An investigation into the roll control of vehicles with hydraulically interconnected suspensions

Publication Type:
Issue Date:
Full metadata record
Files in This Item:
Filename Description Size
01front.pdf381.94 kB
Adobe PDF
02whole.pdf3.78 MB
Adobe PDF
This thesis presents the investigation into a roll-plane hydraulically interconnected suspension (HIS), which is safety-oriented and designed for the vehicle with a high centre of gravity (CG) and a low rollover threshold. As a potential replacement of anti-roll bars, the HIS possesses the ability to resist the roll motion of the vehicle body during cornering or sharp turning by improving the vehicle roll stiffness. The previous research has concluded that the HIS is superior to the anti-roll bars in terms of the anti-roll performance, but its road holding performance has not been thoroughly studied. In this research, the modelling, modal analysis and simulations are conducted to compare the road holding ability of the HIS and anti-roll bars. A multi-function HIS test rig is developed and mounted on a typical Sports Utility Vehicle (SUV). The related experiments are implemented on a four-post test rig. Both the simulation and experiment results confirmed that the HIS is better than anti-roll bars from the perspective of the road holding performance. To overcome the drawback of the HIS that it is unable to handle the large roll motion and the vehicle roll caused by uneven roads, the HIS is then developed to be actively controlled by a control unit. Only one servo valve is included in the control unit of the active HIS so that the system’s cost and energy consumption are much lower in comparison with the conventional active suspensions with four independent motor-actuators. An output feedback H∞ controller is developed based on an empirically estimated active HIS model at a half-car level. The active HIS controlled by the designed H∞ controller is experimentally validated on the test rig with considerable roll angle reductions. For further verifying the controllability of the active HIS and also comparing the effect of different categories of control methods on the active HIS, other three representative control algorithms are also applied to the active HIS equipped vehicle. They are the classic control methods: proportional-integral-derivative (PID) control, the optimal control algorithm: linear-quadratic regulator (LQR) control and the intelligent control algorithm: fuzzy logic control. The obtained fuzzy, fuzzy-PID and LQR controllers are implemented in simulations. The experiments of the fuzzy and fuzzy-PID controllers are also conducted. The fuzzy-PID controller presents the most promising and stable control performance among these three controllers. After that, an attempt is made to improve the control performance of the model-based controllers to enhance the roll resistance ability of the active HIS further. A nine-degrees-of-freedom (nine-DOF) model that can capture the physical characteristics of the active HIS more accurately is developed. The new system model that addresses the relation between the flow change and pressure variation of the hydraulic system, and the viscous resistance of the fluid is also included. Then an H∞ controller and an LQR controller are designed based on the new model and validated in simulations. The experiment of the H∞ controller is also performed on the test rig and the H∞ controller derived from the new model is compared with the H∞ controller derived from the old model. The results show that the H∞ controller based on the new model improves the control performance slightly. Lastly, an effort is made to reduce the effects of the time delay caused by the fluid system by considering the system time delay when developing the controller. Delay-independent and delay-dependent H∞ state feedback controllers are designed and applied to the half car model. The simulation validations of the obtained controllers are carried out in MATLAB. It is found that the developed delay-dependent H∞ controller can provide stable and acceptably good control performance even with system time delay.
Please use this identifier to cite or link to this item: