| Suspension is an important component of vehicle, which affects the ride comfort of vehicle. Ride comfort with traditional passive suspension is poor and limited to improve. While active suspension has so many shortages such as high power consumption, large size and high cost that it is difficult to popularize in passenger car. Therefore, the semi-active suspension has attracted more and more attention to motor firms in recent years. In this paper, a semi-active suspension control method based on optimal control theory is studied, which is designed to control the attitude of the vehicle body on vertical, pitch and roll. The specific contents are as follows:(1) By searching relevant literatures, the body attitude control with electronic control suspension, adjustable shock absorbers for semi-active suspension system and semi-active suspension control methods are illustrated.(2) The model of road and quarter car were built, and Linear Quadratic Regulator(LQR) optimal control algorithm for semi-active suspension based on quarter car model was established to simulate. Then, the effect of optimal control weighting coefficients for vertical acceleration of vehicle body centroid, suspension dynamic deflection and wheel dynamic load have been discussed.(3) The body attitude controller using LQR control theory based on a seven-degrees-of-freedom(7DOFs) vehicle model has been investigated, in which the control targets include vertical acceleration of body centroid, pitch angular acceleration and roll angular acceleration. As the application of LQR control to real vehicle requires knowledge of all states of the 7DOFs vehicle model, it is difficult to apply LQR control if the states include absolute displacement or absolute angle which is difficult to measure. By using first-order-derivative, the new state equation still conforms to the standard form of optimal control. LQR control based on the new equation of state can run without absolute displacement and absolute angle, and translates control damping force into the control increment. Simulation results show that the incremental LQR control has a good control effect. Finally, an incremental LQR control algorithm with weight coefficients self-adaption according to the deformation of the suspension is proposed.(4) The incremental LQR control algorithm was verified in different conditions by using the Car Sim which provides more accurate vehicle model. The parameters of vehicle model were set according to a real car. A real magneto-rheological damper(MRD) used on the car was tested by test rig. According to the experimental data, a lookup table model of MRD for simulation and an inverse model of MRD for control system based on rotary-stretch-surface fitting were established. In the inverse model, control current of MRD is calculated according to the ideal damper force and the compression rate of suspension. The control effect of incremental LQR control and adaptive incremental LQR control was analyzed by simulation of passive suspension and semi-active suspension with MRD.(5) Practicability of adaptive incremental LQR control algorithm was analyzed through building a hardware in the loop test rig. The hardware in the loop test rig mainly includes: a host computer, a d SPACE hardware board, a controller which integrated driver, a current acquisition module and MRDs used for loaders. First of all, a body attitude controller based on freescale microcontroller is designed. Then, a MRD driver with current feedback is designed based on BUCK circuit and PID control algorithm and response speed of the driver is analyzed through experiment. Finally, the practicality of the control algorithm was verified by the real time simulation of Car Sim RT software which setsss different working conditions. |
PDF Full Text Request