摩托车半主动悬架分层预测控制及仿真

Hierarchical Preview Control and Simulation of Semi-active Motorcycle Suspension

行驶平稳性和乘坐舒适性是衡量车辆悬架性能优劣的重要指标,而基于磁流变技术的半主动预测控制与其它控制方法的结合则是改善悬架性能的一种先进控制方法。为提高控制效果,提出了一种悬架振动分层建模预测控制的新思路,即将前后轮系均看作是相互独立的底层二自由度系统,推导出协调底层关系的上层关联动力学方程,结合轴距预测和LQR优化控制方法,以前轮检测到的路面激励作为后轮下一步输入的激励,然后以簧载质量质心处的垂直加速度和俯仰加速度为上层控制目标进行协调,继而得到所需的半主动控制力。通过对一个四自由度摩托车模型的仿真计算表明,该方法在线计算量少,前后轮系易于采用不同的控制策略以提高控制效果。该方法对于多轮系车辆的振动控制提供了一个崭新的思路。

It is well known that ride comfort and handling performance are all important targets for a vehicle suspension, and combination of preview control and other strategy based on magneto-rheological technology is an advanced technology for the characteristics improvement of suspension. For improving control effect, a preview control strategy based on hierarchical modeling was put forward. In this method, the front and rear wheel systems of the vehicle were all dealt with two independent local 2-DOF systems, and some associated central dynamic equations that coordinate local relations were deduced. The wheelbase preview control that means the road excitation of front wheel is as the known input of rear wheel was combined with a quadric LQR algorithm. The vertical acceleration and pitch acceleration of the suspension center were acted as center control objects to assort with the two quarter suspension systems, so the semi-active control forces could be obtained. A 4-DOF motorcycle model was given and the results of simulation showed that the control strategy had less calculating time and it was convenient to adopt different control strategies for front and rear wheels respectively. The given method provided a new thought for the vibration control of multi-wheel vehicles.