中文摘要：

实时精确的位移信息是磁轴承稳定悬浮的前提,位移传感器降低了系统性能,同时增加系统复杂性和成本。研究了一种基于自适应遗传优化支持向量机的三自由度混合磁轴承转子位移智能自检测方法。通过对该磁轴承结构和原理的分析,基于变刚度系数,构建了悬浮力模型;在此基础上,利用最小二乘支持向量机小样本学习特点、通用逼近能力,通过输入输出变量确定和有效样本数据采集,训练得到磁轴承位移自检测模型;针对支持向量机模型参数选取问题,引入自适应遗传算法进行自动寻优;为验证算法的有效性,引入均方误差和绝对误差作为性能指标对模型进行评价;最后通过位移自检测控制仿真和实验研究验证了所提方法具有较高的检测精度,可为磁轴承悬浮控制提供准确的位移信息。

英文摘要：

Real-time accurate displacement is the premise of magnetic bearing stable suspension. Traditional displacement sensors reduce the system performance and increase the complexity and cost. An intelligent self-sensing method for a 3-degree-of-freedom hybrid magnetic bearing （3-DOF-HMB） based on support vector machine （SVM） optimized by adaptive genetic algorithm was presented. First, the structure and working principle of 3-DOF-HMB were explained and the accurate mathematical model was derived with variable stiffness coefficient. Then, the self-sensing model of 3-DOF-HMB was obtained by training SVM with determined input/output variables and representative sample data. Besides, to improve the performance of the self-sensing model, the adaptive genetic algorithm was used to optimize parameters of SVM, and the mean squared error and absolute error was introduced as a performance index to evaluate the model. Finally, the displacement self-sensing control test was designed. The findings show that the proposed strategy is suitable for displacement selfsensing of 3-DOF-HMB with high precision.