基于GA和模糊小波神经网络的磁轴承控制方法  

A Magnetic Bearing Control Method Based on Genetic Algorithm and Fuzzy-Wavelet Neural Network

作　　者：彭辉[1] 曾碧[2] 

机构地区：[1]广西工业职业技术学院,南宁530001 [2]广东工业大学,广州510006

出　　处：《微特电机》2017年第4期77-81,92,共6页Small & Special Electrical Machines

基　　金：广西教育厅课题项目(KY2015YB440);广西工业职业技术学院课题项目(桂工业院[2014]56号)

摘　　要：针对常规PID技术不能满足磁轴承(MB)的鲁棒性控制要求,提出一种基于模糊小波神经网络(FWNN)的磁轴承控制方法。将差分驱动模式的MB系统构建成一个非线性动态模型,在模糊神经网络(FNN)中嵌入小波神经网络(WNN)形成F-WNN,并通过监督学习机制对F-WNN的参数进行在线学习,同时利用遗传算法优化学习率因子。最后,以转子位置误差作为F-WNN的输入,以控制电压作为输出,实现转子的自适应控制。仿真结果表明,该方法在存在干扰情况下能够精确控制MB转子位置,具有较高的实际应用价值。For the issue that the conventional PID technology can＇t meet the requirements of the magnetic bearing （MB） robustness control, a magnetic bearing control method based on fuzzy-wavelet neural network （F-WNN） is pro- posed. Firstly, a nonlinear dynamic model was built for the MB system with differential drive mode. Then, the wavelet neural network （WNN） was embedded in the fuzzy neural network （FNN） to form F-WNN, and the supervised learning mechanism was used for online learning the F-WNN parameters. At the same time, the genetic algorithm （GA） was adopt- ed to optimize the learning rate factor. Finally, the rotor position error is regard as the input of the F-WNN, and the con- trol voltage as the output, realized adaptive control of the rotor. Experimental results show that the proposed method can ac- curately control the position of MB rotor in the presence of interference, and has high practical application value.

关 键 词：磁轴承 转子位置控制 模糊小波神经网络 遗传算法 

分 类 号：TH133.3[机械工程—机械制造及自动化]