模糊人工神经网络方法在QSAR研究中的应用

使用模糊神经网络提取易于理解的“IF－THEN”模糊规则，并用于亚苄丙二腈类衍生物活性的预测，结果较好。     

基于模块化模糊神经网络的非线性系统故障诊断

提出了一种基于模块化模糊神经网络的非线性系统故障诊断新方法。该方法先使用模糊c-均值聚类法对测量空间进行模块分割,再利用模糊IF－THEN规则对分割后的子空间分别采用局部BP模型进行逼近,最后,通过离线学习获得不同子空间故障输出与测量输入的非线性动力特性。试验表明该网络具有良好的泛化性能,可显著提高非线性系统故障检测的快速性、鲁棒性及准确率。     

一种基于动态聚类的模糊分类规则的生成方法

介绍了一种基于动态聚类的模糊分类规则的生成方法，这种方法能决定规则数目，隶属函数的位置及形状．首先，介绍了基于超圆雏体隶属函数的模糊分类规则的基本形式；然后，介绍动态聚类算法，该算法能将每一类训练模式动态的分为成簇，对于每簇，则建立一个模糊规则；通过调整隶属函数的斜度，来提高对训练模式分类识别率，达到对模糊分类规则进行优化调整的目的；用两个典型的数据集评测了这篇文章研究的方法，这种方法构成的分类系统在识别率与多层神经网络分类器相当，但训练时间远少于多层神经网络分类器的训练时间．     

一种改进型T-S模糊神经网络

对T-S模糊神经网络进行了分析,提出了一种新型T-S模糊神经网络,改进了前件网络的结构及学习算法,减少了模糊规则层的节点数,有效地克服了T-S模糊神经网络模糊规则冗余的缺点。这种新型T-S模糊神经网络具有学习算法简单、收敛速度快等优点。把该网络应用到卷取温度控制中进行仿真,得到了满意的结果。     

图象压缩的模糊竞争矢量量化方法

在分析神经网络竞争学习算法和模糊Ｃ均值算法的基础上，提出了模糊竞争学习算法，并对模糊隶属度函数进行了探讨。理论分析和实验结果表明，模糊竞争学习算法用于图象矢量量化压缩编码是一种非常有效的方法。     

一种基于椭圆区域的进化式模糊分类系统

介绍了一种进化式模糊分类系统.首先,介绍系统的基本特征及结构框架.然后,介绍了一种动态聚类算法,并运用动态聚类算法对输入的训练模式进行动态聚类,每一簇创建一条模糊规则.规则所对应的区域为类椭圆形区域.规则调整的策略是连续改变模糊分类规则的一个参数,使得分类系统对训练模式识别率不能再提高,对不能达到要求的调整,采用遗传算法进行调整.分析了规则调整的方法,给出了调整算法,也介绍了规则的插入和聚合策略.用两个典型的数据集来评测研究的系统,研究的分类系统在识别率与多层神经网络分类器相当,但训练时间远少于多层神经网络分类器的训练时间.     

一种基于模糊规则的分类方法

本文提出了一种基于模糊规则的分类方法。首先介绍了一种新的模糊规则提取方法，然后基于所提取的模糊规则给出了一个采用二级判决的分类算法，并利用IRIS数据对此分类算法进行了仿真测试。结果表明，该算法在训练样本较少的情况下，仍能得到很好的分类效果.     

基于递阶遗传算法模糊加权神经网络的模糊规则自动获取

针对模糊规则的自动获取一直是模糊系统的一个瓶颈问题,提出一种基于递阶结构的混合编码遗传算法与进化规划相结合的模糊加权神经网络学习新算法,利用该算法同时优化模糊加权神经网络的结构和参数,最后说明了从网络中提取模糊规则的方法,从而自动获得最优的模糊规则。分析和实验结果表明,本文方法在规则提取和分类准确性等方面比其他方法更好。     

基于神经网络的自适应模糊控制器

本文提出了一种基于神经网络的自适应模糊控制器，控制器为５层前向结构，其输入和输出均为数值量。根据给定的训练数据，通过学习算法，能够实现前件参数和后件参数的辨识，提取控制规则，最后通过仿真实验证明了这种方法的有效性。     

一种基于模糊规则的神经网络结构及其学习算法研究

文章提出了一种基于模糊规则的神经网络结构,并用形式化语言进行描述。基于模糊规则的神经网络由输入层、规则层和输出层三层网络结构组成,以隶属度函数（语义值）作为网络权值,输入值沿权值的传播即进行隶属度计算。在充分分析三角形函数特征的基础上,应用启发式方法,导出了FRBNN网络的学习算法。最后应用FRBNN评价船舶碰撞危险度,表明FRBNN兼备神经网络和模糊推理系统的优点。     

The chaining syllogism in fuzzy logic

We analyze the validity of the chaining syllogism in fuzzy systems, i.e., whether two fuzzy rules IF F, THEN G, and IF G, THEN H imply the rule IF F, THEN H. Conditions are given under which this basic deduction scheme holds. "If A is predicated of all B, and B of all C, A must necessarily be predicated of all C." ;-The chaining syllogism according to Aristotle's Prior Analytics.     

基于熵聚类模糊神经网络味觉信号识别系统的研究

提出了一种基于熵聚类的模糊神经网络味觉信号识别系统模型，该模型利用聚类方法实现模糊输入空间划分和模糊IF-THEN规则提取，并使用梯度下降法对系统参数进行精炼，系统兼具有良好的可解释性和学习能力，对11种矿泉水味觉信号的识别实验结果表明了该系统的可行性和有效性。     

Logical radial basis function networks a hybrid intelligent model for function approximation

The aim of this article is to introduce a new approach for fuzzy neural network models which can be used effectively in function approximation problems. The proposed model is introduced as an adaptive two-level fuzzy inference system. The architecture of the model is basically a two-layer network of new types of fuzzy-neurons which act as fuzzy IF–THEN rules. The model can be considered as a logical version of the Radial Basis Function networks (RBF). Genetic Algorithms have been adopted as the learning mechanism of the proposed model. Simulations show both highly nonlinear mapping and reasoning capabilities together with simpler structure and better performance when compared with classical neural networks.     

Application of an artificial immune system-based fuzzy neural network to a RFID-based positioning system

Due to the rapid development of globalization, which makes supply chain management more complicated, more companies are applying radio frequency identification (RFID), in warehouse management. The obvious advantages of RFID are its ability to scan at high-speed, its penetration and memory. In addition to recycling, use of a RFID system can also reduce business costs, by indentifying the position of goods and picking carts. This study proposes an artificial immune system (AIS)-based fuzzy neural network (FNN), to learn the relationship between the RFID signals and the picking cart’s position. Since the proposed network has the merits of both AIS and FNN, it is able to avoid falling into the local optimum and possesses a learning capability. The results of the evaluation of the model show that the proposed AIS-based FNN really can predict the picking cart position more precisely than conventional FNN and, unlike an artificial neural network, it is much easier to interpret the training results, since they are in the form of fuzzy IF–THEN rules.     

一种自学习模糊神经网络多变量自适应控制器

本文在将文（６）的参数学习算法推广到多变量系统和对爬山法加以改进的基础上,提出了一种新的基于Ｐｉ－ｓｉｇｍａ混合型自适应模糊神经网络的多变量的自适应控制器。该控制器能在不需过多先验知识的情况下在线自学前件和后件参数,仿真结果表明,该控制器有效的。     

Fuzzy Identification of Nonlinear Systems via Orthogonal Transform

This paper is concerned with the application of orthogonal transforms and fuzzy competitive learning to extract fuzzy rules from data. The least square algorithm with orthogonal transforms is proposed to supervise the progress of fuzzy competitive learning. First of all, competitive learning takes place in the product space of system inputs and outputs and each cluster corresponds to a fuzzy IF–THEN rule. The fuzzy relation matrix, confirmed by fuzzy competitive learning, is studied by the orthogonal least square algorithm. The validity of fuzzy rules is obtained by analyzing the effect of orthogonal vectors in the fuzzy model, and subsequently removing less important ones. The structure identification and parameter identification of the fuzzy model are simultaneously confirmed in the proposed algorithm. Using simulation results as an example, the fuzzy model of non‐linear systems can be built by using the proposed algorithm. Copyright © 2011 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society     

Development of fault diagnosis strategies based on qualitative predictions of symptom evolution behaviors

A SDG-based reasoning procedure is presented in this paper to qualitatively predict all possible symptom patterns and also their progression sequences caused by fault propagation in any given process system. These intrinsic features of the symptom evolution behaviors are then captured with IF–THEN rules in a two-layer fuzzy inference system. The proposed diagnostic system can be used to identify not only the locations of fault origins but also their magnitude levels with relatively high resolution. Numerical simulation studies have been carried out to verify the feasibility and effectiveness of the proposed approach.     

Fuzzy inference to risk assessment on nuclear engineering systems

This paper presents a nuclear case study, in which a fuzzy inference system (FIS) is used as alternative approach in risk analysis. The main objective of this study is to obtain an understanding of the aging process of an important nuclear power system and how it affects the overall plant safety. This approach uses the concept of a pure fuzzy logic system where the fuzzy rule base consists of a collection of fuzzy IF–THEN rules. The fuzzy inference engine uses these fuzzy IF–THEN rules to determine a mapping from fuzzy sets in the input universe of discourse to fuzzy sets in the output universe of discourse based on fuzzy logic principles. The risk priority number (RPN), a traditional analysis parameter, was calculated and compared to fuzzy risk priority number (FRPN) using scores from expert opinion to probabilities of occurrence, severity and not detection. A standard four-loop pressurized water reactor (PWR) containment cooling system (CCS) was used as example case. The results demonstrated the potential of the inference system for subsiding the failure modes and effects analysis (FMEA) in aging studies.     

A knowledge-based system for the prediction of surface roughness in turning process

In the present work, a knowledge-based system is developed for the prediction of surface roughness in turning process. Neural networks and fuzzy set theory are used for this purpose. Knowledge acquired from the shop floor is used to train the neural network. The trained network provides a number of data sets, which are fed to a fuzzy-set-based rule generation module. A large number of IF–THEN rules are generated, which can be reduced to a smaller set of rules by using Boolean operations. The developed rule base may be used for predicting surface roughness for given process variables as well as for the prediction of process variables for a given surface roughness. The concise set of rules helps the user in understanding the behavior of the cutting process and to assess the effectiveness of the model. The performance of the developed knowledge-based system is studied with the experimental data of dry and wet turning of mild steel with HSS and carbide tools.