结合数据变换和启发式方法的模糊系统辨识

取代应用原始数据样本构造模糊模型的传统方法,提出应用数据变换技术和启发式方法简化模糊建模过程.对于变换后的数据,首先通过启发式方法确定模糊If-Then规则结论部分非模糊单值(即实数)的初始值,然后通过梯度下降学习方法进行精调.该方法不仅模型精度较高,而且收敛速度快.仿真实例验证了所提出的模型优于传统的方法.     

Structured Learning and Decomposition of Fuzzy Models for Robotic Control Applications

In this paper a specially designed structured-optimization procedure is used for learning the parameters of the Takagi–Sugeno (TS) type fuzzy models. It is well-known that the number of learning parameters increases exponentially with the number of model inputs. Therefore an appropriate learning scheme with preliminary structuring of the learning parameters into two groups: antecedent parameters and consequent parameters can be helpful for speeding-up the learning process. Two different optimization algorithms for tuning the antecedent and consequent parameters respectively are used in a sequence of repetitive loops (epochs). The stop criterion is defined as a number of repetitions of the loops or as a desired minimal error. Random walk algorithm with variable step size is used in this paper for tuning the antecedent parameters of the membership functions. For tuning the consequent parameters of the singletons, a specially proposed local learning algorithm is used. The problem of dimensionality reduction in fuzzy modeling is also considered in the paper from another viewpoint, namely as a hierarchical fuzzy model structure. It is accomplished by a decomposition of the complete fuzzy model into a feedforward hierarchical structure of sub-models called partial fuzzy models each one with two inputs and one output. Then the local models are learned separately in a preliminary specified and repetitive order. Such decomposition scheme has a potential for a significant reduction of the number of model parameters to be tuned thus reducing the total learning time. It has been experimentally shown that both concepts for dimensionality reduction in learning fuzzy models have benefits in learning speed and accuracy. A comparison with simultaneous optimization of all parameters of a single fuzzy model is also given. It shows that the proposed structured learning as well as the decomposition of the fuzzy model into a hierarchical fuzzy model structure lead to reducing the learning time and creating more accurate fuzzy models. Finally an application for learning a fuzzy controller of a two-link robot motion is shown and analysed.     

Hybrid identification of fuzzy rule‐based models

In this study, we propose a hybrid identification algorithm for a class of fuzzy rule‐based systems. The rule‐based fuzzy modeling concerns structure optimization and parameter identification using the fuzzy inference methods and hybrid structure combined with two methods of optimization theories for nonlinear systems. Two types of inference methods of a fuzzy model concern a simplified and linear type of inference. The proposed hybrid optimal identification algorithm is carried out using a combination of genetic algorithms and an improved complex method. The genetic algorithms determine initial parameters of the membership function of the premise part of the fuzzy rules. In the sequel, the improved complex method (being in essence a powerful auto‐tuning algorithm) leads to fine‐tuning of the parameters of the respective membership functions. An aggregate performance index with a weighting factor is proposed in order to achieve a balance between performance of the fuzzy model obtained for the training and testing data. Numerical examples are included to evaluate the performance of the proposed model. They are also contrasted with the performance of the fuzzy models existing in the literature. © 2002 John Wiley & Sons, Inc.     

The Development of Incremental Models

In this study, we introduce and discuss a concept of an incremental granular model. In contrast to typical rule-based systems encountered in fuzzy modeling, the underlying principle exploited here is to consider a two-phase development of fuzzy models. First, we build a standard regression model which could be treated as a preliminary construct capturing the linear part of the data and in this way forming a backbone of the entire construct. Next, all modeling discrepancies are compensated by a collection of rules that become attached to the regions of the input space where the error is localized. The incremental model is constructed by building a collection of information granules through some specialized fuzzy clustering, called context-based (conditional) fuzzy C-means that is guided by the distribution of error of the linear part of the model. The architecture of the model is discussed along with the major algorithmic phases of its development. In particular, the issue of granularity of fuzzy sets of context and induced clusters is discussed vis-a-vis the performance of the model. Numeric studies concern some low-dimensional synthetic data and several datasets coming from the machine learning repository.     

A new approach to fuzzy modeling

This paper proposes a new approach to fuzzy modeling. The suggested fuzzy model can express a given unknown system with a few fuzzy rules as well as Takagi and Sugeno's model (1985), because it has the same structure as that of Takagi and Sugeno's model. It is also as easy to implement as Sugeno and Yasukawa's model (1993) because its identification mimics the simple identification procedure of Sugeno and Yasukawa's model. The suggested algorithm is composed of two steps: coarse tuning and fine tuning. In coarse tuning, fuzzy C-regression model (FCRM) clustering is used, which is a modified version of fuzzy C-means (FCM). In fine tuning, gradient descent algorithm is used to precisely adjust parameters of the fuzzy model instead of nonlinear optimization methods used in other models. Finally, some examples are given to demonstrate the validity of this algorithm     

Fuzzy force learning controller of flexible wiper system

Wiper blade of automobile is among those types of flexible system that is required to be operated in quite high velocity to be efficient in high load conditions. This causes some annoying noise and deteriorated vision for occupants. The modeling and control of vibration and low-frequency noise of an automobile wiper blade using soft computing techniques are focused in this study. The flexible vibration and noise model of wiper system are estimated using artificial intelligence system identification approach. A PD-type fuzzy logic controller and a PI-type fuzzy logic controller are combined in cascade with active force control (AFC)-based iterative learning (IL). A multi-objective genetic algorithm is also used to determine the scaling factors of the inputs and outputs of the PID-FLC as well as AFC-based IL gains. The results from the proposed controller namely fuzzy force learning (FFL) are compared with those of a conventional lead–lag-type controller and the wiper bang–bang input. Designing controllers based on classical methods could become tedious, especially for systems with high-order model. In contrast, FFL controller design requires only tuning of some scaling factors in the control loop and hence is much simpler and efficient than classical design methods.     

/spl epsiv/-insensitive fuzzy c-regression models: introduction to /spl epsiv/-insensitive fuzzy modeling

This paper introduces a new epsilon-insensitive fuzzy c-regression models (epsilonFCRM), that can be used in fuzzy modeling. To fit these regression models to real data, a weighted epsilon-insensitive loss function is used. The proposed method make it possible to exclude an intrinsic inconsistency of fuzzy modeling, where crisp loss function (usually quadratic) is used to match real data and the fuzzy model. The epsilon-insensitive fuzzy modeling is based on human thinking and learning. This method allows easy control of generalization ability and outliers robustness. This approach leads to c simultaneous quadratic programming problems with bound constraints and one linear equality constraint. To solve this problem, computationally efficient numerical method, called incremental learning, is proposed. Finally, examples are given to demonstrate the validity of introduced approach to fuzzy modeling.     

Design and tuning of standard additive model based fuzzy PID controllers for multivariable process systems.

This paper describes a design and two-level tuning method for fuzzy proportional-integral derivative (FPID) controllers for a multivariable process where the fuzzy inference uses the inference of standard additive model. The proposed method can be used for any n x n multi-input-multi-output process and guarantees closed-loop stability. In the two-level tuning scheme, the tuning follows two steps: low-level tuning followed by high-level tuning. The low-level tuning adjusts apparent linear gains, whereas the high-level tuning changes the nonlinearity in the normalized fuzzy output. In this paper, two types of FPID configurations are considered, and their performances are evaluated by using a real-time multizone temperature control problem having a 3 x 3 process system.     

FITSK: online local learning with generic fuzzy input Takagi-Sugeno-Kang fuzzy framework for nonlinear system estimation.

Existing Takagi-Sugeno-Kang (TSK) fuzzy models proposed in the literature attempt to optimize the global learning accuracy as well as to maintain the interpretability of the local models. Most of the proposed methods suffer from the use of offline learning algorithms to globally optimize this multi-criteria problem. Despite the ability to reach an optimal solution in terms of accuracy and interpretability, these offline methods are not suitably applicable to learning in adaptive or incremental systems. Furthermore, most of the learning methods in TSK-model are susceptible to the limitation of the curse-of-dimensionality. This paper attempts to study the criteria in the design of TSK-models. They are: 1) the interpretability of the local model; 2) the global accuracy; and 3) the system dimensionality issues. A generic framework is proposed to handle the different scenarios in this design problem. The framework is termed the generic fuzzy input Takagi-Sugeno-Kang fuzzy framework (FITSK). The FITSK framework is extensible to both the zero-order and the first-order FITSK models. A zero-order FITSK model is suitable for the learning of adaptive system, and the bias-variance of the system can be easily controlled through the degree of localization. On the other hand, a first-order FITSK model is able to achieve higher learning accuracy for nonlinear system estimation. A localized version of recursive least-squares algorithm is proposed for the parameter tuning of the first-order FITSK model. The local recursive least-squares is able to achieve a balance between interpretability and learning accuracy of a system, and possesses greater immunity to the curse-of-dimensionality. The learning algorithms for the FITSK models are online, and are readily applicable to adaptive system with fast convergence speed. Finally, a proposed guideline is discussed to handle the model selection of different FITSK models to tackle the multi-criteria design problem of applying the TSK-model. Extensive simulations were conducted using the proposed FITSK models and their learning algorithms; their performances are encouraging when benchmarked against other popular fuzzy systems.     

Gradient‐based back‐propagation dynamical iterative learning scheme for the neuro‐fuzzy inference system

In this paper, a gradient‐based back propagation dynamical iterative learning algorithm is proposed for structure optimization and parameter tuning of the neuro‐fuzzy system. Premise and consequent parameters of the neuro‐fuzzy model are initialized randomly and then tuned by the proposed iterative algorithm. The learning algorithm is based on the first order partial derivative of the output with respect to the structure parameters. The first order derivative of the model output with respect to the structure parameters determines the sensitivity of the model to structure parameters. The sensitivity values are then used to set the tuning factors and parameters updating step sizes. Therefore, an adaptive dynamical iterative scheme is achieved which adapts the learning procedure to the current state of the performance during the optimization process. Larger tuning step sizes make the convergence speed higher and vice versa. In this regard, this parameter is treated according to the calculated sensitivity of the model to the parameter. The proposed learning algorithm is compared with the least square back propagation method, genetic algorithm and chaotic genetic algorithm in the neuro‐fuzzy model structure optimization. Smaller mean square error and shorter learning time are sought in this paper, and the performance of the proposed learning algorithm is versified regarding these criteria.     

Rough Neural Network Based on Bottom-Up Fuzzy Rough Data Analysis

Based on bottom-up fuzzy rough data analysis, a new rough neural network decision-making model is proposed. Through supervised Gaustafason–Kessel (G–K) clustering algorithm, proper fuzzy clusters are found to partition the input data space. At the same time cluster number is searched by monotone increasing process. If the cluster number matches with that exactly exist in data sets then excellent fuzzy rough data modeling (FRDM) model can be built. And by integrating it with neural network technique, corresponding rough neural network is constructed. Our method overcomes the defects of conventional top-down based rough logic neural network (RLNN) method, and it also achieves adaptive learning ability and comprehensive soft decision-making ability compared with FRDM model. The experiment results indicate that our method has stronger generalization ability and more compact network structure than conventional RLNN.     

一种基于Rough Sets和模糊神经网络的规则获取的方法

该文提出了一种基于RoughSets思想获取初始规则,并通过模糊神经网络优化,最后再进行简化获取模糊规则,及模糊系统参数学习的方法。并通过实例进行了自动列车运行系统仿真。文中还基于上述实例,将这种基于模糊神经网络的学习与控制方法与标准的BP网络和基本的模糊系统方法进行了比较,并总结了这种方法的特点。结论表明,该文所提出的模糊规则生成和模糊系统学习方法是行之有效的。     

A systematic neuro-fuzzy modeling framework with application tomaterial property prediction

A systematic neural-fuzzy modeling framework that includes the initial fuzzy model self-generation, significant input selection, partition validation, parameter optimization, and rule-base simplification is proposed in this paper. In this framework, the structure identification and parameter optimization are carried out automatically and efficiently by the combined use of a sell-organization network, fuzzy clustering, adaptive back-propagation learning, and similarity analysis-based model simplification. The proposed neuro-fuzzy modeling approach has been used for nonlinear system identification and mechanical property prediction in hot-rolled steels from construct composition and microstructure data. Experimental studies demonstrate that the predicted mechanical properties have a good agreement with the measured data by using the elicited fuzzy model with a small number of rules.     

Designing neurofuzzy system based on ICART algorithm and its application for modeling jet fuel endpoint of hydrocracking process

In this paper, a neuro-fuzzy system based on improved CART algorithm (ICART) is presented, in which the ICART algorithm is used to design neuro-fuzzy system. It is worth noting that ICART algorithm partitions the input space into tree structure adaptively, which avoids the curse of dimensionality (number of rules goes up exponentially with number of input variables). Moreover, it adopts density function to construct the local model for every node in order to overcome the discontinuous boundaries existed in CART algorithm. In addition, a supervised scheme is used to adjust parameters to minimize the network output error and construct more accurate fuzzy model on the basis of the ICART algorithm. Finally, to illustrate the validity of the proposed method, a simulation research and a practical application are done. The results show that the proposed method can provide optimal model structure and parameters for fuzzy modeling, possesses high learning efficiency, and is smoother than CART algorithm. It can be successfully applied to modeling jet fuel endpoint of hydrocracking processing.     

A genetic-based neuro-fuzzy approach for modeling and control ofdynamical systems

Linguistic modeling of complex irregular systems constitutes the heart of many control and decision making systems, and fuzzy logic represents one of the most effective algorithms to build such linguistic models. In this paper, a linguistic (qualitative) modeling approach is proposed. The approach combines the merits of the fuzzy logic theory, neural networks, and genetic algorithms (GAs). The proposed model is presented in a fuzzy-neural network (FNN) form which can handle both quantitative (numerical) and qualitative (linguistic) knowledge. The learning algorithm of a FNN is composed of three phases. The first phase is used to find the initial membership functions of the fuzzy model. In the second phase, a new algorithm is developed and used to extract the linguistic-fuzzy rules. In the third phase, a multiresolutional dynamic genetic algorithm (MRD-GA) is proposed and used for optimized tuning of membership functions of the proposed model. Two well-known benchmarks are used to evaluate the performance of the proposed modeling approach, and compare it with other modeling approaches.     

Wavenet fuzzy PID controller for nonlinear MIMO systems: Experimental validation on a high-end haptic robotic interface

A novel global PID control scheme for nonlinear MIMO systems is proposed and implemented for a robot as study case, this scheme is called AWFPID from its adaptive wavelet fuzzy PID control structure. Basically, it identifies inverse error dynamics using a radial basis neural network with daughter RASP1 wavelets activation function; its output is in cascaded with an infinite impulse response (IIR) filter to prune irrelevant signals and nodes as well as to recover a canonical form. Then, online adaptive fuzzy tuning of a discrete PID regulator is proposed, whose closed-loop guarantees global regulation for nonlinear dynamical plants. The wavelet network includes a fuzzy inference system for online tuning of learning rates. A real-time experimental study on a three degrees of freedom haptic interface, the PHANToM Premium 1.0A, highlights the regulation with smooth control effort without using the mathematical model of the robot.     

基于矩阵的双论域模糊概率粗糙集增量更新算法

双论域模糊概率粗糙集是针对双论域信息系统的一种新的数据挖掘模型,现实应用中的双论域信息系统总是处于动态更新中,针对该问题提出一种基于矩阵的双论域模糊概率粗糙集增量式更新方法.首先,通过矩阵方法重新对双论域模糊概率粗糙集进行表示;然后,在矩阵表示模型的基础上,分别研究双论域信息系统两个论域中对象增加和减少时模型的增量式更...     

基于自适应神经网络的不确定非线性系统的模糊跟踪控制

提出了一种基于模糊模型和自适应神经网络的跟踪控制方法.在系统具有未知不确定非线性特性的情况下,首先利用T＿S模糊模型对系统的已知特性进行近似建模,对基于模糊模型的模糊H∞跟踪控制律进行输出跟踪控制.并在此基础上,进一步采用RBF神经网络完全自适应控制,通过在线自适应调整RBF神经网络的权重、函数中心和宽度,从而有效地消除系统的未知不确定性和模糊建模误差的影响,保证了非线性闭环系统的稳定性和系统的H∞跟踪性能,而不要求系统的不确定项和模糊建模误差满足任何匹配条件或约束.最后,将所提出的方法应用到一非线性混沌系统,仿真结果表明了所提出的方案不仅能够有效地稳定该混沌系统,而且能使系统输出跟踪期望输出.     

Fuzzy modeling based on generalized conjunction operations

An approach to fuzzy modeling based on the tuning of parametric conjunction operations is proposed. First, some methods for the construction of parametric generalized conjunction operations simpler than the known parametric classes of conjunctions are considered and discussed. Second, several examples of function approximation by fuzzy models, based on the tuning of the parameters of the new conjunction operations, are given and their approximation performances are compared with the approaches based on a tuning of membership functions and other approaches proposed in the literature. It is seen that the tuning of the conjunction operations can be used for obtaining fuzzy models with a sufficiently good performance when the tuning of membership functions is not possible or not desirable.     

基于改进模糊语法增量式算法的文本分类方法

针对现有很多文本分类算法必须进行训练-测试-再训练的缺点以及通用模型的语法表现度较差等问题,提出一种改进的模糊语法算法(IFGA)。首先根据一些选取的文本片段建立学习模型,为了适应轻微变化,采用增量式模型;然后将选取的文本片段转化到底层架构中,即模糊语法。最后利用模糊联合操作将单个文本片段语法进行结合,并将所学习的文本片段转化成更加一般的表示。与决策表算法、朴素贝叶斯等算法进行了两组对比实验,第一个实验结果表明IFGA和其他机器学生算法性能并无明显差异。第二个实验结果说明增量式学习算法比标准机器学习算法更加具有优势。其性能较平稳,数据的尺寸影响更小。另外,提出的算法具有较低的模型重新训练时间。