混沌系统智能辨识与控制研究
发布时间:2018-08-01 12:26
【摘要】:混沌是自然界中广泛存在的一种复杂运动形式。近些年来,混沌与其它学科互相渗透,无论是在物理学、生物学、数学、电子学、心理学、信息科学,还是在气象学、经济学、天文学,甚至在音乐、艺术等领域都得到了广泛的应用。为了更好地利用混沌或者消除混沌的不良影响,辨识混沌系统的模型并实施相应的控制具有重要的意义。近些年来,以神经网络和模糊理论为代表的智能控制理论在复杂系统建模、控制方面得到了长足发展。本文采用智能控制理论研究了混沌系统的辨识与控制,具体研究工作如下:首先,提出了一种基于区间Ⅱ型模糊系统的混沌系统辨识方法。该方法采用网格对角线法来划分模糊空间,Ⅱ型模糊集主隶属度函数为对称三角形隶属函数。在保持前件参数不变的情况下,采用带遗忘因子的递推最小二乘法辨识结论参数。为了解决采样数据受到噪声污染的问题,对采样数据进行Sigmoid数据变换,并采用粒子群算法优化变换函数的关键参数和隶属函数宽度,避免了隶属函数的调整,提高了Ⅱ型模糊模型的辨识精度。此方法应用到Mackey-Glass混沌系统的建模中,仿真结果验证了本文方法的有效性。其次,利用混沌系统的部分结构信息,提出了一种基于Wiener-LSSVM模型的混沌系统辨识方法。Wiener-LSSVM模型由一个线性动态子系统和LSSVM组成,比较适合描述大部分的混沌系统。给出了同时辨识线性动态部分和最小二乘支持向量机的最小二乘算法。然后,提出了一种基于Hammerstein-ELM模型的混沌系统辨识方法。Hammerstein-ELM模型由一个极值学习机神经网络和一个线性动态部分组成。推导出了用于同时辨识ELM神经网络和线性动态子系统参数的广义极值学习算法。该算法采用矩阵伪逆确定辨识参数,提高了辨识的准确性。最后,基于模糊理论,提出了两种Hénon混沌系统的控制与同步算法。第一种方法采用T-S模型来辨识Hénon混沌系统,得到Hénon混沌系统的局部动态线性模型,基于此模型设计了模糊广义预测控制算法来实现Hénon混沌系统的跟踪与同步控制。第二种方法采用模糊逆方法建立Hénon混沌系统的模糊逆模型,基于此模糊逆模型设计了Hénon混沌系统的自适应逆控制与同步算法。仿真结果验证了所提方法的有效性。
[Abstract]:Chaos is a complex form of motion widely existing in nature. In recent years, chaos has interpenetrated with other disciplines, whether in physics, biology, mathematics, electronics, psychology, information science, meteorology, economics, astronomy, or even music. Art and other fields have been widely used. In order to make better use of chaos or eliminate the bad effects of chaos, it is of great significance to identify the model of chaotic system and implement the corresponding control. In recent years, intelligent control theory, represented by neural network and fuzzy theory, has made great progress in complex system modeling and control. In this paper, the identification and control of chaotic systems are studied by using intelligent control theory. The research work is as follows: firstly, a chaotic system identification method based on interval 鈪,
本文编号:2157574
[Abstract]:Chaos is a complex form of motion widely existing in nature. In recent years, chaos has interpenetrated with other disciplines, whether in physics, biology, mathematics, electronics, psychology, information science, meteorology, economics, astronomy, or even music. Art and other fields have been widely used. In order to make better use of chaos or eliminate the bad effects of chaos, it is of great significance to identify the model of chaotic system and implement the corresponding control. In recent years, intelligent control theory, represented by neural network and fuzzy theory, has made great progress in complex system modeling and control. In this paper, the identification and control of chaotic systems are studied by using intelligent control theory. The research work is as follows: firstly, a chaotic system identification method based on interval 鈪,
本文编号:2157574
本文链接:https://www.wllwen.com/shoufeilunwen/jckxbs/2157574.html