theses-3554 |
| Date | 01 October 2019 |
| Creators | Wight, Winston James |
| Publisher | DigitalCommons@CalPoly |
| Source Sets | California Polytechnic State University |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Master's Theses |
Page generated in 0.0033 seconds