汽车主动悬架LQR控制器平顺性控制仿真

Simulation of Ride Comfort Control of Vehicle Active Suspension LQR Controller

LQR控制器的汽车主动悬架中控制器的权重系数矩阵Q主要依靠计算机的不断计算或人员的先验知识来确定,难以得到全局最优的LQR控制器。因此,提出一种新的混合优化算法PSO-DE来优化系数矩阵Q,PSO-DE混合优化算法能够保持粒子多样性,同时还可以对粒子群进行全局搜索以获得全局最优解。将混合优化算法PSO-DE-LQR与DE-LQR控制算法相比较,并结合某车参数在Simulink中建立混合型1/4悬架仿真模型。将被动悬架、DE-LQR主动悬架和混合优化算法PSO-DE-LQR主动悬架的车身加速度、悬架动行程及轮胎动位移三项性能指标的均方根值进行了对比分析。仿真结果表明,混合优化算法PSO-DE-LQR明显优于DE-LQR主动悬架和被动悬架,在很大程度上能减少路面冲击对车身振动的冲击,能显著地改善乘客乘坐的平顺性与驾驶员的操作稳定性。

The weight matrix Q of the controller in active suspension of the LQR controller is mainly determined by continuous computation of computers or depends on experience, and it is difficult to obtain the global optimal LQR controller. Therefore, a new hybrid optimization algorithm, PSO - DE, is proposed to optimize the coefficient matrix Q. The PSO - DE hybrid optimization algorithm can keep the diversity of particles and the global search ability of the particle swarm to obtain the global optimal solution. Compared the hybrid optimization algorithm PSO - DE - LQR with DE - LQR control algorithm combined with the parameters of some vehicle in Simulink, a hybrid 1/4 suspension simulation model was established. Based on it, the root - mean - square values of the vehicle body acceleration, sus- pension dynamic travel and tire dynamic displacement of the active suspension, the DE - LQR active suspension and the hybrid optimization algorithm PSO - DE - LQR active suspension were compared and analyzed. The simulation results show that the PSO - DE - LQR hybrid optimization algorithm is superior to the DE - LQR active suspension and the passive suspension, which can greatly reduce the impact of road impact on the vibration of vehicle body and can significantly improve the ride comfort and the driver＇s operational stability.