CPI数据预测计算的ANFIS模型结构探讨

依据CPI经济序列数据确定性混沌原理,探讨自适应神经模糊推理系统模型构造,并给出此类混沌数据列预测的ANFIS系统结构形式,进行CPI经济序列数据预测.并用实例拟合、预测数据证明:ANFIS模型是一种精度较高的混沌数据序列预报系统.为CPI数据预测提供了一种计算方法.     

自适应神经模糊推理系统在交通污染物浓度预测中的应用

城市交通带来的废气排放已经成为城市大气污染的主要来源之一。交通污染问题的成因和机理较为复杂,变化规律具有较强非线性和周期性特征。将自适应神经模糊推理系统(adaptive neuro fuzzy inference system,ANFIS)应用于交通污染物浓度时序数据预测时呈现出良好的泛化能力。本文以长沙市CO小时浓度数据为研究目标,通过分析CO浓度时序数据的自相关性、偏自相关性,以及交通流对CO浓度的时滞性影响,确定ANFIS预测模型的输入变量。结果表明,相较于传统的时间序列预测模型以及机器学习模型,ANFIS模型预测结果具有更高的精度,能够对交通环境污染进行预测及预警,为防止城市灾害性大气污染事件发生奠定理论研究基础并提供有效决策支持。     

模糊推理的灵敏度

借助于模糊逻辑连接词的灵敏度，定义了模糊推理系统的灵敏度，研究了几种常见的模糊推理系统的灵敏度，进一步估算了各种模糊推理机的灵敏度，并将模糊推理系统的灵敏度与模糊连接词灵敏度的关系用等式表示出来。     

基于FI-代数的一个逻辑系统

建立一种基于FI-代数的模糊命题演算的形式演绎系统．并讨论了该系统语义的完备性。其目的在于使通常的众多模糊推理系统能够纳入该逻辑系统之中．以便在一个更加广泛的代数和逻辑框架下来研究模糊推理的逻辑基础问题。     

区间值模糊推理的三Ⅰ算法

模糊推理在控制和人工智能等领域已得到了广泛的应用，但其理论基础还不完善，王国俊教授提出的模糊命题逻辑的形式演绎系统和三Ⅰ算法为模糊推理奠定了严格的逻辑基础，本文把三Ⅰ算法用于区间值模糊推理，并且指出一般模糊推理是在区间退化为点时的特殊区间值模糊推理，从而把一般模糊推理纳入于区间值模糊推理的框架之内。     

一种基于神经元的自适应模糊推理网络

本文根据Ｔ－Ｓ模糊模型提出了一种新的基于神经元的自适应模糊推理网络，给出了连接结构和学习算法，它能自动学习和修正隶属函数及模糊规则，将其用于Ｂｏｘ的煤气炉，太阳黑子预报以及降雨量预报等不同类型的复杂系统建模，仿真结果表明，该模糊神经网络具有收敛速度快，辨识精度高，泛化能力强和适应范围广等特点，可当作复杂系统建模的一种有效工具。     

水泥产品结构的综合智能优化及应用研究

根据水泥市场需求信息,运用自适应模糊推理系统对水泥产品结构需求进行系统建模,应用并行遗传算法对模型求解,得到了来年的最优水泥产品结构需求计划,为水泥企业的生产规划及其经济效益的提高提供了重要的参考价值.     

基于Mamdani型模糊推理的震中烈度预测模型

对应用模糊推理进行系统预测进行了深入的研究,建立了以震级和震源深度为输入的基于Mamdani型模糊推理的震中烈度预测模型.并以四川地区震例数据为例,对数据信息提取,模糊规则建立等关键环节进行了详细的介绍,预测结果分析表明推理模型是可行和有效的.     

基于多元Hamacher-ANFIS的股票价格预测

应用减法-模糊聚类算法、多元Hamacher算子以及自适应神经模糊推理系统(ANFIS)提出了一种中国股票市场价格建模及预测的多元Hamacher-ANFIS模型.首先多元Hamacher算子与ANFIS相结合,对ANFIS种各规则的隶属度测度机制和规则参数更新机制进行了修正,建立基于减法-模糊聚类的多元HamacherANIFS模型;再从沪深两市各选取了总市值最大的5支股票,计算出它们在同一时间段的历史波动率,并以此为依据得到模型对该股票预测性能的权重;最后运用减法-模糊聚类算法初始化模型参数,对每个数据组进行5重交叉检验,并根据之前得到的权重计算出模型关于检验集的综合R2值.实验结果证明,与现有方法相比,该模型增强了对复杂目标函数的学习能力,提高了对股票价格的预测精度.     

高阶不确定非线性系统切换自适应镇定

研究了一类控制系数未知高阶不确定非线性系统的全局自适应稳定控制设计问题.值得指出的是,更高的系统幂次和未知的控制系数显示系统容许更强的非线性和更严重的不确定性/未知性,因而所研究的系统更具一般性.这里不是采用传统自适应技术,而是通过灵活运用基于切换的自适应技术和增加幂积分方法,给出了该控制问题的结构简单且易于实现的新型切换自适应控制器.所设计控制器保证闭环系统状态全局有界且最终趋于零.     

An adaptive neuro‐fuzzy inference system for islanding detection in wind turbine as distributed generation

This article proposes a new integrated diagnostic system for islanding detection by means of a neuro‐fuzzy approach. Islanding detection and prevention is a mandatory requirement for grid‐connected distributed generation (DG) systems. Several methods based on passive and active detection scheme have been proposed. Although passive schemes have a large non‐detection zone (NDZ), concern has been raised on active method due to its degrading power‐quality effect. Reliably detecting this condition is regarded by many as an ongoing challenge as existing methods are not entirely satisfactory. The main emphasis of the proposed scheme is to reduce the NDZ to as close as possible and to keep the output power quality unchanged. In addition, this technique can also overcome the problem of setting the detection thresholds inherent in the existing techniques. In this study, we propose to use a hybrid intelligent system called ANFIS (the adaptive neuro‐fuzzy inference system) for islanding detection. This approach utilizes rate of change of frequency (ROCOF) at the target DG location and used as the input sets for a neuro‐fuzzy inference system for intelligent islanding detection. This approach utilizes the ANFIS as a machine learning technology and fuzzy clustering for processing and analyzing the large data sets provided from network simulations using MATLAB software. To validate the feasibility of this approach, the method has been validated through several conditions and different loading, switching operation, and network conditions. The proposed algorithm is compared with the widely used ROCOF relays and found working effectively in the situations where ROCOF fails. Simulation studies showed that the ANFIS‐based algorithm detects islanding situation accurate than other islanding detection algorithms. © 2014 Wiley Periodicals, Inc. Complexity 21: 10–20, 2015     

Adaptive network-based fuzzy inference system with leave-one-out cross-validation approach for prediction of surface roughness

The prediction of surface roughness is a challengeable problem. In order to improve the prediction accuracy in end milling process, an improved approach is proposed to model surface roughness with adaptive network-based fuzzy inference system (ANFIS) and leave-one-out cross-validation (LOO-CV) approach. This approach focuses on both architecture and parameter optimization. LOO-CV, which is an effective measure to evaluate the generalization capability of mode, is employed to find the most suitable membership function and the optimal rule base of ANFIS model for the issue of surface roughness prediction. To find the optimal rule base of ANFIS, a new “top down” rules reduction method is suggested. Three machining parameters, the spindle speed, feed rate and depth of cut are used as inputs in the model. Based on the same experimental data, the predictive results of ANFIS with LOO-CV are compared with the results reported recently in the literature and ANFIS with clustering methods. The comparisons indicate that the presented approach outperforms the opponent methods, and the prediction accuracy can be improved to 96.38%. ANFIS with LOO-CV approach is an effective approach for prediction of surface roughness in end milling process.     

An adaptive network based fuzzy inference system–auto regression–analysis of variance algorithm for improvement of oil consumption estimation and policy making: The cases of Canada,United Kingdom,and South Korea

This paper presents an adaptive network based fuzzy inference system (ANFIS)–auto regression (AR)–analysis of variance (ANOVA) algorithm to improve oil consumption estimation and policy making. ANFIS algorithm is developed by different data preprocessing methods and the efficiency of ANFIS is examined against auto regression (AR) in Canada, United Kingdom and South Korea. For this purpose, mean absolute percentage error (MAPE) is used to show the efficiency of ANFIS. The algorithm for calculating ANFIS performance is based on its closed and open simulation abilities. Moreover, it is concluded that ANFIS provides better results than AR in Canada, United Kingdom and South Korea. This is unlike previous expectations that auto regression always provides better estimation for oil consumption estimation. In addition, ANOVA is used to identify policy making strategies with respect to oil consumption. This is the first study that introduces an integrated ANFIS–AR–ANOVA algorithm with preprocessing and post processing modules for improvement of oil consumption estimation in industrialized countries.     

A mathematical modeling for incorporating energy price hikes into total natural gas consumption forecasting

In some countries that energy prices are low, price elasticity of demand may not be significant. In this case, large increase or hike in energy prices may impact energy consumption in a way which cannot be drawn from historical data. This paper proposes an integrated adaptive fuzzy inference system (FIS) to forecast long-term natural gas (NG) consumption when prices experience large increase. To incorporate the impact of price hike into modeling, a novel procedure for construction and adaptation of Takagi–Sugeno fuzzy inference system (TS-FIS) is suggested. Linear regressions are used to construct a first order TS-FIS. Furthermore, adaptive network-based FIS (ANFIS) is used to forecast NG consumption in power plants. To cope with random uncertainty in small historical data sets, Monte Carlo simulation is utilized to generate training data for ANFIS. To show the applicability and usefulness of the proposed model, it is applied for forecasting of annual NG consumption in Iran where removing energy subsidies has resulted in a hike in NG prices.     

Shape optimal design of arch dams using an adaptive neuro-fuzzy inference system and improved particle swarm optimization

An efficient methodology is proposed to find the optimal shape of arch dams including fluid–structure interaction subject to earthquake ground motion. In order to reduce the computational cost of optimization process, an adaptive neuro-fuzzy inference system (ANFIS) is built to predict the dam effective response instead of directly evaluating it by a time-consuming finite element analysis (FEA). The presented ANFIS is compared with a widespread neural network termed back propagation neural network (BPNN) and it appears a better performance generality for estimating the dam response. The optimization task is implemented using an improved version of particle swarm optimization (PSO) named here as IPSO. In order to assess the effectiveness of the proposed methodology, the optimization of a real world arch dam is performed via both IPSO–ANFIS and PSO–BPNN approaches. The numerical results demonstrate the computational advantages of the proposed IPSO–ANFIS for optimal design of arch dams when compared with the PSO–BPNN approach.     

Design of hybrid genetic- SVD ANFIS networks for fatigue life modelling of GRP composites

Genetic algorithm (GA) and singular value decomposition (SVD) are deployed for the optimal design of both Gaussian membership functions of antecedents and the vector of linear coefficients of consequents, respectively, of adaptive neurofuzzy inference systems (ANFIS) networks that are used for fatigue life modelling and prediction of unidirectional GRP Composites. The aim of such modelling is to show how the fatigue life varies with the variation of important parameters namely, maximum stress, stress ratio, fiber angle. It is demonstrated that SVD can be effectively used to optimally find the vector of linear coefficients of conclusion parts in ANFIS models and their Gaussian membership functions in premise parts are determined by GA.     

Modeling ductile to brittle transition temperature of functionally graded steels by ANFIS

In the present paper, a model based on adaptive network-based fuzzy inference systems (ANFIS) for predicting ductile to brittle transition temperature of functionally graded steels in both crack divider and crack arrester configurations has been presented. Functionally graded steels containing graded ferritic and austenitic regions together with bainite and martensite intermediate layers were produced by electroslag remelting. To build the model, training and testing using experimental results from 140 specimens were conducted. The used data as inputs in ANFIS models are arranged in a format of six parameters that cover the FGS type, the crack tip configuration, the thickness of graded ferritic region, the thickness of graded austenitic region, the distance of the notch from bainite or martensite intermediate layer and temperature. According to these input parameters, in the ANFIS models, the ductile to brittle transition temperature of each FGS specimen was predicted. The training and testing results in the ANFIS models have shown a strong potential for predicting the ductile to brittle transition temperature of each FGS specimen.     

Adaptive network fuzzy inference system modeling for the input selection and prediction of anaerobic digestion effluent quality

This paper presents the development and evaluation of three adaptive network fuzzy inference system (ANFIS) models for a laboratory scale anaerobic digestion system outputs with varied input selection approaches. The aim was the investigation of feasibility of the approach-based-control system for the prediction of effluent quality from a sequential upflow anaerobic sludge bed reactor (UASBR) system that produced a strong nonlinearship between its inputs and outputs. As ANFIS demonstrated its ability to construct any nonlinear function with multiple inputs and outputs in many applications, its estimating performance was investigated for a complex wastewater treatment process at increasing organic loading rates from 1.1 to 5.5 g COD/L d. Approximation of the ANFIS models was validated using correlation coefficient, MAPE and RMSE. ANFIS was successful to model unsteady data for pH and acceptable for COD within anaerobic digestion limits with multiple input structure. The prediction performance showed a high feasibility of the model-based-control system on the anaerobic digester system to produce an effluent amenable for a consecutive aerobic treatment unit.     

Efficient modeling of MIMO systems through Timed Automata based Neuro-Fuzzy Inference Engine

This paper introduces a novel neuro-fuzzy approach for learning and modeling so-called Multi-Input Multi-Output Coupling (MIMO) systems, i.e., systems where the output variables may depend upon all system's input variables. This strong coupling makes the MIMO systems behavior highly oscillatory in time and, as a consequence, it makes these systems not particularly suitable to be learned and represented by using conventional approaches. In order to address this issue, our proposal presents an adaptive supervised learning algorithm capable of forming a suitable collection of Timed Automata based Fuzzy Systems that model the dynamic behavior of a given MIMO system. The adaptive learning is accomplished by taking advantage of the theories coming from the area of times series analysis (such as the Adaptive Piecewise Constant Approximation method) with a well-known neuro-fuzzy framework of the Adaptive Neuro Fuzzy Inference System (ANFIS). In experiments, where our proposal has been tested on the Fuzz-IEEE 2011 Fuzzy Competition dataset, the proposed supervised learning algorithm significantly reduces the output error measure and achieves better performance than the one provided by a conventional application of the ANFIS algorithm.     

Predicting effluent from the wastewater treatment plant of industrial park based on fuzzy network and influent quality

In this study, three types of adaptive neuro fuzzy inference system (ANFIS) were employed to predict effluent suspended solids (SS_eff), chemical oxygen demand (COD_eff), and pH_eff from a wastewater treatment plant in industrial park. For comparison, artificial neural network (ANN) was also used. The results indicated that ANFIS statistically outperformed ANN in terms of effluent prediction. The minimum mean absolute percentage errors of 2.67%, 2.80%, and 0.42% for SS_eff, COD_eff, and pH_eff could be achieved using ANFIS. The maximum values of correlation coefficient for SS_eff, COD_eff, and pH_eff were 0.96, 0.93, and 0.95, respectively. The minimum mean square errors of 0.19, 2.25, and 0.00, and the minimum root mean square errors of 0.43, 1.48, and 0.04 for SS_eff, COD_eff, and pH_eff could also be achieved. ANFIS’s architecture can overcome the limitations of traditional neural network. It also revealed that the influent indices could be applied to the prediction of effluent quality.