复杂系统的递阶模糊辨识

针对Takagi_Sugeno模糊模型 (T_S模型 )严重的维数灾问题, 借鉴GMDH算法, 提出了一种新的复杂系统递阶模糊辨识方法. 本文首先详细描述了由两输入变量的特殊T_S模型所组成的递阶模糊模型 ;然后提出了具体的辨识该递阶模糊模型的方法. 该方法的特点是 :a)在结构辨识阶段, 用FCM模糊聚类方法评价系统中每个输入变量的重要性, 以便构造合理的递阶模糊模型 ;b)预先合理地确定了所要辨识的参数的初始值, 用扩展卡尔曼滤波方法可很快地得到这些参数. 最后, 给出的仿真实例说明了本文辨识方法的有     

平面二级倒立摆的圆周行走与镇定控制

采用拉格朗日方程建立了平面二级倒立摆的非线性动力学模型, 将其在平衡位置线性化, 得到系统在两个正交控制方向解耦的近似模型. 针对每一个方向上由互相耦合的基座小车定位子系统和摆杆镇定子系统所组成的六阶欠驱动系统, 设计了自适应滑模模糊控制器, 实现了基座小车沿圆周行走条件下摆杆的镇定控制, 验证了该控制算法对欠驱动、不稳定、多变量耦合系统的有效性.     

不确定非方广义分数阶T–S模糊系统的鲁棒镇定

针对不确定非方广义分数阶T–S模糊系统阶数为0<α<1的鲁棒镇定问题,提出了一个有效的判据.首先,利用一个新的T–S模糊动态补偿器,将不确定非方广义分数阶T–S模糊系统转化为一个增广的不确定方形广义分数阶T–S模糊系统.由于增广变量的引入,动态补偿器的设计问题可以等价为静态输出反馈控制器的设计问题.其次,设计一个分数阶导数反馈控制器对得到的增广系统进行正常化处理.然后,对正常化得到的不确定分数阶T–S模糊系统进行研究,得到一个系统渐近稳定的充分条件.该条件保守性小并且形式简洁.最后,通过一个数值算例和一个实际例子验证了本文所提出结论的正确性和有效性.     

基于递阶遗传算法的模糊控制系统的模糊集优化研究

由于模糊系统的输入、输出变量模糊集中模糊集的多少直接影响到模糊规则集的完备程度，并决定着控制系统的性能，所以本文提出了一种基于递阶遗传算法的模糊控制系统的模糊集优化方法，同时对模糊集的个数和形状进行优化，使其能尽量完美地反映系统输入、输出变量，减少了规则的冗余．对二阶动态系统的仿真研究证明了该方法的有效性．     

空间分布系统的空间模糊建模

本文对空间分布系统提出了一种新的模糊建模方法.首先,在3-D模糊集的基础上给出一种改进的空间模糊集,包含了传统的模糊集和添加的一维空间信息.在空间轴方向上,将传统模糊隶属度函数沿输入变量的随空间的物理变化曲线进行扩展,通过隶属度的连续变化描述输入变量在空间中的变化.其次,基于空间模糊集,采用Mamdani模糊模型形式,设计了对空间分布系统的空间模糊模型的建模方法.最后,通过仿真算例对方法进行了验证.     

基于模糊动态模型的多变量系统模糊控制

采用模糊动态模型对多变量复杂非线性系统进行模糊控制.首先针对局部线性动态模 型设计状态反馈控制器,然后利用模糊推理确定整个系统的控制;在一系列局部模型通过模糊 隶属函数连接得到的连续的全局模型中,全面考虑其它关联子系统对标称线性系统的摄动,并 利用大系统分散控制关联稳定性的概念和方法,得到了闭环模糊系统稳定的充分条件.仿真例 子验证了该设计方法的正确性.     

采用正交小波网络的非线性系统辨识方法

基于多分辨分析的递阶思想,采用正交小波网络研究了输入样本空间分布非均匀时非线性系统的辨识问题,重点讨论了样本非均匀时网络系的设计问题,并给出了基于该网络系的在线递阶辨识算法,最后利用正交小波网络分别对非线性静态和动态系统进行了仿真辨识。     

考虑重要输入变量选择的非线性系统模糊辨识

针对有数百个可能输入的复杂非线性动态系统模糊建模问题,本文提出一种新的考虑重要输入变量选择的模糊辨识方法.首先采用两阶段模糊曲线方法(TSFC)从大量可选择的输入变量中给出各输入变量与输出之间的关联度权重,根据输入变量指标快速选择出重要的输入变量,然后采用模糊聚类(FCM)和高斯(Gaussian)型隶属函数确定模糊模型前提参数,采用递推最小二乘(RLS)辨识出模糊模型结论参数.最后通过对Mackey-Glass混沌时间序列和Box-Jenkins煤气炉数据两个国际标准例题模糊建模验证了该方法的有效性,并将该方法应用到一个实际气动变载荷加载系统的模糊建模中,验证了该方法的实用性.     

非线性系统辨识方法的新进展

对现有的非线性系统辨识方法进行了简要综述。介绍了多层递阶辨识方法，以及把神经网络、模糊逻辑、遗传算法等知识应用于非线性系统辨识而得到的一些新型辨识方法，最后慨括了非线性系统辨识未来的发展方向。     

一类模糊系统的多层递阶预报

本文讨论一类模型系统的多层递阶预报问题，提出了相应的模糊多层递阶模型，模糊参数估计方法及预报步骤。通过应用实例说明效果良好。     

Exact analytical inverse mapping of decomposable TS fuzzy systems with singleton and linear consequents

In this study, a new method is proposed for the exact analytical inverse mapping of Takagi–Sugeno fuzzy systems with singleton and linear consequents where the input variables are described by using strong triangular partitions. These fuzzy systems can be decomposed into several fuzzy subsystems. The output of the fuzzy subsystem results in multi-linear form in singleton consequent case or multi-variate second order polynomial form in linear consequent case. Since there exist explicit analytical formulas for the solutions of first and second order equations, the exact analytical inverse solutions can be obtained for decomposable Takagi–Sugeno fuzzy systems with singleton and linear consequents. In the proposed method, the output of the fuzzy subsystem is represented by using the matrix multiplication form. The parametric inverse definition of the fuzzy subsystem is obtained by using appropriate matrix partitioning with respect to the inversion variable. The inverse mapping of each fuzzy subsystem can then easily be calculated by substituting appropriate parameters of the fuzzy subsystem into this parametric inverse definition. So, it becomes very easy to find the analytical inverse mapping of the overall Takagi–Sugeno fuzzy system by composing inverse mappings of all fuzzy subsystems. The exactness and the effectiveness of the proposed inversion method are demonstrated on trajectory tracking problems by simulations.     

Second-order terminal sliding mode control of uncertain multivariable systems

A second-order terminal sliding mode controller for uncertain multivariable systems is proposed in this paper. The controller adopts the hierarchical control structure. The paper derives the state transform matrices which are used to transform a multivariable linear system to the block controllable form consisting of two subsystems, an input–output subsystem and a stable internal dynamic subsystem. The proposed controller utilizes a non-singular terminal sliding mode manifold for the input–output subsystem to realize fast convergence and better tracking precision. Meanwhile, a chattering-free second-order terminal sliding mode control law is presented. The stability of uncertain multivariable systems can be realized using the proposed controller. A derivative estimator is utilized in the paper to estimate the derivatives of the sliding mode functions for the controller. The simulation results are presented to validate the design method.     

Systematic design method of stabilization fuzzy controllers for pendulum systems

A systematic method to construct stabilization fuzzy controllers for a single pendulum system and a series-type double pendulum system is presented based on the single input rule modules (SIRMs) dynamically connected fuzzy inference model. The angle and angular velocity of each pendulum and the position and velocity of the cart are selected as the input items. Each input item is given with a SIRM and a dynamic importance degree. All the SIRMs have the same rule setting. The dynamic importance degrees use the absolute value(s) of the angle(s) of the pendulum(s) as the antecedent variable(s). The dynamic importance degrees are designed such that the upper pendulum angular control takes the highest priority and the cart position control takes the lowest priority when the upper pendulum is not balanced upright. The control priority orders are automatically adjusted according to control situations. The simulation results show that the proposed fuzzy controllers have high generalization ability to completely stabilize a wide range of single pendulum systems and series-type double pendulum systems in short time. By extending the architecture, a stabilization fuzzy controller for a series-type triple pendulum system is even possible. © 2001 John Wiley & Sons, Inc.     

多级模糊控制的月球车避障研究

针对月球车这个MIMO系统,考虑其在未知环境下的避障问题,设计了一个多级模糊控制器,将输入变量分组设计子模糊作为第1级,再将第1级的输出作为第2级的输入变量设计第2级子模糊系统,最后控制系统输出平移速度和角速度。这样就避免了常规模糊系统中规则数目过大和设计繁琐问题,提高了整个控制系统的性能。该控制器模糊规则数量少,控制方法简单,响应速度快。最后的实验证明了该方法的有效性。     

一类离散非线性不确定互联系统的模糊分散控制

利用模糊控制方法研究一类离散非线性互联系统的分散控制问题.首先采用模糊(T-S)模型对离散非线性不确定互联系统进行模糊建模,应用并行分布补偿算法(PDC)给出状态反馈分散模糊控制方案,并基于李亚普诺夫函数方法证明了闭环系统的稳定性.然后当系统的状态不完全可测时,设计模糊分散观测器来估计各子系统的状态,从而给出基于观测器的状态反馈分散模糊控制设计的方法.因为该分散模糊控制设计问题是以线性矩阵不等式的形式给出,所以很容易用凸优化方法求解.仿真结果验证了所提出控制方法的有效性.     

Subsystem inference representation for fuzzy systems based uponproduct-sum-gravity rule

A new representation which expresses a product-sum-gravity (PSG) inference in terms of additive and multiplicative subsystem inferences of single variable is proposed. The representation yields additional insight into the structure of a fuzzy system and produces an approximate functional characterization of its inferred output. The form of the approximating function is dictated by the choice or polynomial, sinusoidal, or other designs of subsystem inferences. With polynomial inferences, the inferred output approximates a polynomial function the order of which is dependent on the numbers of input membership functions. Explicit expressions for the function and corresponding error of approximation are readily obtained for analysis. Subsystem inferences emulating sinusoidal functions are also discussed. With proper scaling, they produce a set of orthonormal subsystem inferences. The orthonormal set points to a possible “modal” analysis of fuzzy inference and yields solution to an additive decomposable approximation problem. This work also shows that, as the numbers of input membership functions become large, a fuzzy system with PSG inference would converge toward polynomial or Fourier series expansions. The result suggests a new framework to consider fuzzy systems as universal approximators     

广义递阶Mamdani模糊系统及其泛逼近性

从解决模糊系统的“规则爆炸”问题出发,本文首先给出广义递阶M amdan i模糊系统的定义,然后证明其与具有中间变量的广义M amdan i模糊系统等价,并借助方形分片线性函数构造性的证明了在最大模和积分模意义下该系统是泛逼近器.最后仿真实例证实了该系统的有效性.     

H∞ decentralized fuzzy model reference tracking controldesign for nonlinear interconnected systems

In general, due to the interactions among subsystems, it is difficult to design an H_∞ decentralized controller for nonlinear interconnected systems. The model reference tracking control problem of nonlinear interconnected systems is studied via H_∞ decentralized fuzzy control method. First, the nonlinear interconnected system is represented by an equivalent Takagi-Sugeno type fuzzy model. A state feedback decentralized fuzzy control scheme is developed to override the external disturbances such that the H∞ model reference tracking performance is achieved. Furthermore, the stability of the nonlinear interconnected systems is also guaranteed. If states are not all available, a decentralized fuzzy observer is proposed to estimate the states of each subsystem for decentralized control. Consequently, a fuzzy observer-based state feedback decentralized fuzzy controller is proposed to solve the H_∞ tracking control design problem for nonlinear interconnected systems. The problem of H _∞ decentralized fuzzy tracking control design for nonlinear interconnected systems is characterized in terms of solving an eigenvalue problem (EVP). The EVP can be solved very efficiently using convex optimization techniques. Finally, simulation examples are given to illustrate the tracking performance of the proposed methods     

非匹配不确定系统的终端滑模分解控制

针对非匹配多变量模型不确定系统,提出了一种终端滑模分解控制方法.通过状态变换和去耦合处理将系统转换为块能控标准型,它由匹配扰动的值域空间子系统和稳定的非匹配扰动的零动态子系统组成.提出了特殊的终端滑模超曲面,采用滑模控制策略,使值域空间子系统的状态在有限时间内收敛至平衡点,随后非匹配扰动的零动态子系统渐近收敛至平衡点附近的邻域内,且建立了该邻域的范围与系统的非匹配不确定性范围之间的数学关系,并用于系统的设计与分析.所提方法对于维数较高的非匹配不确定系统的控制具有较大的意义,可简化设计,实现递阶控制.仿真实