基于奇异值分解的非均匀采样非线性系统的模糊模型辨识

针对复杂、不确定、非均匀采样数据的非线性系统,提出一种基于矩阵奇异值分解(SVD)的模型结构辨识和参数估计的建模方法.首先,利用矩阵奇异值(SVD)分解算法分析各局部模型与奇异值、积累贡献率的关系,确定模糊模型的规则数,从而实现模型的结构优化;然后,为了克服递推最小二乘出现的误差积累、传递现象,采用奇异值分解的递推最小二乘估计模型的结论参数;最后,通过仿真实例验证所提出算法的有效性.     

灰色复数矩阵SVD的无参考模糊图像质量评价

为了进一步突出重要的图像结构特征,采用复数矩阵表示图像,提出了基于灰色复数奇异值分解的无参考模糊图像质量评价方法。该方法首先将原始模糊图像经点扩散函数生成二次模糊图像,再采用复数矩阵的形式表示原始图像和二次模糊图像的结构特征,在此基础上,对原始模糊图像和二次模糊图像进行分块复数矩阵奇异值分解,获得区域相关度,采用灰色关联度评价模糊图像质量。在3个数据库上的实验结果表明,该方法评价结果合理,与主观评价具有较好的一致性。     

区间二型模糊熵及其在图像分割中的应用

针对计算区间二型模糊熵存在的不足,构造一种新的区间二型模糊熵。新的区间二型模糊熵同时考虑了隶属度的均值和本身的模糊性对熵值的影响,而且满足四条公理,较为全面客观地测度了区间二型模糊集的不确性,并被应用于图像分割。数值实例和仿真实验表明了所提出区间二型模糊熵的合理性和实用性。     

二型模糊集的模糊熵研究

研究基于质心的二型模糊集的模糊熵和加权模糊熵,构造了两个二型模糊集的模糊熵度量.针对二型模糊集的特殊情形,提出一种新的区间值模糊集的模糊熵度量,既弥补了现有区间值模糊集退化为普通模糊集时熵为零的不足,又克服了两个明显不同的区间值模糊集熵相等的缺点.数值实例和仿真实验表明了所提出模糊熵的合理性和实用性.     

基于聚类和SVD算法的模糊逻辑系统结构辨识

为了研究模糊逻辑系统新的结构辨识方法, 提出采用基于山峰函数的减法聚类算法构造模糊逻辑系统的初始结构, 并利用奇异值分解(SVD)算法分析了模糊规则与奇异值、累积贡献率以及索引向量的关系, 从而实现了模糊逻辑结构的优化. 最后, 对该算法的可行性和有效性进行了仿真验证和性能比较, 取得了较好的效果.     

基于自组织模糊神经网络的出水总磷预测

针对污水处理过程出水总磷预测问题,本文提出一种基于改进Levenberg--Marquardt(improved Levenberg--Marquardt,ILM)学习算法和奇异值分解(singular value decomposition,SVD)的适于在线建模的自组织模糊神经网络(fuzzy neural network,FNN)预测方法.ILM-SVDFNN采用改进LM学习算法对隶属函数中心、宽度和输出权值进行训练.在参数自适应学习的同时,采用单边Jacobi变换实现规则层输出阵的奇异值分解,根据奇异值定义增长和修剪指标实现规则层神经元在线动态调整.此外,证明了所提方法在网络结构固定和调整阶段的收敛性.最后,利用典型非线性系统辨识、Mackey-Glass时间序列预测和实际污水处理过程出水总磷预测实验进行验证.仿真结果显示所设计的自组织模糊神经网络结构紧凑且预测精度较高,较好地满足了污水处理厂对出水总磷检测精度和实时性的要求.     

基于OWA算子的区间值加权模糊推理

针对如何对区间值模糊产生式规则赋予合理权值的问题,将OWA算子引入到区间值模糊推理中。介绍一种基于OWA算子的区间值赋权方法,根据此方法给出区间值模糊集上的加权模糊产生式规则的推理算法。在采用该算法的过程中,为合理地计算输入事实与规则前件的匹配程度,引入基于OWA算子的区间值模糊匹配函数值和总体贴近度的计算方法。实例分析表明了所给出的区间值模糊推理算法的有效性和可行性。     

区间值关系数据库上模糊关联规则的预测方法

讨论了区间值关系数据库上模糊关联规则的挖掘算法与预测方法。采用一种比RFCM算法省时的FCMdd算法将记录在属性的取值划分成若干个模糊集,并提出区间值关系数据库上模糊关联规则的挖掘算法。仿真实例说明挖掘算法能够通过挖掘有意义的模糊关联规则来发现区间值关系数据库中蕴涵的关联性。区间值关系数据库上模糊关联规则的预测方法改进了标准可加性模型,并通过遗传算法调整模糊关联规则中三角模糊数的参数来提高预测的精度。     

区间二型模糊子空间0阶TSK系统

人们倾向于使用少量的有代表性的特征来描述一条规则,而忽略极为次要的冗余的信息。经典的区间二型TSK(Takagi-Sugeno-Kang)模糊系统,在规则前件和后件部分会使用完整的数据特征空间,对于高维数据而言,易导致系统的复杂度增加和可解释性的损失。针对于此,提出了区间二型模糊子空间0阶TSK系统。在规则前件部分,使用模糊子空间聚类和网格划分相结合的方法生成稀疏的规整的规则中心,在规则后件部分,使用简化的0阶形式,从而得到规则语义更为简洁的区间二型模糊系统。在模拟和真实数据上的实验结果表明该方法分类效果良好,可解释性更好。     

基于奇异值分解的固定区间平滑新方法*

本文提出一种基于奇异值分解（ＳＶＤ）的固定区间平滑新方法，该方法基于Ｒａｕｃｈ－Ｔｕｎｇ－Ｓｔｒｉｅｂｅｌ固定区间平滑方法，利用奇异值分解作为计算工具，将原算法中协方差阵进行奇异值分解，不仅具有很好的数值稳定性和鲁棒性，而且避免了矩阵的求逆，此外，采用ＳＶＤ分解，具有明显的物理意义。仿真计算结果证明了本文方法的有效性和优越性。     

Interval type-2 fuzzy expert system for prediction of carbon monoxide concentration in mega-cities

This paper presents an indirect approach to interval type-2 fuzzy logic system modeling to forecaste the level of air pollutants. The type-2 fuzzy logic system permits us to model the uncertainties among rules and the parameters related to data analysis. In this paper, we propose an indirect method to create an interval type-2 fuzzy logic system from a historical data, where Footprint of Uncertainties of fuzzy sets are extracted by implementation of an interval type-2 FCM algorithm and based on an upper and lower value for the level of fuzziness m in FCM. Finally, the proposed model is applied for prediction of carbon monoxide concentration in Tehran air pollution. It is shown that the proposed type-2 fuzzy logic system is superior in comparison to type-1 fuzzy logic systems in terms of two performance indices.     

对区间二型模糊集的EKM降型法的改进

二型模糊集的质心计算称为降型,目前的降型方法大多计算成本较高,其中EKM (Enhanced Karnik-Mendel)法可计算区间二型模糊集的质心。然而,由于EKM算法中求取切换点的初始化方法还不完善,计算时间较长,使其在实际应用中受到一定限制。对此,提出一种新的改进EKM法,对原有方法进行了两处改进：更改切换点的初始化条件和改进查找切换点的方法。所提出的方法可实现向上和向下搜索,计算量大大减小,降型更有效。仿真结果验证了新的改进EKM法的有效性。     

基于区间直觉模糊的动态多属性灰色关联决策方法

针对属性值为区间直觉模糊数且属性权重未知的一类决策问题,利用灰色关联分析方法的思想,构建了一种动态区间直觉模糊数多属性决策方法。首先利用区间直觉模糊数的运算法则和性质设计各时间段的正负理想方案,并以与正理想方案灰色关联度偏差最小化为目标构建了多目标规划模型,确定属性权重;然后通过计算各时间段各方案对正、负理想方案的区间直觉模糊数的灰色关联度,构建方案优属度模型,并求解方案优属度的表达式,确定方案的优势度;最后通过一个案例验证了所提出的构建方法的有效性和可行性。     

一种基于量子遗传算法的扩展T-S 模型辨识

在传统T-S模型的基础上,提出一种扩展T-S模型。该模型由一组模糊规则组成,由规则前件实现输入空间的划分,将成员函数及其函数变换引入规则后件以实现对输入子空间的非线性映射。对于该模型的建立,使用改进量子遗传算法优化规则前件,递推最小二乘法确定规则后件参数。通过对两个典型非线性系统辨识,仿真结果表明了该模型可以显著提高辨识精度,且具有很好的泛化性能。     

Development of an interval type-2 fuzzy sets based hierarchical MADM model by combining DEMATEL and TOPSIS

In this study, a new interval type-2 fuzzy multiple-attribute decision making model is developed by integrating Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) and Decision Making Trial and Evaluation Laboratory (DEMATEL). The proposed model utilizes hierarchical decomposition approach for reducing inherent complexity of the decision making problems. Additionally, interdependencies among problem attributes are taken into consideration by using interval type-2 fuzzy DEMATEL method. Finally, ranking orders of the alternatives are obtained by hierarchical interval type-2 fuzzy TOPSIS method. As there are several forms of interactions among criteria in real life settings, decision makers should be provided with the expert and intelligent systems which can overcome the preferential independence assumption. The proposed method is able to model causal dependencies by using interval type-2 trapezoidal fuzzy sets. The proposed method is implemented in a Strengths, Weaknesses, Opportunities, and Threats (SWOT)-based strategy selection problem.     

Local consistency adjustment strategy and DEA – driven interval type-2 trapezoidal fuzzy decision-making model and its application for fog-haze factor assessment problem

This paper aims to develop a novel decision-making method with interval type-2 trapezoidal fuzzy preference (IT2TrFPR), which can deal with the complex decision information presented in the form of interval type-2 trapezoidal fuzzy numbers. In this paper, we mainly propose a novel interval type-2 trapezoidal fuzzy decision-making method with local consistency adjustment strategy and data envelopment analysis (DEA). Considering the harm of fog-haze pollution to the environment and human life, we apply the decision-making method to the problem about influence factors of for-haze weather. Firstly, we introduce the definition of IT2TrFPR that sufficiently expresses the uncertainty of original decision-making information. After that, we show the definition of the order consistency and additive consistency for IT2TrFPR. Considering that the original IT2TrFPR given by decision-makers usually does not satisfy the characteristic of consistency, to transform the unacceptable additive consistent IT2TrFPRs into acceptable ones, we design a consistency-improving algorithm that uses the local adjustment approach to preserve the original decision-making information as much as possible and avoids the original information loss. Then, an output-oriented interval type-2 trapezoidal fuzzy DEA model and the concept for quasi interval type-2 trapezoidal fuzzy priority weight are developed to derive the interval type-2 trapezoidal fuzzy priority weight vector (IT2TrFPW) and obtain the final ranking result of alternatives. Finally, the effectiveness of the proposed decision-making method is demonstrated by a numerical example of selecting the most crucial fog-haze influence factor. Meanwhile, we also conduct a comparative analysis by comparing our method with the existing methods to show some merits of the proposed method.

     

Position tracking of a 3-PSP parallel robot using dynamic growing interval type-2 fuzzy neural control

Parallel robots have complicated structures as well as complex dynamic and kinematic equations, rendering model-based control approaches as ineffective due to their high computational cost and low accuracy. Here, we propose a model-free dynamic-growing control architecture for parallel robots that combines the merits of self-organizing systems with those of interval type-2 fuzzy neural systems. The proposed approach is then applied experimentally to position control of a 3-PSP (Prismatic–Spherical–Prismatic) parallel robot. The proposed rule-base construction is different from most conventional self-organizing approaches by omitting the node pruning process while adding nodes more conservatively. This helps preserve valuable historical rules for when they are needed. The use of interval type-2 fuzzy logic structure also better enables coping with uncertainties in parameters, dynamics of the robot model and uncertainties in rule space. Finally, the adaptation structure allows learning and further adapts the rule base to changing environment. Multiple simulation and experimental studies confirm that the proposed approach leads to fewer rules, lower computational cost and higher accuracy when compared with two competing type-1 and type-2 fuzzy neural controllers.     

An ELECTRE I-based multi-criteria group decision making method with interval type-2 fuzzy numbers and its application to supplier selection

As an extension of type-2 fuzzy numbers (T2 FNs), interval type-2 fuzzy numbers (IT2 FNs) are able to deal more effectively with uncertainty, have better processing abilities, and simpler computations. Because of these abilities, IT2 FNs have been widely applied indecision support systems (DSS). In this paper, to ensure more effective multi-criteria group decision-making in uncertain environments, the elimination and choice translating reality (ELECTRE) method is extended using interval type-2 fuzzy numbers. An α-based distance method is first proposed to measure the proximity between the interval type-2 fuzzy numbers. Then, an entropy measure for the IT2 FNs and an entropy weight model are developed to objectively determine the criteria weights without any weight information. By applying an α-based distance method, the concordance and discordance for each alternative are measured to determine the partial-preference outranking order. A complementary analysis is then conducted to obtain the full rank order of all alternatives. Finally, the feasibility and applicability of the proposed method are detailed using two different practical examples. A sensitivity and comparative analysis are also conducted to demonstrate the strength and practicality of the proposed method.     

Interval type-2 fuzzy inverse controller design in nonlinear IMC structure

In the recent years it has been demonstrated that type-2 fuzzy logic systems are more effective in modeling and control of complex nonlinear systems compared to type-1 fuzzy logic systems. An inverse controller based on type-2 fuzzy model can be proposed since inverse model controllers provide an efficient way to control nonlinear processes. Even though various fuzzy inversion methods have been devised for type-1 fuzzy logic systems up to now, there does not exist any method for type-2 fuzzy logic systems. In this study, a systematic method has been proposed to form the inverse of the interval type-2 Takagi-Sugeno fuzzy model based on a pure analytical method. The calculation of inverse model is done based on simple manipulations of the antecedent and consequence parts of the fuzzy model. Moreover, the type-2 fuzzy model and its inverse as the primary controller are embedded into a nonlinear internal model control structure to provide an effective and robust control performance. Finally, the proposed control scheme has been implemented on an experimental pH neutralization process where the beneficial sides are shown clearly.     

A Self-Evolving Interval Type-2 Fuzzy Neural Network With Online Structure and Parameter Learning

This paper proposes a self-evolving interval type-2 fuzzy neural network (SEIT2FNN) with online structure and parameter learning. The antecedent parts in each fuzzy rule of the SEIT2FNN are interval type-2 fuzzy sets and the fuzzy rules are of the Takagi–Sugeno–Kang (TSK) type. The initial rule base in the SEIT2FNN is empty, and the online clustering method is proposed to generate fuzzy rules that flexibly partition the input space. To avoid generating highly overlapping fuzzy sets in each input variable, an efficient fuzzy set reduction method is also proposed. This method independently determines whether a corresponding fuzzy set should be generated in each input variable when a new fuzzy rule is generated. For parameter learning, the consequent part parameters are tuned by the rule-ordered Kalman filter algorithm for high-accuracy learning performance. Detailed learning equations on applying the rule-ordered Kalman filter algorithm to the SEIT2FNN consequent part learning, with rules being generated online, are derived. The antecedent part parameters are learned by gradient descent algorithms. The SEIT2FNN is applied to simulations on nonlinear plant modeling, adaptive noise cancellation, and chaotic signal prediction. Comparisons with other type-1 and type-2 fuzzy systems in these examples verify the performance of the SEIT2FNN.