论文部分内容阅读
以含磁流变阻尼器的1/4车辆非线性半主动悬架模型为研究对象,在充分考虑该非线性系统未建模动态的基础上提出了具体的神经网络与滑模变结构控制相结合的智能控制策略,有效抑制了悬架系统的振动,使车辆行驶的平顺性与舒适性得以提高。应用神经网络的在线学习能力对非线性动力学模型的不确定部分及外界未知扰动进行了神经网络估计,确定了未知函数的上确界,构造了控制系统的滑模变量并且合理设计神经网络的自适应规律使状态变量快速接近原点。通过稳定性分析证明了此种控制方法是全局渐近收敛的,并且对未建模动态具有强鲁棒性。数值仿真结果验证了该种控制方法的有效性,得到了阻尼器两端控制电压的变化规律。
Taking a quarter-vehicle nonlinear semi-active suspension model with magnetorheological damper as the research object, this paper presents a concrete neural network and sliding mode variable structure control phase based on the unmodeled dynamics of the nonlinear system The combination of intelligent control strategy, effectively inhibiting the vibration of the suspension system, so that the ride comfort and vehicle can be improved. The neural network’s online learning ability is used to estimate the uncertain part of the nonlinear dynamic model and the unknown disturbance. The upper bound of the unknown function is determined, and the sliding mode variables of the control system are constructed and the neural network is reasonably designed Adaptive rules make the state variables quickly approach the origin. The stability analysis shows that this control method is globally asymptotically convergent and strongly robust to unmodeled dynamics. Numerical simulation results verify the effectiveness of this control method and obtain the variation of the control voltage across the damper.