A TSK-Type Recurrent Neuro-Fuzzy Systems for Fault Prognosis

As a result from the demanding of process safety, reliability and environmental constraints, a called of fault detection and diagnosis system become more and more important. In this article some basic aspects of TSK (Takigi Sugeno Kang) neuro-fuzzy techniques for the prognosis and diagnosis of manufacturing systems are presented. In particular, a neuro-fuzzy model that can be used for the identification and the simulation of faults prognosis models is described. The presented model is motivated by a cooperative neuro-fuzzy approach based on a vectorized recurrent neural network architecture. The neuro-fuzzy architecture maps the residuals into two classes: a one of fixed direction residuals and another one of faults belonging to rotary kiln.

An Automatic Threshold Selection Using ALO for Healthcare Duplicate Record Detection with Reciprocal Neuro-Fuzzy Inference System

ESystems based on EHRs(Electronic health records)have been in use for many years and their amplified realizations have been felt recently.They still have been pioneering collections of massive volumes of health data.Duplicate detections involve discovering records referring to the same practical components,indicating tasks,which are generally dependent on several input parameters that experts yield.Record linkage specifies the issue of finding identical records across various data sources.The similarity existing between two records is characterized based on domain-based similarity functions over different features.De-duplication of one dataset or the linkage of multiple data sets has become a highly significant operation in the data processing stages of different data mining programmes.The objective is to match all the records associated with the same entity.Various measures have been in use for representing the quality and complexity about data linkage algorithms,and many other novel metrics have been introduced.An outline of the problem existing in themeasurement of data linkage and de-duplication quality and complexity is presented.This article focuses on the reprocessing of health data that is horizontally divided among data custodians,with the purpose of custodians giving similar features to sets of patients.The first step in this technique is about an automatic selection of training examples with superior quality from the compared record pairs and the second step involves training the reciprocal neuro-fuzzy inference system(RANFIS)classifier.Using the Optimal Threshold classifier,it is presumed that there is information about the original match status for all compared record pairs(i.e.,Ant Lion Optimization),and therefore an optimal threshold can be computed based on the respective RANFIS.Febrl,Clinical Decision(CD),and Cork Open Research Archive(CORA)data repository help analyze the proposed method with evaluated benchmarks with current techniques.

Data Analytics on Unpredictable Pregnancy Data Records Using Ensemble Neuro-Fuzzy Techniques

The immune system goes through a profound transformation during pregnancy,and certain unexpected maternal complications have been correlated to this transition.The ability to correctly examine,diagnoses,and predict pregnancy-hastened diseases via the available big data is a delicate problem since the range of information continuously increases and is scalable.Many approaches for disease diagnosis/classification have been established with the use of data mining concepts.However,such methods do not provide an appropriate classification/diagnosis model.Furthermore,single learning approaches are used to create the bulk of these systems.Classification issues may be made more accurate by combining predictions from many different techniques.As a result,we used the Ensembling of Neuro-Fuzzy(E-NF)method to perform a high-level classification of medical diseases.E-NF is a layered computational model with self-learning and self-adaptive capabilities to deal with specific problems,such as the handling of imprecise and ambiguous data that may lead to uncertainty concerns that specifically emerge during the classification stage.Preprocessing data,Training phase,Ensemble phase,and Testing phase make up the complete procedure for the suggested task.Data preprocessing includes feature extraction and dimensionality reduction.Besides such processes,the training phase includes the fuzzification process of medical data.Moreover,training of input data was done using four types of NF techniques:Fuzzy Adaptive Learning Control Network(FALCON),Adaptive Network-based Fuzzy Inference System(ANFIS),Self Constructing Neural Fuzzy Inference Network(SONFIN)and/Evolving Fuzzy Neural Network(EFuNN).Later,in the ensemble phase,all the NF methods’predicted outcomes are integrated,and finally,the test results are evaluated in the testing phase.The outcomes indicate that the method could predict impaired glucose tolerance,preeclampsia,gestational hypertensive abnormalities,bacteriuria,and iron deficiency anaemia better than the others.In addition,the model exposed the capability to be utilized as an autonomous learning strategy,specifically in the early stages of pregnancy,examinations,and clinical guidelines for disease interventions.

Training Neuro-Fuzzy by Using Meta-Heuristic Algorithms for MPPT

It is one of the topics that have been studied extensively on maximum power point tracking(MPPT)recently.Traditional or soft computing methods are used for MPPT.Since soft computing approaches are more effective than traditional approaches,studies on MPPT have shifted in this direction.This study aims comparison of performance of seven meta-heuristic training algorithms in the neuro-fuzzy training for MPPT.The meta-heuristic training algorithms used are particle swarm optimization(PSO),harmony search(HS),cuckoo search(CS),artificial bee colony(ABC)algorithm,bee algorithm(BA),differential evolution(DE)and flower pollination algorithm(FPA).The antecedent and conclusion parameters of neuro-fuzzy are determined by these algorithms.The data of a 250 W photovoltaic(PV)is used in the applications.For effective MPPT,different neuro-fuzzy structures,different membership functions and different control parameter values are evaluated in detail.Related training algorithms are compared in terms of solution quality and convergence speed.The strengths and weaknesses of these algorithms are revealed.It is seen that the type and number of membership function,colony size,number of generations affect the solution quality and convergence speed of the training algorithms.As a result,it has been observed that CS and ABC algorithm are more effective than other algorithms in terms of solution quality and convergence in solving the related problem.

An Adaptive Neuro-Fuzzy Inference System to Improve Fractional Order Controller Performance

The design and analysis of a fractional order proportional integral deri-vate(FOPID)controller integrated with an adaptive neuro-fuzzy inference system(ANFIS)is proposed in this study.Afirst order plus delay time plant model has been used to validate the ANFIS combined FOPID control scheme.In the pro-posed adaptive control structure,the intelligent ANFIS was designed such that it will dynamically adjust the fractional order factors(λandµ)of the FOPID(also known as PIλDµ)controller to achieve better control performance.When the plant experiences uncertainties like external load disturbances or sudden changes in the input parameters,the stability and robustness of the system can be achieved effec-tively with the proposed control scheme.Also,a modified structure of the FOPID controller has been used in the present system to enhance the dynamic perfor-mance of the controller.An extensive MATLAB software simulation study was made to verify the usefulness of the proposed control scheme.The study has been carried out under different operating conditions such as external disturbances and sudden changes in input parameters.The results obtained using the ANFIS-FOPID control scheme are also compared to the classical fractional order PIλDµand conventional PID control schemes to validate the advantages of the control-lers.The simulation results confirm the effectiveness of the ANFIS combined FOPID controller for the chosen plant model.Also,the proposed control scheme outperformed traditional control methods in various performance metrics such as rise time,settling time and error criteria.

Intuitionistic Neuro-Fuzzy Optimization in the Management of Medical Diagnosis

Diabetes has become a major concern nowadays and its complications are affecting various organs of a diabetic patient. Therefore, a multi-dimensional technique including all parameters is required to detect the cause, its proper diagnostic procedure and its prevention. In this present work, a technique has been introduced that seeks to build an implementation for the intelligence system based on neural networks. Moreover, it has been described that how the proposed technique can be used to determine the membership together with the non-membership functions in the intuitionistic environment. The dataset has been obtained from Pima Indians Diabetes Database (PIDD). In this work, a complete diagnostic procedure of diabetes has been introduced with seven layered structural frameworks of an Intuitionistic Neuro Sugeno Fuzzy System (INSFS). The first layer is the input, in which six factors have been taken as an input variable. Subsequently, a neural network framework has been developed by constructing IFN for all the six input variables, and then this input has been fuzzified by using triangular intuitionistic fuzzy numbers. In this work, we have introduced a novel optimization technique for the parameters involved in the INSFS. Moreover, an inference system has also been framed for the neural network known as INFS. The results have also been given in the form of tables, which describe each concluding factor.

基于Neuro-Fuzzy方法的Web服务器访问流量预测

Neuro Fuzzy方法是将神经网络和模糊逻辑有机的结合 ,用于解决复杂的非线性问题 ;用它来进行Web服务器流量预测 ,是一种新的思路和方法 .主要介绍了模型构造的基本思想、结构。

基于Neuro-Fuzzy方法的Web服务器访问流量预测

Neuro-Fuzzy方法是将神经网络和模糊逻辑进行有机的结合，用于解决复杂的非线性问题；用它来进行Web服务器流量预测，是一种新的思路和方法。该文介绍了模型构造的基本思想、结构、算法，也介绍了进化式聚类方法和预测过程；同时，给出了实验数据及分析。

Characteristics Prediction Method of Electro-hydraulic Servo Valve Based on Rough Set and Adaptive Neuro-fuzzy Inference System

Synthesis characteristics of the electro-hydraulic servo valve are key factors to determine eligibility of the hydraulic production.Testing all synthesis characteristics of the electro-hydraulic servo valve after assembling leads to high repair rate and reject rate,so accurate prediction for the synthesis characteristics in the industrial production is particular important in decreasing the repair rate and the reject rate of the product.However,the research in forecasting synthesis characteristics of the electro-hydraulic servo valve is rare.In this work,a hybrid prediction method was proposed based on rough set(RS) and adaptive neuro-fuzzy inference system(ANFIS) in order to predict synthesis characteristics of electro-hydraulic servo valve.Since the geometric factors affecting the synthesis characteristics of the electro-hydraulic servo valve are from workers' experience,the inputs of the prediction method are uncertain.RS-based attributes reduction was used as the preprocessor,and then the exact geometric factors affecting the synthesis characteristics of the electro-hydraulic servo valve were obtained.On the basis of the exact geometric factors,ANFIS was used to build the final prediction model.A typical electro-hydraulic servo valve production was used to demonstrate the proposed prediction method.The prediction results showed that the proposed prediction method was more applicable than the artificial neural networks(ANN) in predicting the synthesis characteristics of electro-hydraulic servo valve,and the proposed prediction method was a powerful tool to predict synthesis characteristics of the electro-hydraulic servo valve.Moreover,with the use of the advantages of RS and ANFIS,the highly effective forecasting framework in this study can also be applied to other problems involving synthesis characteristics forecasting.

A novel approach to determine residual stress field during FSW of AZ91 Mg alloy using combined smoothed particle hydrodynamics/neuro-fuzzy computations and ultrasonic testing

The faults in welding design and process every so often yield defective parts during friction stir welding(FSW).The development of numerical approaches including the finite element method(FEM)provides a way to draw a process paradigm before any physical implementation.It is not practical to simulate all possible designs to identify the optimal FSW practice due to the inefficiency associated with concurrent modeling of material flow and heat dissipation throughout the FSW.This study intends to develop a computational workflow based on the mesh-free FEM framework named smoothed particle hydrodynamics(SPH)which was integrated with adaptive neuro-fiizzy inference system(ANFIS)to evaluate the residual stress in the FSW process.An integrated SPH and ANFIS methodology was established and the well-trained ANIS was then used to predict how the FSW process depends on its parameters.To verify the SPH calculation,an itemized FSW case was performed on AZ91 Mg alloy and the induced residual stress was measured by ultrasonic testing.The suggested methodology can efficiently predict the residual stress distribution throughout friction stir welding of AZ91 alloy.

Detection of small bowel tumor in wireless capsule endoscopy images using an adaptive neuro-fuzzy inference system

Automatic diagnosis tool helps physicians to evaluate capsule endoscopic examinations faster and more accurate.The purpose of this study was to evaluate the validity and reliability of an automatic post-processing method for identifying and classifying wireless capsule endoscopic images, and investigate statistical measures to differentiate normal and abnormal images. The proposed technique consists of two main stages, namely, feature extraction and classification. Primarily, 32 features incorporating four statistical measures(contrast, correlation, homogeneity and energy) calculated from co-occurrence metrics were computed. Then, mutual information was used to select features with maximal dependence on the target class and with minimal redundancy between features. Finally, a trained classifier, adaptive neuro-fuzzy interface system was implemented to classify endoscopic images into tumor, healthy and unhealthy classes. Classification accuracy of 94.2% was obtained using the proposed pipeline. Such techniques are valuable for accurate detection characterization and interpretation of endoscopic images.

Robust adaptive neuro-fuzzy control of uncertain nonholonomic systems

In this paper, we present an adaptive neuro-fuzzy controller design for a class of uncertain nonholonomic systems in the perturbed chained form with unknown virtual control coefficients and strong drift nonlinearities. The robust adaptive neuro-fuzzy control laws are developed using state scaling and backstepping. Semiglobal uniform ultimate bound-edness of all the signals in the closed-loop are guaranteed, and the system states are proven to converge to a small neigh-borhood of zero. The control performance of the closed-loop system is guaranteed by appropriately choosing the design parameters. By using fuzzy logic approximation, the proposed control is free of control singularity problem. An adaptive control-based switching strategy is proposed to overcome the uncontrollability problem associated with x 0 (t 0 ) = 0.

A chemical-reaction-optimization-based neuro-fuzzy hybrid network for stock closing price prediction

Accurate prediction of stock market behavior is a challenging issue for financial forecasting.Artificial neural networks,such as multilayer perceptron have been established as better approximation and classification models for this domain.This study proposes a chemical reaction optimization(CRO)based neuro-fuzzy network model for prediction of stock indices.The input vectors to the model are fuzzified by applying a Gaussian membership function,and each input is associated with a degree of membership to different classes.A multilayer perceptron with one hidden layer is used as the base model and CRO is used to the optimal weights and biases of this model.CRO was chosen because it requires fewer control parameters and has a faster convergence rate.Five statistical parameters are used to evaluate the performance of the model,and the model is validated by forecasting the daily closing indices for five major stock markets.The performance of the proposed model is compared with four state-of-art models that are trained similarly and was found to be superior.We conducted the Deibold-Mariano test to check the statistical significance of the proposed model,and it was found to be significant.This model can be used as a promising tool for financial forecasting.

Method to Improve Airborne Pollution Forecasting by Using Ant Colony Optimization and Neuro-Fuzzy Algorithms

This contribution shows the feasibility of improving the modeling of the non-linear behavior of airborne pollution in large cities. In previous works, models have been constructed using many machine learning algorithms. However, many of them do not work for all the pollutants, or are not consistent or robust for all cities. In this paper, an improved algorithm is proposed using Ant Colony Optimization (ACO) employing models created by a neuro-fuzzy system. This method results in a reduction of prediction error, which results in a more reliable prediction models obtained.

Column breakthrough studies for the removal and recovery of phosphate by lime-iron sludge:Modeling and optimization using artificial neural network and adaptive neuro-fuzzy inference system

Increases in the treatment of water to meet the growing water demand ultimately result in unmanageable quantities of residuals,the handling,and disposal of which is a major environmental issue.Consequently,research into beneficial reuse of water treatment residuals continues unabated.This study investigated the applicability of lime-iron sludge for phosphate adsorption by fixed-bed column adsorption.Laboratory-scale experiments were conducted at varying flow rates and bed depths.Fundamental and empirical models(Thomas,Yan,Bohart-Adams,Yoon-Nelson,and Wolboroska)as well as artificial intelligence techniques(Artificial neural network(ANN)and Adaptive neuro-fuzzy inference system(ANFIS))were used to simulate experimental breakthrough curves and predict column dynamics.Increase in flow rate resulted in reduced adsorption capacity.However,adsorption capacity was not affected by bed depth.ANN was superior in predicting breakthrough curves and predicted breakthrough times with high accuracy(R^2>0.9962).Na OH(0.5 mol·L^-1)was successfully used to regenerate the adsorption bed.After nine cyclic adsorption/desorption runs,only a marginal decrease in adsorption and desorption efficiencies of 10%and 8%respectively was observed.The same regenerate Na OH solution was reused for all desorption cycles.After nine cycles the eluent desorbed a total of 1550 mg phosphate exhibiting potential for further reuse.

A Neuro-Fuzzy Approach to Road Traffic Congestion Prediction

The fast-paced growth of artificial intelligence applications provides unparalleled opportunities to improve the efficiency of various systems.Such as the transportation sector faces many obstacles following the implementation and integration of different vehicular and environmental aspects worldwide.Traffic congestion is among the major issues in this regard which demands serious attention due to the rapid growth in the number of vehicles on the road.To address this overwhelming problem,in this article,a cloudbased intelligent road traffic congestion prediction model is proposed that is empowered with a hybrid Neuro-Fuzzy approach.The aim of the study is to reduce the delay in the queues,the vehicles experience at different road junctions across the city.The proposed model also intended to help the automated traffic control systems by minimizing the congestion particularly in a smart city environment where observational data is obtained from various implanted Internet of Things(IoT)sensors across the road.After due preprocessing over the cloud server,the proposed approach makes use of this data by incorporating the neuro-fuzzy engine.Consequently,it possesses a high level of accuracy by means of intelligent decision making with minimum error rate.Simulation results reveal the accuracy of the proposed model as 98.72%during the validation phase in contrast to the highest accuracies achieved by state-of-the-art techniques in the literature such as 90.6%,95.84%,97.56%and 98.03%,respectively.As far as the training phase analysis is concerned,the proposed scheme exhibits 99.214% accuracy. The proposed prediction modelis a potential contribution towards smart cities environment.

An evolutionary adaptive neuro-fuzzy inference system for estimating field penetration index of tunnel boring machine in rock mass

Field penetration index(FPI) is one of the representative key parameters to examine the tunnel boring machine(TBM) performance.Lack of accurate FPI prediction can be responsible for numerous disastrous incidents associated with rock mechanics and engineering.This study aims to predict TBM performance(i.e.FPI) by an efficient and improved adaptive neuro-fuzzy inference system(ANFIS) model.This was done using an evolutionary algorithm,i.e.artificial bee colony(ABC) algorithm mixed with the ANFIS model.The role of ABC algorithm in this system is to find the optimum membership functions(MFs) of ANFIS model to achieve a higher degree of accuracy.The procedure and modeling were conducted on a tunnelling database comprising of more than 150 data samples where brittleness index(BI),fracture spacing,α angle between the plane of weakness and the TBM driven direction,and field single cutter load were assigned as model inputs to approximate FPI values.According to the results obtained by performance indices,the proposed ANFISABC model was able to receive the highest accuracy level in predicting FPI values compared with ANFIS model.In terms of coefficient of determination(R^(2)),the values of 0.951 and 0.901 were obtained for training and testing stages of the proposed ANFISABC model,respectively,which confirm its power and capability in solving TBM performance problem.The proposed model can be used in the other areas of rock mechanics and underground space technologies with similar conditions.

NEURO-FUZZY NETWORKS IN CAPP

The neuro-fuzzy network (NFN) is used to model the rules and experience of the process planner. NFN is to select the manufacturing operations sequences for the part features. A detailed description of the NFN system development is given. The rule structure utilizes sigmoid functions to fuzzify the inputs, multiplication to combine the if Part of the rules and summation to integrate the fired rules. Expert knowledge from previous process Plans is used in determinning the initial network structure and parameters of the membership functions. A back-propagation (BP) training algorithm was developed to fine tune the knowledge to company standards using the input-output data from executions of previous plans. The method is illustrated by an industrial example.

Adaptive Neuro-Fuzzy Logic System for Heavy Metal Sorption in Aquatic Environments

In this paper, adaptive neuro-fuzzy inference system ANFIS is used to assess conditions required for aquatic systems to serve as a sink for metal removal;it is used to generate information on the behavior of heavy metals (mercury) in water in relation to its uptake by bio-species (e.g. bacteria, fungi, algae, etc.) and adsorption to sediments. The approach of this research entails training fuzzy inference system by neural networks. The process is useful when there is interrelation between variables and no enough experience about mercury behavior, furthermore it is easy and fast process. Experimental work on mercury removal in wetlands for specific environmental conditions was previously conducted in bench scale at Concordia University laboratories. Fuzzy inference system FIS is constructed comprising knowledge base (i.e. premises and conclusions), fuzzy sets, and fuzzy rules. Knowledge base and rules are adapted and trained by neural networks, and then tested. ANFIS simulates and predicts mercury speciation for biological uptake and mercury adsorption to sediments. Modeling of mercury bioavailability for bio-species and adsorption to sediments shows strong correlation of more than 98% between simulation results and experimental data. The fuzzy models obtained are used to simulate and forecast further information on mercury partitioning to species and sediments. The findings of this research give information about metal removal by aquatic systems and their efficiency.

Software Reliability Assessment Using Hybrid Neuro-Fuzzy Model

Software reliability is the primary concern of software developmentorganizations, and the exponentially increasing demand for reliable softwarerequires modeling techniques to be developed in the present era. Small unnoticeable drifts in the software can culminate into a disaster. Early removal of theseerrors helps the organization improve and enhance the software’s reliability andsave money, time, and effort. Many soft computing techniques are available toget solutions for critical problems but selecting the appropriate technique is abig challenge. This paper proposed an efficient algorithm that can be used forthe prediction of software reliability. The proposed algorithm is implementedusing a hybrid approach named Neuro-Fuzzy Inference System and has also beenapplied to test data. In this work, a comparison among different techniques of softcomputing has been performed. After testing and training the real time data withthe reliability prediction in terms of mean relative error and mean absolute relativeerror as 0.0060 and 0.0121, respectively, the claim has been verified. The resultsclaim that the proposed algorithm predicts attractive outcomes in terms of meanabsolute relative error plus mean relative error compared to the other existingmodels that justify the reliability prediction of the proposed model. Thus, thisnovel technique intends to make this model as simple as possible to improvethe software reliability.