共查询到19条相似文献,搜索用时 109 毫秒
1.
2.
对不确定混沌系统控制问题, 研究了一种基于径向基函数神经网络(radial basis function neural network, RBFNN)的反馈补偿控制方法. 该方法首先用RBFNN对混沌系统的动力学特性进行学习, 然后用训练好的RBFNN模型对混沌系统进行反馈补偿控制. 该方法的特点是不需要被控混沌系统的数学模型,可以快速跟踪任意给定的参考信号. 数值仿真试验表明了该控制方法不仅具有响应速度快、控制精度高, 而且具有较强的抑制混沌系统参数摄动能力和抗干扰能力. 相似文献
3.
4.
5.
设计一种新型混合模糊神经推理系统,该系统仅从期望输入输出数据集即可达到获取知识、确定模糊初始规则基的目的.再利用神经网络学习能力便不难修改规则库中的模糊规则以及隶属函数和网络权值等参数,这样大大减少了规则匹配过程,加快了推理速度,从而极大程度地提高了系统的自适应能力.用它对Mackey-Glass混沌时间序列进行预测试验,结果表明利用该网络模型无论离线还是在线学习均能对Mackey-Glass混沌时间序列进行准确的预测,证明了该系统的有效性.
关键词:
神经网络模型
模糊逻辑
混合推理系统
混沌时间序列 相似文献
6.
提出一种模糊边界模块化神经网络(FBMNN)的混沌时间序列预测方法,该方法先对混沌时间序列观测点重构的相空间进行模块化划分,划分点的选取由遗传算法自动寻优.然后定义一个模糊隶属度函数,在划分边界一侧按照一定的模糊隶属度设定模糊边界带,通过模糊化处理,解决了各模块划分点附近预测结果的跳跃问题.最后每一模块,及其模糊边界的样本点都对应一个递归神经网络进行训练,通过预测合成模块输出结果.该方法对三个混沌时间序列基准数据集Mackey-Glass,Lorenz,Henon进行实验,结果表明该方法有效地提高了混沌时间序列预测效果.
关键词:
模糊边界
模块化神经网络
混沌时间序列
预测 相似文献
7.
8.
9.
10.
11.
In this paper we propose a new scheme to achieve chaos control and synchronization in Bragg acousto-optic bistable systems. In the scheme, we use the output of one system to drive two identical chaotic systems. Using the maximal conditional Lyapunov exponent (MCLE) as the criterion, we analyze the conditions for realizing chaos synchronization. Numerical calculation shows that the two identical systems in chaos with negative MCLEs and driven by a chaotic system can go into chaotic synchronization whether or not they were in chaos initially. The two systems can go into different periodic states from chaos following an inverse period-doubling bifurcation route as well when driven by a periodic system. 相似文献
12.
混沌控制与反混沌控制是一对逆问题. 通过研究系统状态变量的关联性, 分析了在电流型连续电流模式Boost变换器关联系数变化的情况下, 实现系统的混沌控制与反混沌控制的方法, 为实际应用打下理论基础. 建立了系统的离散数学模型, 利用单值矩阵理论解释了变换器混沌控制与反混沌控制的机理. 研究结果表明, 在只改变系统状态变量的关联系数的情况下, 该控制策略能够将处于任意状态的Boost变换器控制到周期1, 2, 4轨道以及混沌态, 系统的输出可实现混沌与反混沌控制. 仿真结果证明了所提出方法以及研究结果的正确性. 相似文献
13.
A Cournot duopoly, with a bounded inverse demand function and different constant marginal production costs, can be modeled as a discrete-time dynamical system, which exhibits complex bifurcating and chaotic behaviors. Based on some essential features of the model, we show how bifurcation and chaos can be controlled via the delayed feedback control method. We then propose and evaluate an adaptive parameter-tuning algorithm for control. In addition, we discuss possible economic implications of the chaos control strategies described in the paper. 相似文献
14.
针对不确定混沌系统控制问题, 研究了一种基于共轭梯度法(conjugate gradient algorithm, CGA)的多项式函数模型 (polynomial-basis-functions model, PBFM)的补偿控制方法. 该方法首先用PBFM对混沌系统的动力学特性进行拟合, 然后用拟合好的PBFM模型对不确定混沌系统进行前馈补偿控制. 该方法的特点是不需要被控混沌系统的数学模型, 可以快速跟踪任意给定的参考信号. 数值仿真试验表明了该方法不仅具有响应速度快、控制精度高, 而且具有较强的抑制混沌系统参数摄动能力和抗干扰能力.
关键词:
混沌控制
多项式函数模型
共轭梯度法 相似文献
15.
传统的混沌控制方法大多需要获知混沌系统的模型知识, 但是工业实际中系统的参数经常是未知的,与此同时系统建模过程当中经常会不可避免地存在未建模的动态不确定性, 这种情况下常规的混沌控制方法不能取得优化的控制性能指标.为解决此问题, 提出了一类基于无模型方法的混沌系统自适应控制算法.该算法基于数据驱动, 无需混沌系统的先验知识, 无需训练过程, 在线调整参数较少, 是一种低成本的控制器.数学证明了该控制系统的稳定性, 仿真结果说明了这种理论的有效性.
关键词:
混沌控制
自适应
无模型
数据驱动 相似文献
16.
In this Letter, a kind of novel model, called the generalized Takagi-Sugeno (T-S) fuzzy model, is first developed by extending the conventional T-S fuzzy model. Then, a simple but efficient method to control fractional order chaotic systems is proposed using the generalized T-S fuzzy model and adaptive adjustment mechanism (AAM). Sufficient conditions are derived to guarantee chaos control from the stability criterion of linear fractional order systems. The proposed approach offers a systematic design procedure for stabilizing a large class of fractional order chaotic systems from the literature about chaos research. The effectiveness of the approach is tested on fractional order Rössler system and fractional order Lorenz system. 相似文献
17.
18.
A NEW STRATEGY OF CHAOS CONTROL AND A UNIFIED MECHANISM FOR SEVERAL KINDS OF CHAOS CONTROL METHODS 下载免费PDF全文
Based on a general principle of physics that a physical system is in the most stable state if it is of the lowest energy stale, a new method for chaos control is proposed. A calculable generalized energy function in a nonlinear system is suggested for measuring control process, The Henon map and Lorenz system are taken as two typical examples to demonstrate the method. A series of stabilized periodic orbits as well aa inverse sequence of chaotic bands are obtained. At the same time, a unified mechanism of physics for several kinds of current cbaos control methods is studied using the idea proposed in this paper. 相似文献
19.
Junji Ohtsubo 《Optical Review》1999,6(1):1-15
Semiconductor laser with feedback is an excellent model for nonlinear optical system which shows chaotic dynamics. It is interesting not only from the fundamental physical study but also application standpoints of view. The dynamics of feedback induced instability and chaos, especially for optical feedback, and their applications are reviewed in this paper. The model of such a system is described by the laser rate equations. At first the dynamic behaviors of feedback induced instability and chaos in semiconductor lasers are discussed on the basis of the theory and experiments. Instability and chaos may be stabilized by the method of chaos control. Then we apply the method to suppress the noise induced by the feedback in a semiconductor laser. The synchronization of chaos between two similar systems is also an important issue in chaos applications and we discuss secure communications based on chaos synchronization. Some other examples of applications of feedback induced chaos are also described. 相似文献