磁悬浮开关磁阻电机的迭代学习控制

Iterative Learning Control of Bearingless Switched Reluctance Motor

对磁悬浮开关磁阻电机进行控制时,往往需要进行复杂的解耦和反馈线性化步骤,如何避免这些繁琐的设计过程还无有效的方法。文章利用以往时刻对当前控制量的影响程度提出了一种具有影响函数的迭代学习控制器,为此首先针对目前现有的磁悬浮开关系统进行了数学模型的修正及扩展,进而通过分析了迭代学习算法存在的问题,设计了一种具有影响函数的迭代学习控制,最后对磁悬浮开关磁阻电机启动状态进行了实验控制,表明算法在收敛速度以及高精度跟踪性能方面的良好优势。

The decoupling or feedback linearization is always used to the bearingless switched reluctance motor（BSRM） controlling system.Unfortunately there is no effective way to avoid this fussy process.So a kind of iterative learning control（ILC）with influence function was proposed,which used the influence of the previously controlling information adding to the currently controlling.The model of the BSRM existing was amended and extended at first.After analyzing problems of ILC arithmetic,the iterative learning control with influence function（IFILC）was used in the BSRM system.The results show good performances of the controller,and the IFILC has faster convergence rate than ILC.