主动电磁轴承μ控制器的多目标优化设计

Multi-objective Optimal Design of m-controller for Active Magnetic Bearing

提出一种用于主动电磁轴承μ控制器的多目标优化设计。由于用电磁轴承支撑的转子系统其固有的负刚度本质是不稳定的,需要采用闭环控制才可以稳定运行,因此环路中控制器的设计成为重要环节。实际中,电磁轴承的动态控制模型需要在平衡点附近作线性化,因此,由于运行过程中状态的不同,会引起位移刚度和电流刚度的变化等不确定性。为考虑其影响,采用不确定参数摄动法建立电磁轴承转子模型,然后采用多目标蚁群算法进行6-DK迭代来综合μ控制器。优化结果保证闭环系统在给定不确定度时有2.57倍的稳定裕度;仿真和实验也表明所提控制器的响应性能相对于PID有明显的提升,具有更强的干扰抑制能力,该电磁轴承转子系统具有优越的内部和外部稳定性。

A multi-objective optimized design ofm-controller for active magnetic bearing is presented.Rotors suspended with electromagnetic bearings are inherently unstable and closed-loop control is an essential part of their stable operation.Therefore,the design of the controller in the loop becomes an important step.Actually,the actuator dynamics need to be linearized around an operating point.Hence,due to different states during the operation,the changing model parameters bring about uncertainties such as the changing displacement stiffness and current stiffness of the electromagnetic bearing.In order to consider the influence of these uncertainties,the uncertain parameter perturbation method is used to model the system,and then a multi-objective ant colony algorithm is used to perform 6-DK iterations to synthesize them-controller.The optimization results ensure that the closed-loop system has a stability margin of 2.57 times at given uncertainties.Simulations and experiments also show that the response performance of the controller proposed in this paper is significantly improved compared with PID,and it has stronger disturbance rejection;thus the rotor system has excellent internal and external stability.