【作者】 徐林；

【导师】 张静；

【作者基本信息】 哈尔滨理工大学 ， 模式识别与智能系统， 2009， 硕士

【摘要】 磁悬浮系统是学习和研究控制理论的重要平台之一。对磁悬浮系统的研究可以归结为对非线性系统和不稳定系统的研究,这类复杂控制对象难以用数学公式进行精确描述,采用经典控制方法也难取得好的控制效果。因此,对于磁悬浮系统,获取不依赖于数学模型的控制方法具有十分重要的意义。本文基于德国Amira公司设计的MA401型磁悬浮实验设备,首先介绍单自由度磁悬浮系统的构成及工作原理,详细分析了系统的动态模型,在线性化基础上对系统进行定性分析,指出磁悬浮系统为本质不稳定对象,并应用奇异摄动法建立了磁悬浮系统的非线性数学模型。其次,针对磁悬浮系统难以建立精确数学模型的特性,构造了一个二维模糊控制器,加入积分环节消除误差,实验结果表明系统具有很强的鲁棒性。最后,在MATLAB环境下利用S-函数建立磁悬浮的非线性动态模型。针对系统的开环不稳定和强烈非线性的特性,在基本模糊控制和常规PID控制的基础上提出模糊自适应PID(FAPID)控制策略,虽然改善了系统的动态性能,却造成了稳态误差的增大。为此文中提出了一种改进的模糊自适应PID控制方法(IFAPID),将微分环节从模糊控制器中剥离出来,只对比例和积分参数进行模糊整定,并针对磁悬浮系统的强烈非线性,在模糊控制器中采用非线性模糊化。仿真结果表明,IFAPID控制系统的抗干扰和适应参数变化的能力都优于PID和FAPID控制方法,并具有更好的动态特性和稳定性。更多还原

【Abstract】 The magnetic levitation system is one of the most important platforms to study and research control theory.The research on the magnetic levitation system can be attributed to the research on the nonlinear and unstable system. This kind of complicated control object is difficult to describe precisely with mathematic formula, and is very difficult to obtain the better effect using the classical control methods. Therefore, it’s great significant to obtain controlling means without depending on system mathematical model.Based on the magnetic levitation equipment MA401 that is made in Germany Amira Company, firstly, the structure and working principle of the single degree freedom magnetic levitation system are introduced, and it’s dynamic model is analyzed in detail. The qualitative analysis to the magnetic levitation system based on linearization is carried on, it can be concluded that the magnetic levitation system is an unstable object. And the nonlinear mathematical model of the magnetic levitation system is established, where the singular perturbation method is used.Next, considering the difficulty of giving the exact mathematical model of magnetic levitation, a two-dimension fuzzy controller adding a integration element to avoid the error is constructed and the result shows the magnetic levitation has strong robustness.Finally, the nonlinearity dynamic model of maglev system is programmed by S-function in MATLAB. Based on both the advantages of PID controller and fuzzy controller, a fuzzy adaptive PID controller(FAPID) is proposed for the characteristics of magnetic levitation system, such as open-loop instability and strong nonlinearity, which the FAPID controller aggrandize the error of stability by simulation, though it’s dynamic property is improved. Therefore, the algorithm of the improved fuzzy adaptive PID controller(IFAPID) is advanced and designed that by parted D controller from PID controller, the problem of strong nonlinearity of magnetic levitation system are solved by using nonlinear fuzzy mapping in the fuzzy controller, and the PID controller’s P and I parameters are adapted by fuzzy controller. The results showed that the IFAPID controller has the advantages of higher anti-interference ability and adaptability to parameters’ changing than conventional PID controller and FAPID, and better dynamic property and stability.更多还原