基于遗传算法的车辆座椅悬架最优控制器的应用

常规LQG控制器在车辆座椅主动悬架应用中,存在权值矩阵确定的问题。设计了一种基于遗传算法优化LQG权值矩阵的方法。该方法利用遗传算法的全局优化算法,以主动座椅悬架性能指标为目标函数对权值矩阵进行优化设计,并在Matlab/Simulink环境下建立了系统模型并进行仿真模拟。仿真结果表明:基于遗传算法优化的LQG控制器能够解决了最优控制器中存在的权值确定问题,提高悬架的设计效率,同时对于乘坐舒适性的改善有良好的效果。

The optimization of LQG Controller in Vehicle Seat Suspension Based on the Genetic Algorithm method

Due to the shortcoming of the LQG(Linear Quadratic Gaussian) when defining the weight matrices,a method using the genetic algorithm is designed to optimize the LQG controller.This method utilizes the global optimization algorithm of the genetic algorithm to optimize and design the weight matrices with the target function of chassis performance indexes.This method aims at improving the design efficiency and control performance of the LQG controller.And then a simulation experiment is provided for the active vehicle chassis control.The result shows that the performance of the actively controlled vehicle chassis using LQG controller optimized by the genetic algorithm such as the vertical acceleration of seat,the dynamic displacement of the seat suspension and the tire can be greatly improved,compared to the seat chassis controlled by the normal LQG and the passive one,.Meanwhile,the problem of defining the weight matrices is greatly solved based on the advantage of the normal LQG is sufficiently utilized.