神经网络自动生成模糊系统

在模糊系统的生成过程中，最主要的任务是隶属函数和模糊规则的提取和调整，但用传统方法，其工作量往往随变量数的增长而爆炸性地增加。为了解决这一问题，本文提出了一种新颖的，利用神经网络来自动地撮模糊系统 隶属函数和规则。     

CDMA蜂窝系统中基于模糊神经网络的反向链路功率控制

CDMA是一个干扰受限系统,反向链路功率控制对于克服“远近效应”和增加系统容量是非常重要的.本文提出了一种基于模糊神经网络(FNN)的自适应闭环功率控制算法,该算法动态地调整功率控制增量,使基站接收到的每个用户的发射功率相等.仿真结果表明,由于模糊神经网络能够较好地识别反向链路的时变特性,FNN功率控制算法比传统的固定步长功率控制方法取得了更好的控制性能和更大的系统容量.而且,FNN能够通过神经网络训练自动地调整隶属度函数和模糊规则,从而适合于实现在线系统识别和自适应控制.     

一种新型模糊神经网络结构确定的研究

本文给出了生种改进的模糊神经网络模型。这种模糊神经网络结构的建立只需要根据系统的输入输出数据来得到，且含义非常明晰。通常网络神经元的个数据根据经验来推，而我们通过糊数据曲线来确定网络模型，具有更高的准确度和优化性。     

Evolutionary algorithms for fuzzy control system design

This paper provides an overview on evolutionary learning methods for the automated design and optimization of fuzzy logic controllers. In a genetic tuning process, an evolutionary algorithm adjusts the membership functions or scaling factors of a predefined fuzzy controller based on a performance index that specifies the desired control behavior. Genetic learning processes deal with the automated design of the fuzzy rule base. Their objective is to generate a set of fuzzy if-then rules that establishes the appropriate mapping from input states to control actions. We describe two applications of genetic-fuzzy systems in detail: an evolution strategy that tunes the scaling and membership functions of a fuzzy cart-pole balancing controller and a genetic algorithm that learns the fuzzy control rules for an obstacle-avoidance behavior of a mobile robot     

Adaptive Sliding Mode Control of Robot Manipulator Based on Second Order Approximation Accuracy and Decomposed Fuzzy Compensator

A kind of adaptive sliding mode control scheme for tracking control of robot manipulator which has structured uncertainties and unstructured uncertainties is proposed in this paper. Multi-input Multi-output fuzzy logical system (FLS) is used as a compensator in the control law to compensate all the uncertainties. To reduce the number of the fuzzy rules and the burden of the computation, we design FLS based on second order approximation theorem which can approximate the uncertain function with less fuzzy rules at arbitrary precision than traditional FLS. Besides, to further reduce the number of the fuzzy rules and the amount of calculation, a new decomposed fuzzy logical system based on the decomposition of membership function is proposed. From the simulation results we can see that the control scheme and the fuzzy compensator proposed in this paper can perform fairly.     

Direct Yaw-Moment Control of an In-Wheel-Motored Electric Vehicle Based on Body Slip Angle Fuzzy Observer

A stabilizing observer-based control algorithm for an in-wheel-motored vehicle is proposed, which generates direct yaw moment to compensate for the state deviations. The control scheme is based on a fuzzy rule-based body slip angle (beta) observer. In the design strategy of the fuzzy observer, the vehicle dynamics is represented by Takagi-Sugeno-like fuzzy models. Initially, local equivalent vehicle models are built using the linear approximations of vehicle dynamics for low and high lateral acceleration operating regimes, respectively. The optimal beta observer is then designed for each local model using Kalman filter theory. Finally, local observers are combined to form the overall control system by using fuzzy rules. These fuzzy rules represent the qualitative relationships among the variables associated with the nonlinear and uncertain nature of vehicle dynamics, such as tire force saturation and the influence of road adherence. An adaptation mechanism for the fuzzy membership functions has been incorporated to improve the accuracy and performance of the system. The effectiveness of this design approach has been demonstrated in simulations and in a real-time experimental setting.     

A new fuzzy logic-based controller design method for DC and ACimpressed-voltage drives

This paper deals with the design of fuzzy logic-based controllers (FLBC) for DC and AC electric drives. Industrial drives employ the cascaded PI control with a subordinated current control loop to make sure that the current does not exceed the admissible value and improve dynamic performances. The nonlinear FLBC characteristics permit one to achieve the performances of the cascaded control using only one control loop. This is feasible by a suitable choice of the scaling factors together with the rules of the fuzzy controller. The authors propose a minimum number of rules and the criteria, based on physical considerations, to determine the input and output gains instead of using the trial and error procedure. The designed FLBC is able to control the speed of a DC drive as well as the rotor speed and flux of a vector-controlled induction motor drive. Computer simulations show the effectiveness of the new fuzzy-controller design method. The reduced number of rules and membership functions and the application flexibility together with the possible implementation on low cost μPs lead the authors to think that the proposed tuning criteria will be widely adopted     

自寻优模糊集的自调整模糊控制器及通用模糊控制器的硬件实现

在文[1][2]的基础上,本文给出了自寻优模糊集自调整模糊控制器算法.并在这种软件设计的基础上,与文[4][5]提出的硬件构想相结合,实现了通用模糊控制器的硬件实现,其原理是用软件确定最佳的模糊集和控制规则,而用CLC连续逻辑电路实现控制作用.将这种控制器与最佳整定的DDC控制效果进行了比较,表明本文所提出的模糊控制器是简单可行的,模糊控制器对时变系统、非线性系统等有着广阔的应用前景.这种硬件具有很强的通用性,为模糊控制的进一步推广创造了条件.     

一种启发式变阶直觉模糊时间序列预测模型

论文针对已有高阶模糊时间序列模型在预测精度和预测范围上的限制,结合直觉模糊集理论,提出一种启发式变阶直觉模糊时间序列预测模型。模型首先应用直接模糊聚类算法对论域进行非等分划分;然后,针对直觉模糊时间序列的数据特性,改进现有直觉模糊集隶属度和非隶属度函数的建立方法;最后,采用阶数随序列实时变化的高阶预测规则进行预测,并将历史数据发展趋势的启发知识引入解模糊过程,使模型的预测范围得到扩展。在Alabama大学入学人数和北京市日均气温两组数据集上分别与典型方法进行对比实验,结果表明该模型有效克服了传统模型的缺点,拥有较高的预测精度,证明了模型的有效性和优越性。     

Fuzzy logic approach to VLSI placement

A contemporary definition of VLSI placement problem is characterized by multiple objectives. These objectives are: timing, chip area, interconnection length and possibly others. In this paper, fuzzy logic has been used to facilitate multiobjective decision-making in placement for standard cell design style. A placement process has been defined in terms of linguistic variables, linguistic values and membership functions. Various objectives have been related by hierarchical fuzzy logic rules implemented as object-oriented programming objects. It is demonstrated that a designed fuzzy logic system is flexible in selecting goals and considering tradeoffs. Details of implementation, experimental results and comparisons with other systems are provided     

改进的模糊神经网络及其在人脸识别中的应用

对基于模糊神经网络的人脸图像分类器进行研究。将多输入单输出模糊推理系统改造成多输入多输出的模糊神经分类器,并提出了一种改进的模糊神经分类器,改进模型的计算量明显减少。在将模糊规则库与训练样本集对应的基础上提出了一种模糊隶属函数参数的初始化方法。该初始化方法的优点在于它充分利用了训练样本所包含的鉴别信息。在ORL人脸的原始图像空间中用上述方法设计分类器,获得了较好的实验结果。     

Experimental and statistical study in adhesion features of bonded interfaces of IC packages

The existence of defects in the corresponding interfaces can gradually degrade the interfacial adhesion when the flip chip package is exposed to the high temperature and humidity. This study used the fuzzy controller to stabilize both the moisture weight gain and adhesion strength together for the button shear test specimen. Parameters design, although based on the Taguchi method, can optimize the performance characteristic through the setting of process parameters and can reduce the sensitivity of the system performance to sources of variation. However, most published Taguchi applications to date have been concerned with the optimization of a single performance characteristic only. For example, it is only concerned a single performance characteristic with the optimization of the adhesion strength or the moisture gain weight individually. In this study, the control rules of the fuzzy controller were formed using the authors’ experience and knowledge of the IC packaging process. The control parameters to be tuned were the membership functions. Therefore, the controller’s performance depended on the membership functions. The fuzzy controller combined Taguchi parameter design, which makes the control performance insensitive to the operating condition changes and noise was used to determine the membership functions.     

Design of a fuzzy traffic controller for ATM networks

This paper presents the design of a fuzzy traffic controller that simultaneously manages congestion control and call admission control for asynchronous transfer mode (ATM) networks. The fuzzy traffic controller is a fuzzy implementation of the two-threshold congestion control method and the equivalent capacity admission control method extensively studied in the literature. It is an improved, intelligent implementation that not only utilizes the mathematical formulation of classical control but also mimics the expert knowledge of traffic control. We appropriately choose input linguistic variables of the fuzzy traffic controller so that the controller is a closed-loop system with stable and robust operation. We extract knowledge of conventional control methods from numerous analytical data using a clustering technique and then use this knowledge to set parameters of the membership functions and fuzzy control rules via fuzzy set manipulation (linguistically stated but mathematically treated) with the aid of an optimization technique named genetic algorithm (GA). Simulation results show that the proposed fuzzy admission control improves system utilization by a significant 11%, while maintaining the quality of service (QoS) contract comparable with that of the conventional equivalent capacity method. The performance of the proposed fuzzy congestion control method is also 4% better than that of the conventional two-threshold congestion control method     

Circuit implementation of a fully programmable and continuously slope tunable triangular/trapezoidal membership function generator

In this article, a fully programmable membership function generator (MFG) is proposed. This MFG is capable of generating triangular, trapezoidal as well as both S-shaped and Z-shaped membership functions simultaneously. Utilizing a differential pair as an analog switch leads to relax the design of fuzzy systems control part. This MFG has the ability of adapting itself with various fuzzy controllers which produce different control voltage ranges. Unlike the available reported literatures, this MFG uses a new analog programmable current mirror (APCM) instead of digitally programmable current mirrors to adjust the slopes of membership functions. Extensive time domain simulations have been carried out using Hspice by level 49 parameters (BSIM3v3) in standard CMOS technology to validate the effective performance of the proposed MFG.     

An Optimum Vertical Handoff Decision Algorithm Based on Adaptive Fuzzy Logic and Genetic Algorithm

Handoff decision making is one of the most important topics in wireless heterogeneous networks architecture as there are many parameters which have to be considered when triggering handoff and selecting suitable access point. More intelligent approaches which reckon user profiles, application requirements, and network conditions must be improved so that desired performance results for both user and network could be provided. In this paper we introduce a new adaptive vertical handoff decision making algorithm in which fuzzy membership functions are optimized by means of genetic algorithm. Genetic algorithm is an adaptive search technique based on natural selection and genetic rules. In addition to that, it takes places in various scientific applications and can be used to adjust the membership functions in fuzzy systems. The purpose of the study is to adjust the shape of fuzzy membership functions, properly, using genetic algorithm in order to achieve optimum handoff performance. The results show that, compared to the several different algorithms performance of the proposed approach with genetic algorithm is significantly improved for both user and network in terms of number of handoff while the other requirements are still satisfied.     

Minimal fuzzy memberships and rules using hierarchical geneticalgorithms

A new scheme to obtain optimal fuzzy subsets and rules is proposed. The method is derived from the use of genetic algorithms, where the genes of the chromosome are classified into two different types. These genes can be arranged in a hierarchical form, where one type of gene controls the other. The effectiveness of this genetic formulation enables the fuzzy subsets and rules to be optimally reduced and, yet, the system performance is well maintained. In this paper, the details of formulation of the genetic structure are given. The required procedures for coding the fuzzy membership function and rules into the chromosome are also described. To justify this approach to fuzzy logic design, the proposed scheme is applied to control a constant water pressure pumping system. The obtained results, as well as the associated final fuzzy subsets, are included in this paper. Because of its simplicity, the method could lead to a potentially low-cost fuzzy logic implementation     

Membership function learning in fuzzy classification

A novel three-layer fuzzy neural network (FNN) is proposed which possesses ihe structure and learning ability of artificial neural networks, and the classification ability of fuzzy algorithms for pattern classification problems. During learning, the proposed FNN learns the membership function of each fuzzy class from training samples and adaptively organizes its hidden layer. The learning and recall times of the FNN are fast. Simulation results are also presented.     

On interpretation of graffiti digits and characters for eBooks: neural-fuzzy network and genetic algorithm approach

This paper presents the rule optimization, tuning of the membership functions, and optimization of the number of fuzzy rules, of a neural-fuzzy network (NFN) using a genetic algorithm (GA). The objectives are achieved by training a proposed NFN with rule switches. The proposed NFN and GA are employed to interpret graffiti number inputs and commands for electronic books (eBooks).     

A recurrent fuzzy cellular neural network system with automatic structure and template learning

It is widely accepted that using a set of cellular neural networks (CNNs) in parallel can achieve higher level information processing and reasoning functions either from application or biologics points of views. Such an integrated CNN system can solve more complex intelligent problems. In this paper, we propose a novel framework for automatically constructing a multiple-CNN integrated neural system in the form of a recurrent fuzzy neural network. This system, called recurrent fuzzy CNN (RFCNN), can automatically learn its proper network structure and parameters simultaneously. The structure learning includes the fuzzy division of the problem domain and the creation of fuzzy rules and CNNs. The parameter learning includes the tuning of fuzzy membership functions and CNN templates. In the RFCNN, each learned fuzzy rule corresponds to a CNN. Hence, each CNN takes care of a fuzzily separated problem region, and the functions of all CNNs are integrated through the fuzzy inference mechanism. A new online adaptive independent component analysis mixture-model technique is proposed for the structure learning of RFCNN, and the ordered-derivative calculus is applied to derive the recurrent learning rules of CNN templates in the parameter-learning phase. The proposed RFCNN provides a solution to the current dilemma on the decision of templates and/or fuzzy rules in the existing integrated (fuzzy) CNN systems. The capability of the proposed RFCNN is demonstrated on the real-world defect inspection problems. Experimental results show that the proposed scheme is effective and promising.     

通用硬件模糊控制器研究

提出了一种基于多值逻辑电路的模糊控制器硬件实现方案,采用规则分时进行硬件模糊推理,不同规则的推理结果合并后形成模糊输出,经模糊判决后形成精确量输出。该方案的复杂性不受规则数量的影响,执行速度不受语言变量维数的影响,该方案通过改变存储器数据可以方便地调整隶属度函数和模糊控制规则,克服了硬件模糊控制器灵活性差这一重大缺陷,该方案便于以VLSI实现。