基于遗传算法的主动悬架最优控制方法研究

针对线性二次型调节器（LQR）在工程应用中所存在的权重矩阵确定困难的问题,利用遗传算法（GA）优化LQR权重矩阵,提出了一种改进的最优控制方法--遗传线性二次型调节器（GALQR）。该方法采用了分区淘汰搜索机制,兼顾了大范围全局搜索与局部细微搜索,保证了权重矩阵最优解的全局性和精确性。将GALQR方法应用于主动悬架控制规律的设计,其结果与经验型LQR方法进行了对比,表明GALQR方法增加了寻求最优权重矩阵的可能性,使系统获得更佳的设计性能。

Optimal Control of the Active Suspension Based on the Genetic Algorithm

The genetic algorithm(GA) was added to the linear quadratic regulators(LQR) and a new optimal control method-Genetic Algorithm Linear Quadratic Regulators(GALQR) was presented.The method used GA to optimize the weight matrices of LQR and adopted divisional removing searching mechanism to insure getting the global optimal results. The method was applied to design the control schedules of Active Suspension System.Through comparing with the experiential LQR,the results show the GALQR can increase the possibility of finding the optimal weight matrices and make the system gain better design performance.