Neural Fuzzy Research Articles

Neural Network Identification and Direct Adaptive Fuzzy Neural Network (DAFNN) Controller for Unknown Nonlinear Non-affine Pneumatic Servo System

In this paper, a direct adaptive fuzzy neural network (DAFNN) controller for trajectory tracking control of the non-linear non-affine pneumatic servo system is presented. First, using a neural network identifier, the non-linear dynamics of a real pneumatic servo system is simulated. By comparing the output of the neural network and the output of the experimental setup, it is observed that the non-linear pneumatic actuator system is well-identified using neural networks. By incorporating the Lyapunov stability theorem, the adaptive laws for the parameters of the controller are obtained, parameter boundedness and stability of the closed-loop system are guaranteed. Finally, practical results are successfully implemented for trajectory tracking control of the pneumatic servo system, in which the effect of the simultaneous updating of the antecedent and consequent’s parameters of the fuzzy neural network controller has been investigated. The tracking error ±1.3mm and ±1 mm for proposed updating method compared to ±2.5mm and ±3.5mm, for a case that the weigh parameters are merely adjusted, are obtained. The results indicate the proposed adjustment method improves the performance of the controller in the presence of unknown nonlinearities and dynamics uncertainty.

A Neuro-Fuzzy Model for Conflict Prediction in the Niger Delta Region of Nigeria

This paper proposes a soft computing system driven by Neural Networks, Fuzzy Logic and Principal Component Analysis (PCA) for the prediction of conflict in the Niger Delta (ND) region of Nigeria. Identifying conflicts along the level of severity with which they arise in oil bearing host communities (OBHC) is of primary importance to government and other stakeholders, as the accompanying conflict risks mitigation course administered could be planned based on the level of severity of the conflict situations. The system is implemented using MATLAB and Microsoft Excel running on Microsoft Windows 10 operating system. The data set chosen for classification and experimental simulation is based on a statistical data obtained from a three-year field study of the nine states of the ND region of Nigeria. The average training and testing errors of 0.015514 and 0.053247 were obtained at 50 epochs for the model using a hybrid algorithm. PCA reduced the dimension of the original data set at a Cronbach’s alpha of over 0.8 with a 10-fold cross validation, thereby reducing the computational complexity and inference time of the model. The model predicted the conflict risk with an average accuracy of 92.85% and this compared favourably with domain experts conventional conflict prediction approaches. The result obtained gives a promising conclusion that the model is effective in predicting at high level of accuracy, the degree of conflict and presents a veritable decision support for conflict resolution and mediation agencies and stakeholders.

Convolutional Neural Network and Fuzzy Logic-based Hybrid Melanoma Diagnosis System

Studies on the detection of early stage melanoma have recently gained significant interest. Computer aided diagnosis systems based on neural networks, machine learning, convolutional neural networks (CNNs), and deep learning help early stage detection considerably. The colour and shapes of the images created by the pixels are crucial for the CNNs, as the pixels and associated pictures are interrelated just as a person’s fingerprint is unique. By observing this relationship, the pixel values of each picture with its neighborhoods were determined by a fuzzy logic-based system and a unique fingerprint matrix named Fuzzy Correlation Map (FCov-Map) was produced. The fuzzy logic system has four inputs and one output. The advantage of CNNs trained with fuzzy covariance maps is to eliminate both the limited availability of medical grade training data and the need for extensive image preprocessing. The fuzzy logic output is fed to the pretrained AlexNet CNN algorithm. To deliver a reliable result, a deep CNN needs a large amount of data to process. However, to obtain and use the required sufficient data for diseases is not cost- and time-effective. Therefore, the suggested fuzzy logic-based fuzzy correlation map is tackling this issue to solve the limitedness of training CNN data set.

A Comparative Analysis of Hidden Markov Model, Hybrid Support Vector Machines, and Hybrid Artificial Neural Fuzzy Inference System in Reservoir Inflow Forecasting (Case Study: The King Fahd Dam, Saudi Arabia)

The precise prediction of the streamflow of reservoirs is of considerable importance for many activities relating to water resource management, such as reservoir operation and flood and drought control and protection. This study aimed to develop and evaluate the applicability of a hidden Markov model (HMM) and two hybrid models, i.e., the support vector machine-genetic algorithm (SVM-GA) and artificial neural fuzzy inference system-genetic algorithm (ANFIS-GA), for reservoir inflow forecasting at the King Fahd dam, Saudi Arabia. The results obtained by the HMM model were compared with those for the two hybrid models ANFIS-GA and SVM-GA, and with those for individual SVM and ANFIS models based on performance evaluation indicators and visual inspection. The results of the comparison revealed that the ANFIS-GA model and ANFIS model provided superior results for forecasting monthly inflow with satisfactory accuracy in both training (R2 = 0.924, 0.857) and testing (R2 = 0.842, 0.810) models. The performance evaluation results for the developed models showed that the GA-induced improvement in the ANFIS and SVR forecasts was matched by an approximately 25% decrease in RMSE and around a 13% increase in Nash–Sutcliffe efficiency. The promising accuracy of the proposed models demonstrates their potential for applications in monthly inflow forecasting in the present semiarid region.

A Genetic-Based Hybrid Algorithm Harmonic Minimization Method for Cascaded Multilevel Inverters with ANFIS Implementation

Selective harmonic elimination pulse-width modulation (SHEPWM) is a widely adopted method to eliminate harmonics in multilevel inverters, yet solving harmonic amplitude equations is both time consuming and not accurate. This method is applied here for a 7-level cascaded multilevel inverter (CMLI) with erroneous DC sources. To meet the seven harmonic amplitude equations, two notches are applied with the use of higher switching frequency than nominal. These notches can be placed in six different positions in the voltage wave, and each was assessed in a separate manner. In order to solve the equations, a hybrid algorithm composed of genetic algorithm (GA) and Newton–Raphson (N-R) algorithm is applied to achieve faster convergence and maintain the accuracy of stochastic methods. At each step of the modulation index (M), different positions for the notches are compared based on the distortion factor (DF2%) benchmark, and the position with lowest DF2% is selected to train an artificial neural fuzzy interface system (ANFIS). ANFIS will receive the DC sources’ voltages together with required M and will produce one output; thus, eight ANFISs are applied to produce seven firing angles, and the remaining one is to determine which one of the notches’ positions should be used. Software simulations and experimental results confirm the validity of this proposed method. The proposed method achieves THD 8.45% when M is equal to 0.8 and is capable of effectively eliminating all harmonics up to the 19th order.

Optimization of Wells turbine performance using a hybrid artificial neural fuzzy inference system (ANFIS) - Genetic algorithm (GA)

Renewable energy resources are considered as alternative sources of energy for future generations. Wells turbines can be used to convert the reciprocating energy of waves in oceans and seas into rotational energy. Low aerodynamic efficiency is the main drawback of Wells turbines. The present study showed that the addition of the guide vanes, endplate, or end ring to the base blade significantly enhanced the performance of Wells turbines. Also, we used ANFIS-GA-CFD for optimization of the several design variables (the blade tip thickness, the blade form, and the input-output angels of guide vanes) simultaneously. Therefore, artificial neural fuzzy networks were used to forecast the three benchmark parameters (pressure drop coefficient, torque coefficient, and efficiency) and a genetic algorithm was utilized to approximate the optimized design variables when there are maximum torque, maximum efficiency, and minimum pressure drop. The optimal blade increased the upper limit of efficiency up to 18%, torque coefficient up to 8%, and reduced the pressure drop coefficient up to almost 3.45% compared to the best previous models. Moreover, flow separation was delayed.

Design Controller depend in his work on Neural Fuzzy Networks to Control DC Motor's Speed

Design Controller depend in his work on Neural Fuzzy Networks to Control DC Motor's Speed

Design Controller depend in his work on Neural Fuzzy Networks to Control DC Motor's Speed

Design Controller depend in his work on Neural Fuzzy Networks to Control DC Motor's Speed

Using Ultrasonic Sensors and a Knowledge-Based Neural Fuzzy Controller for Mobile Robot Navigation Control

This study proposes a knowledge-based neural fuzzy controller (KNFC) for mobile robot navigation control. An effective knowledge-based cultural multi-strategy differential evolution (KCMDE) is used for adjusting the parameters of KNFC. The KNFC is applied in PIONEER 3-DX mobile robots to achieve automatic navigation and obstacle avoidance capabilities. A novel escape approach is proposed to enable robots to autonomously avoid special environments. The angle between the obstacle and robot is used and two thresholds are set to determine whether the robot entries into the special landmarks and to modify the robot behavior for avoiding dead ends. The experimental results show that the proposed KNFC based on the KCMDE algorithm has improved the learning ability and system performance by 15.59% and 79.01%, respectively, compared with the various differential evolution (DE) methods. Finally, the automatic navigation and obstacle avoidance capabilities of robots in unknown environments were verified for achieving the objective of mobile robot control.

Prediction of Occupational Risks Using an Adaptive Neural Fuzzy Inference System in AZARAB Company

Prediction of Occupational Risks Using an Adaptive Neural Fuzzy Inference System in AZARAB Company

Towards a Better Integration of Fuzzy Matches in Neural Machine Translation through Data Augmentation

We identify a number of aspects that can boost the performance of Neural Fuzzy Repair (NFR), an easy-to-implement method to integrate translation memory matches and neural machine translation (NMT). We explore various ways of maximising the added value of retrieved matches within the NFR paradigm for eight language combinations, using Transformer NMT systems. In particular, we test the impact of different fuzzy matching techniques, sub-word-level segmentation methods and alignment-based features on overall translation quality. Furthermore, we propose a fuzzy match combination technique that aims to maximise the coverage of source words. This is supplemented with an analysis of how translation quality is affected by input sentence length and fuzzy match score. The results show that applying a combination of the tested modifications leads to a significant increase in estimated translation quality over all baselines for all language combinations.

Remote sensing based rain fall prediction using big data assisted integrated routing framework

At present, soft computing technology is the latest, most challenging technology used in prediction analysis of weather parameters. Therefore, the accuracy of the prediction in the area of rainfall analysis is very important. However, present methods in predicting rainfalls have less accuracy. In many cases, weather prediction models function poorly. Machine Learning methods ignore the effect of physical factors that make rainfall forecasting. It improves the accuracy of rainfall prediction and analysis and aims to predict rainfall with the related air density, relative humidity and temperature parameters. The most recent advanced remote sensing techniques with Artificial Neural Network model and Fuzzy Logic Control (FLC) associated with big data Assisted Integrated Routing and Surplus Memory (BIRSM) model has been developed with the available India Meteorological Department (IMD) data for the year from to Data preprocessing techniques are applied to the volume of available data. An appropriate and error-reducing evaluation for rainfall prediction is performed in combination with the proposed BIRSM model and Artificial Neural Network. Accurate and time series prediction can be very helpful for agriculture and flood management etc.

WITHDRAWN: Utilizing schemes for detecting the threat levels for chronic kidney disease

Abstract The severe kidney ailment (CKD) is normally symbolized by a regular loss of features with kidneys with over time because of several metrics. The initial detection and treatment save the kidney and stops the improvement of CKD. The CKD disease is perceived as a universal public health dispute over earlier times. The biggest risk for this fatal disease in developing countries with gathering therapy is costly. The vitality of diagnosing the individual with the threat of CKD is by making using grouping schemes cannot be underrated as they can change the series of ailments. The location of these silent killer ailments initially offered the best chances for executing the possible policies for minimizing the possibility of kidney losses. The neural fuzzy scheme is employed for finding the threats of CKD in patients. The detection is carried out by using the neural fuzzy with the precision of 97%. The selected characteristics and detection of CKD ailments are carried out to locate the threats. The outcomes of the detecting are grouped to locate the ratio of patients with improved risk of kidney ailments with improved possibilities of being diabetic. Based on the tree-based grouping three groups are created with strong associations prevailing amongst the chronic kidney and diabetes.

Safety Prediction of Deep Foundation Pit Based on Neural Network and Entropy Fuzzy Evaluation

The monitoring data can effectively reflect the safety status of the project during the construction of deep foundation pit, and the risks existing in the project can be discovered in time and the development trend can be reasonably predicted through the processing and analysis of the existing monitoring data. In this paper, a deep foundation pit in compound soil area of a coastal city was taken as an example, the BP neural network was taken to predict the monitoring data in the next stage, the entropy method was utilized to determine the weight of the evaluation index according to the predicted value, and the fuzzy comprehensive evaluation method was used to quantitatively describe the future safety status, so as to formulate targeted countermeasures and improve the construction safety.

A Comparative Study Between Artificial Neural Networks and Fuzzy Inference System for Estimation and Filling of Missing Runoff Data at Al-Jawadiyah Station

Runoff is one of the most important components of the hydrological cycle, and having complete series of runoff data is essential for any hydrological modelling process. This study compares artificial neural networks and fuzzy inference systems for estimation of runoff data at the Al-Jawadiyah hydrometric station. This study used only the runoff data at Al-Jawadiyah station in addition to the runoff values measured at Al-Amiri station on the Syrian-Lebanese border. Many experiments were conducted, and a very large number of artificial neural networks were trained with changing the number of hidden layers, the number of neurons, and the training algorithms until the best network was reached according to the regression criteria and the root mean of the error squares between the measured values and the predicted values. Also, many fuzzy inference models have been prepared, changing the number and type of membership functions until the most accurate model has been reached. The results showed the high reliability of both the artificial neural network models and the fuzzy inference models in estimating runoff in the study area, and the comparison between the results showed the great convergence of the two models with a slight preference for the fuzzy models. This study recommends using the rest of the artificial intelligence models and comparing them to arrive at the most accurate model. This helps prepare a complete series of hydrological and climatic measurements that form a basis for preparing an accurate hydrological model for the study area.

Investigation of the Possibility of Building a Neural Fuzzy Logic Controller with Discrete Terms for Controlling and Automating Oil and Gas Engineering Facilities

The issue of increasing the efficiency of functioning of classical control systems for technological processes and objects of oil and gas engineering is investigated. The relevance of this topic lies in the need to improve the quality of the control systems for the production and transportation of oil and gas. The purpose of the scientific work is to develop a neuro-fuzzy logic controller with discrete terms for the control and automation of pumping units and pumping stations. It is noted that fuzzy logic, neural network algorithms, together with control methods based on adaptation and synthesis of control objects, make it possible to learn the automation system and work under conditions of uncertainty. Methods for constructing classical control systems are studied, the advantages and disadvantages of fuzzy controllers, as the main control system, are analyzed. A method for constructing a control system based on a neuro-fuzzy controller with discrete terms in conditions of uncertainty and dynamic parameters of the process is proposed. The positive features of the proposed regulator include a combination of fuzzy reasoning about a technological object and mathematical predictive models, a fuzzy control system gains the possibility of subjective description based on neural network structures, as well as adaptation to the characteristics of the object. The graph of dependence for the term-set of the controlled parameter on the degree of membership is presented. A possible implementation of tracking the triggering of one of the rules of the neuro-fuzzy system in the format of functional block diagrams is presented. The process of forming an expert knowledge base in a neuro-fuzzy control system is considered. For analysis, a graph of the dependence of the output parameter values is shown. According to the results obtained, the deviation of the values for the model and the real process does not exceed 18%, which allows us to speak of a fairly stable operation of the neuro-fuzzy controller in automatic control systems.

MANAGING TARGET COSTS AT THE DEVELOPMENT STAGE OF A PRODUCT

At the development stage of a product, target cost management has a significant effect on achievement and maintenance of product competitiveness in a given market and the desired target profit as the primary aim of production. Target cost management, i.e., target profit management, is performed by evaluating individual variations of a conceptual and preliminary design solution for a new product and by choosing the solution with the lowest development costs. The paper presents the use of Adaptive Neural Fuzzy Inference System model for managing development costs on the example of double-row angular contact ball bearings as an example of applying the above approach to market positioning in relation to competitors.

Design of a Neural Fuzzy Damper Vibration to Improve the Stability of Electric Power Systems

Design of a Neural Fuzzy Damper Vibration to Improve the Stability of Electric Power Systems

Water quality control of Tasik Kejuruteraan UKM water channel using Artificial Neural Network and Neural Fuzzy Network

The water quality of Tasik Kejuruteraan UKM water channel is an issue that should be emphasized because excess water will be discharged to Sungai Langat where this river is the source of water supply for the residents in some part of Klang Valley area. The water quality parameters must comply with the standard assigned by Malaysian’s Department of Environment (DOE). As a first step towards proper monitoring programme, a proper analysis of the water quality should be conducted thoroughly. As such, this study composed of two parts. First, experimental investigation of the water quality of Tasik Kejuruteraan UKM water channel was carried out to assess its environmental quality. Experiment was conducted to determine the Water Quality Index of Tasik Kejuruteraan water channel based upon guidelines as stipulated by DOE. The results indicate that Tasik Kejuruteraan UKM water channel is polluted and proper water quality control should be applied. In the second part of this study, preliminary analysis of artificial neural network (ANN) and neural fuzzy network for water quality control was conducted to determine the suitability of these controllers in water treatment system. Based on the simulations conducted, neural fuzzy network shows better efficacy in the pH control scheme compared with ANN because there is no oscillation detected in the corresponding controlled variable.

Enriching E Commerce Fraud Detection by using Machine Learning

As there has been a proliferation of the internet platform, it has been increasingly getting affordable for a lot of individuals. The rise has been instrumental in achieving several services including the E-commerce platform. This has led to an unprecedented increase in the amount of fraud that is being committed on this platform. The fraud that is being committed on the E-commerce platforms is very different from the frauds committed on other platforms online. Numerous researches have been performed to combat the evils of credit card frauds and money laundering rings. But there is a severe lack of research on the fraud that is committed on the E-commerce platform. Therefore, this research paper defines an innovative approach for the identification of fraud on E-commerce platforms through the implementation of machine learning approaches. The presented technique utilizes Linear Clustering, Entropy Estimation and Frequent itemset mining in addition to the inclusion of Artificial Neural Networks, Hypergraph formation and Fuzzy classification. The implementation of this system will give more security for E-commerce platform-based transactions by identifying fraudulent activities with better efficiency. The methodology has been tested extensively through rigorous experimentation to evaluate the performance metrics which yielded significantly positive results.