车辆主动悬架二次型最优控制器权矩阵参数优化

在使用线性二次型调节器(LQR)的车辆主动悬架控制器中加权矩阵Q和R的取值经常依靠先验知识选取。粒子群算法具有良好的快速寻优能力,可以对权矩阵参数进行优化。针对目前算法存在的缺点通过在更新的过程中动态调整惯性权重以更好的平衡全局和局部搜索能力,同时在迭代后期加入禁忌搜索避免陷入局部最优解。在matlab中建立1/4二自由度主动悬架仿真模型,对振动控制性能仿真分析结果表明,采用改进粒子群优化LQR与传统LQR方法相比能够很大程度上减少路面变化对车身的冲击,乘坐舒适性和可操纵性得到明显提升。

Parameters Optimization for the Weight Matrices in a Vehicle Suspension LQR Active Control

The values of the weighting matrices Q and R in the vehicle active suspension controller using the LQR linear quadratic regulator often rely on prior knowledge. Particle swarm optimization algorithm has good fast optimization ability and can optimize the weight matrix parameters. In view of the shortcomings of the current algorithm, the inertia weights are dynamically adjusted in the process of updating to better balance the global and local search capabilities, and the tabu search is added later in the iteration to avoid falling into the local optimal solution. The simulation model of 1/4 two-degree-of-freedom active suspension is established in matlab. The simulation analysis results of vibration control performance show that the improved particle swarm optimization LQR can greatly reduce the impact of road surface changes on the vehicle body compared with the traditional LQR method. The ride comfort and maneuverability have been significantly improved.