磁悬浮主轴—转子系统LQR控制优化研究

摘要

In order to improve the control performance of the magnetic motorized spindle-rotor system, the dynamic characteristics of the active magnetic bearing-rotor system on the spindle are investigated from a nonlinear point of view. In view of the cyclical influence of unbalance force caused by rotor eccentricity, a 2-DOF magnetic motorized spindle-rotor dynamics model is established. Through model simplification and linear approximation processing, the system state space model is constructed. LQR controller is proposed as the system state feedback control method, and MPGA is designed to optimize the weight matrix, solving the immature convergence of the SGA to a large extent as well as enhancing the global search ability of GA. The simulation results in Simulink show that the algorithm converges to the optimal value faster, the re-sponse time is reduced by 58% compared with the SGA, and the LQR control system has better dynamic stability.%为提高磁悬浮主轴—转子系统运行状态的控制性能,从非线性角度对电主轴上主动磁力轴承—转子系统的动力学特性进行了研究.在考虑转子偏心造成的不平衡力周期性影响的情况下,建立了2自由度的磁悬浮主轴—转子动力学模型.通过模型简化与线性化近似处理构建系统状态空间模型,选择LQR控制器作为系统全状态反馈控制方法,设计出多种群遗传算法来实现对加权矩阵的优化设计,很大程度上解决了标准遗传算法未成熟收敛问题,增强了遗传算法全局搜索能力.在Simulink中的仿真实验结果表明:该算法收敛至最优值速度更快,响应时间较标准遗传算法减少58% ,同时LQR控制系统具有更好的动态稳定性.