基于模糊神经网络结构的多值联想记忆

文章基于模糊神经网络结构,即通过模糊化,推理,去模糊三个过程,把Kosko提出的模糊联想记忆(FAM)网络模型应用到容错性需要较强的多值联想记忆中,解决了这种网络模型不能对随机噪声模式正确联想的问题,新的网络模型设计简单,大量实验表明文中的联想记忆网络大大提高了FAM网络的容错性能。     

带参数聚合算子的模糊联想记忆网络

基于最大运算Max和t--范数T的神经网络模型Max-T FAM是B.Kosko提出的经典模糊联想记忆(FAM)网络的一种重要的广义形式,其性能有多处不足.本文利用一种参数化聚合算子∨λ,提出了一种计算简单、易于硬件实现的广义模糊联想记忆(GFAM)网络,其连接算子从{∨λ|λ∈[0,1]}中选取;从理论上严格证明了GFAM具有一致连续性,比所有Max-T FAM的映射能力和存储能力强很多;接着运用模糊关系方程理论提出和分析了GFAM的一种所谓的Max-Min-λ学习算法;最后用实验对GFAM和Max-T FAM的完整可靠存储能力进行了比较,并示例了GFAM在图像联想方面的应用.     

武器系统故障诊断中的推理技术研究

推理技术是故障诊断系统的核心,目前针对复杂系统的故障推理仍然是个难题,提出了将模糊神经网络与事例推理相融合的集成诊断模型,通过规则推理、事例推理和神经网络推理三个子系统的互相验证,既解决了事例推理不能得到最优解的问题,又解决了模糊神经网络训练样本难以选取的问题,将该推理方法用于某武器系统故障诊断,取得了较好地诊断效果.     

模糊联想记忆对训练模式对摄动的鲁棒性研究

基于模糊取大算子（V）和T-模的模糊合成,构建了一类模糊联想记忆网络（V-T FAM）。利用T-模的模糊蕴涵算子,给出了这类V-T FAM的学习算法。针对训练模式对小幅摄动可能对模糊神经网络的性能产生副作用,提出V-T FAM对训练模式对摄动的鲁棒性概念。理论研究表明,当T-模满足Lipschitz条件时,采用上述学习算法的V-T FAM对训练模式对摄动幅度,在系数为β的条件下全局拥有好的鲁棒性。最后用V-T FAM在图像联想方面的实验验证了理论结果。     

模糊概率SDG模型及故障推理方法

基于符号有向图(SDG)的故障诊断方法具有良好的完备性和易于解释性,但其存在分辨率差的缺陷,为此提出基于模糊概率SDG模型和贝叶斯推理的半定量故障诊断方法.用模糊变量表示节点变量,用条件概率表(CPT)表达节点间的定性因果关系,利用贝叶斯推理和回溯搜索找出故障源候选解的集合,并对候选解进行排序.最后建立了某卫星一次电源系统的诊断模型.仿真结果表明,该方法有效地提高了诊断的分辨率,适用于航天器在轨故障诊断.     

基于模糊描述逻辑的模糊XML模型的表示与推理

通过分析描述逻辑与XML之间的关系,该文进一步研究了基于模糊描述逻辑的模糊XML模型的表示与推理.首先提出了模糊XML模型(包括模糊XML文档和模糊DTD)的形式化定义.然后针对模糊XML模型的特点和推理需求,提出了一种新的模糊描述逻辑f一ALCQwf-reg,给出了它的语法、语义、知识库以及推理算法.在此基础上,研究...     

支持模糊数据类型表示的模糊描述逻辑F-SHOIQ(G)

分析了现有描述逻辑在模糊知识和数据类型表示方面存在的问题,提出了一种新的模糊描述逻辑F-SHOIQ(G).F-SHOIQ(G)不仅能够表示模糊知识,而且能够表示含有自定义模糊数据类型及自定义模糊数据类型谓词的模糊数据信息.首先,给出了模糊数据类型域的概念和模糊数据类型表示的一般形式,在此基础上,定义了F-SHOIQ(G)的语法、语义及相应的知识库,进而给出了基于模糊Tableaux的F-SHOIQ(G)概念的可满足性推理算法.其次,将经典描述逻辑中的推理结构(该结构将Tableaux扩展规则推理和数据类型推理相分离)用于F-SHOIQ(G)的推理问题,设计了相应的模糊数据类型推理机.最后,详细证明了F-SHOIQ(G)概念的可满足性推理问题是可判定的.在数据类型表示方面,F-SHOIQ(G)具备比FSHOIQ更强的表达能力和推理能力,为语义Web表示和推理模糊数据信息提供了理论基础.     

模糊联想记忆网络对模式摄动的鲁棒性分析

在实际问题中,所获取的模糊神经网络的训练模式对总与客观真实的模式对存在一定的小幅误差(摄动),从而可能导致对某些输入网络的实际输出与期望输出有很大的误差。为此,提出了训练模式集摄动对模糊联想记忆网络(FAM)的鲁棒性概念,并具体讨论了采用一种新的权值学习算法时FAM的这种鲁棒性及其控制方法。最后通过实验证明了采用这种新的权值学习算法时,FAM对模式摄动不会拥有好的鲁棒性。     

面向语义Web语义表示的模糊描述逻辑

分析了语义Web语义表示理论的研究现状及存在的问题,提出了一种新的面向语义Web语义表示的模糊描述逻辑FSHOIQ(fuzzy SHOIQ).给出了FSHOIQ的语法和语义,提出了FSHOIQ的模糊Tableaux的概念,给出了一种基于模糊Tableaux的FSHOIQ的ABox约束下的可满足性推理算法,证明了可满足性推理算法的正确性.提出了FSHOIQ的TBox扩展和去除方法,并证明了FSHOIQ的TBox约束下的包含推理问题可以转化为ABox约束下的可满足性推理问题.FSHOIQ为语义Web表示和推理模糊知识提供了理论基础.     

基于DFL的Agent推理模型

推理是Agent研究的核心问题之一.根据Agent的推理特性,引用DFL理论提出基于动态模糊逻辑(DFL)的Agent推理模型,实现了范例验证系统,即基于动态模糊逻辑(DFL)的Agent推理模型的叮当漫画书城系统.     

Application of Fuzzy Logic for the Solution of Inverse Kinematics and Hierarchical Controls of Robotic Manipulators

In this paper, hierarchical control techniques is used for controlling a robotic manipulator. The proposed method is based on the establishment of a non-linear mapping between Cartesian and joint coordinates using fuzzy logic in order to direct each individual joint. The hierarchical control will be implemented with fuzzy logic to improve the robustness and reduce the run time computational requirements. Hierarchical control consists of solving the inverse kinematic equations using fuzzy logic to direct each individual joint. A commercial Microbot with three degrees of freedom is utilized to evaluate this methodology. A decentralized fuzzy controller is used for each joint, with a Fuzzy Associative Memories (FAM) performing the inverse kinematic mapping in a supervisory mode. The FAM determines the inverse kinematic mapping which maps the desired Cartesian coordinates to the individual joint angles. The individual fuzzy controller for each joint generates the required control signal to a DC motor to move the associated link to the new position. The proposed hierarchical fuzzy controller is compared to a conventional controller. The simulation experiments indeed demonstate the effectiveness of the proposed method.     

Dynamic positioning of drilling vessels with a fuzzy logic controller

A fuzzy logic controller for dynamic positioning of drilling vessels in deep water is presented. The core of the fuzzy controller is a set of fuzzy associative memory (FAM) rules that correlate each group of fuzzy control input sets to a fuzzy control output set. A FAM rule is a logical if-then-type statement based on one's sense of realism and experience or can be provided by an expert operator. The design of the fuzzy controller is very simple and does not require mathematical modelling of the complicated nonlinear system based on first principles. The fuzzy controller uses measured vessel heading, yaw rate, distance and velocity of the vessel relative to the desired position (location and heading) to generate the control outputs to bring the vessel to and maintain it in the desired position. The control outputs include the rudder angle, propeller thrust and lateral bow thrust. The effectiveness and robustness of the fuzzy controller are demonstrated through numerical time-domain simulations of the dynamic positioning of a drill ship of Mariner Class hull with use of nonlinear ship equations of motions.     

Adaptive fuzzy systems for backing up a truck-and-trailer

Fuzzy control systems and neural-network control systems for backing up a simulated truck, and truck-and-trailer, to a loading dock in a parking lot are presented. The supervised backpropagation learning algorithm trained the neural network systems. The robustness of the neural systems was tested by removing random subsets of training data in learning sequences. The neural systems performed well but required extensive computation for training. The fuzzy systems performed well until over 50% of their fuzzy-associative-memory (FAM) rules were removed. They also performed well when the key FAM equilibration rule was replaced with destructive, or ;sabotage', rules. Unsupervised differential competitive learning (DCL) and product-space clustering adaptively generated FAM rules from training data. The original fuzzy control systems and neural control systems generated trajectory data. The DCL system rapidly recovered the underlying FAM rules. Product-space clustering converted the neural truck systems into structured sets of FAM rules that approximated the neural system's behavior.     

在线知识获得智能模糊控制器研究

对一些复杂的系统。传统PID或模糊控制很难得到满意控制效果，本文提出采用基于RBF神经网络和遗传算法的自适应模糊控制器来进行控制。由遗传算法在线优化模糊控制器的比例因子、模糊推理规则和隶属函数。并由RBF网络辨识被控对象的动态特性，以评价模糊控制器控制性能。仿真实验表明。优化后的Fuzzy控制器具有较强的学习和自适应控制能力，控制效果优于没有寻优的Fuzzy控制。     

一种模糊神经解耦混合控制器的研究

该文将模糊神经网络与Pl控制技术相结合构成一种模糊神经解耦混合控制器。新控制器在控制过程中借助模糊神经网络的自学习算法实现控制参数的在线调整。仿真结果表明，该控制方法对非线性时变系统有较好的控制效果。     

On fuzzy associative memory with multiple-rule storage capacity

Kosko's fuzzy associative memory (FAM) is the very first neural network model for implementing fuzzy systems. Despite its success in various applications, the model suffers from very low storage capacity, i.e., one rule per FAM matrix. A lot of hardware and computations are usually required to implement the model and, hence, it is limited to applications with small fuzzy rule-base. In this paper, the inherent property for storing multiple rules in a FAM matrix is identified. A theorem for perfect recalls of all the stored rules is established and based upon which the hardware and computation requirements of the FAM model can be reduced significantly. Furthermore, we have shown that when the FAM model is generalized to the one with max-bounded-product composition, single matrix implementation is possible if the rule-base is a set of semi-overlapped fuzzy rules. Rule modification schemes are also developed and the inference performance of the established high capacity models is reported through a numerical example     

基于有序神经网络的神经模糊预测控制

本文使用有序神经网络和改进的模糊控制器构成了一种新型的神经模糊预测控制方法，有序网络学习速度快，所需神经数目少，用事先训练好的有序网络代替传统的预测模型，以期增强输出预测的准确性；同时，用一种改进的模糊控制器原有的PID控制器，增强系统的鲁棒性。仿真结果表明，所提出的神经模糊预测控制方法可以获得理想的控制效果。     

基于自适应模糊神经网络控制器的网络控制系统

网络控制系统中存在着时延、丢包、网络干扰等问题。针对网络控制系统中存在恶化系统的控制性能,甚至导致系统不稳定的因素,提出了一种基于自适应模糊神经网络控制器的网络控制系统,它能根据系统的实际输出与期望输出误差,利用自适应模糊控制和神经网络自学习的原理进行控制参数的自行调整,以符合控制系统的实际要求,同时,分析了网络延时,丢包率及网络干扰因素对系统性能的影响。利用TrueTime工具箱建立了包含自适应模糊神经网络控制器的网络控制系统的仿真模型,并将其分别与基于常规PID控制器的网络控制系统和基于模糊参数PID控制器的网络控制系统进行了比较。实验结果表明,在相同的网络环境下,基于自适应模糊神经网络控制器的网络控制系统的控制效果比基于常规的PID控制器和基于模糊参数PID控制器的要好,且具有较好的抗干扰能力和鲁棒性能。