Neural Fuzzy System Research Articles

Fuzzy deep learning-based crop yield prediction model for sustainable agronomical frameworks

The evolution in science and innovation has to lead to an immense volume of information from various agricultural fields to be accumulated in the public domain. As a result, an objective arises from the investigation of the accessible information and incorporating them with processes like foreseeing crop yield, plant diseases examination, crops enhancement, etc. Machine learning has grown with tremendous processing methods to conceive new innovations in the multi-disciplinary agricultural sector. In experimenting with machine learning models, there exist certain limitations like improvident nonlinear mapping between the raw data and crop yield values. Hence, deep learning models are comprehensively used to extricate critical crop parameters for prediction. Foreseeing the crop yield depending on climate, soil and water parameters has been a potential research subject. This paper proposes a hybrid deep learning-based crop yield prediction system using deep belief network (DBN) and fuzzy neural networks system (FNN). DBN is a combination of statistics and probability with neural networks. Though DBN performs better for nonlinear systems, the algorithm alone cannot provide satisfactory results in terms of robustness, model accuracy and learning speed, which is predominantly due to gradient diffusion. Hence, a DBN along with FNN has been proposed to overcome the nonlinearity and gradient diffusion problems. The proposed model initially performs an efficient pre-training technique by DBN for enhanced model development and feature vector generation. This characteristic feature vector is fed as an input to the FNN for further processing. The superiority of the proposed fuzzy neural network-based deep belief network is analyzed by comparing it with other deep learning algorithms. The proposed model efficiently predicts the results outperforming the other models by preserving the original data distribution with an accuracy of 92%.

ANFIS-Based Optimum Design of Real Power Transmission Towers with Size, Shape and Panel Design Variables using BBO Algorithm

The present study aims to provide a suitable approach to optimize transmission line towers with size, shape and panel design variables. MSTOWER software was used for modeling, analysis and design of the transmission tower. The design requirements applied to the structure were in accordance with ASCE10-97 standard. Transmission line towers are optimized in two ways. The first method combines Biogeography based Optimization (BBO) algorithm with MSTOWER software called BBO-MSTOWER and the second method, called BBO-ANFIS, uses an adaptive fuzzy neural inference system (ANFIS). To analyze the structure in the optimization process with the aim of reducing the computations and time of the optimization process. In order to evaluate the proposed method, two full-scale transmission towers were optimized as numerical models using the above two methods. Finally, the optimal design and time required in the two optimization methods were compared with each other as well as with the initial design of the tower and the results reported in previous studies. The results showed that with acceptable accuracy, ANFIS resulted in a significant reduction in the calculation time of the entire optimization process.

PATTERN OF TRADING PARTNER SELECTION IN DEPUTIZATION SYSTEMS BASED ON ADAPTIVE NEURO- FUZZY INFERENCE SYSTEM

Purpose: The purpose of this study is to develop a model for selecting a business partner in agency systems based on the method of the adaptive neural-fuzzy system.
 Methodology: The present research is applied in terms of purpose and descriptive in terms of the research method. The statistical population of the study, based on the subject of the research, the objectives of the research, and the spatial scope of the research, includes 98 agencies of Parsian Insurance Company in East Azarbaijan Province. According to the available statistics, the number of agencies of Parsian Insurance Company in East Azarbaijan Province is 98; Given that designed systems require more samples to arrive at the right answer. Therefore, the sample size will be done using the all-count sampling method. A questionnaire was used to collect the data of the input variables and the sales amount of different types of insurance policies was used for the output part. An adaptive neurophysiological system (ANFIS) has been used to analyze the data. Also, to evaluate the performance of each of the designed systems, the characteristics of the mean error squares and the root mean of the mean error squares were used.
 Main Findings: The research findings show that the best model designed to select a business partner in agency systems is a system with foot membership functions, some repetitions of 30, and two membership functions at each input.
 Application of Study: The results of this study can be used in agency systems to select business partners.
 Novelty/Originality: The novelty of this study is developing a model for selecting a business partner in agency systems based on the method of the adaptive neural-fuzzy system.

Adaptive Type-2 Fuzzy Neural Network Inherited Terminal Sliding Mode Control for Power Quality Improvement

This article proposes an adaptive type-2 fuzzy neural network control system to enhance the performance of power quality improvement. First, the dynamic model of APF with lumped uncertainties caused by parameter perturbation of ac inductor and dc capacitor is briefly introduced. Then, an integral-type terminal sliding mode control (TSMC) is developed for the finite-time reference signal tracking. Meanwhile, in terms of the considered chattering problem, saturation function is utilized in the proposed TSMC. Moreover, an adaptive type-2 fuzzy neural network (T2RFSFNN) is derived to achieve the model-free design, by applying the recurrent feature selection algorithm in the type-2 fuzzy neural network. To enhance the capacity to represent the uncertainties, adaptive learning mechanisms for updating the parameters of T2RFSFNN are derived by the Lyapunov theorem. Furthermore, a robust compensator with an adaptive uncertainty estimation law is investigated to relax the requirement for lumped uncertainties. Finally, the control performance using the developed T2RFSFNN is evaluated by some comparative experimental results.

МОДЕЛИРОВАНИЕ УГРОЗ БЕЗОПАСНОСТИ ИНФОРМАЦИИ В ТЕРРИТОРИАЛЬНО-РАСПРЕДЕЛЕННЫХ ИНФОРМАЦИОННЫХ СИСТЕМАХ

Introduction: in the design of information security systems importance is the modeling of security threats, which implies the definition of a list of software threats to the information system, on the basis of which decisions are made to neutralize actual threats. Today the number of urgent threats to information security is increasing, due to the complexity of the infrastructure, information processing technologies, and unprotected communication channels. In this regard, the goal is to simulate security threats in geographically distributed information systems. The analysis of foreign and Russian methodologies for threat modeling showed problems associated with a large amount of data for modeling, as well as expert methods. It has been established that to solve the problem posed use machine learning methods, the theory of adaptive fuzzy neural production systems with fuzzy inference algorithms and the use of Data Science technologies when processing large amounts of data. The paper uses such data protection methods as data protection tools, based on fuzzy neural security systems, it is proposed to determine the actual threats to information security. The proposed methodology is automated and hypothetically eliminates expert errors, increases the number of frequently used topical threats to information security, reduces financial costs for the purchase of information security tools, differs in that the process is automated, has low computational complexity, there is no need attracting highly qualified specialists, allows you to determine the list of actual threats in systems of various types and classes, can be adapted to work with databases. The practical significance lies in the automation of the process — the development of a computer program that implements the proposed methodology. Discussion: further research is advisable to continue determining the best parameters of adaptive fuzzy neural production systems and fuzzy inference algorithms.

Evaluation of the effect of music education on improving students’ mental health based on intelligent fuzzy system

Traditional control methods and modern accurate mathematical model control methods do not perform well in the evaluation of students’ mental health. In order to improve the evaluation effect of students’ mental health, this paper takes the intelligent fuzzy system as the control center and proposes an evaluation system to evaluate the effect of music education in promoting students’ mental health based on fuzzy neural network. Moreover, according to the working characteristics of the music education system, this paper interprets the design requirements of its control system in detail, and has an in-depth understanding of the fuzzy principle, neural network principle and fuzzy god network principle. Secondly, this paper completes the design of the actual orthosis control algorithm applied to the fuzzy neural network control system and the optimization of the fuzzy neural network algorithm. Finally, this paper realizes the intelligent processing of the non-linear pressure signal output by the corresponding strain, and uses music education to evaluate the students’ mental health and manage the rehabilitation effect. From the experimental research results, it can be seen that the system constructed in this paper has a certain effect.

Simpsons 1/3 base fuzzy neural network with passive controller of wheeled mobile robot system

The paper investigates Passivity of Simpson 1/3 rule fuzzy neural network controller by teleoperation-based wheeled mobile robot navigation and wheel slippage. To render the obstacle avoidance for wheeled mobile robot and ensuring incident-free navigation particularly unlimited workspace and surface slippage of coordination for master robot position and slave robot velocity, a passivity controller mode is employed. Effective control strategies are formulated to achieve system stability in which soft-computing methodology is accessed to bypass robot wheel slippage and skidding. Soft-computing based fuzzy neural network system is framed through Simpson 1/3 rule for reducing master/slave robot position error on behalf of unavoidable and acceptable forces, which is confirmed by teleoperation system. We concluded mathematically, the system stability which has been shown via its passivity of the fuzzy neural network controller. The forces which can be handled by the human operator are approximately equal to the forces applied by the environmental force and sensor predictor of slave robot whichever comes under the unlimited workspace. Simulation results are verified with the effect of the proposed controller. Index Terms: Autonomous wheeled mobile robot, Passivity, adaptive network fuzzy inference system, Simpson’s 1/3 rule, teleoperation system.

Kontrollü EGR Soğutma Sistem Tasarımının NOx ve BSFC Üzerine Etkisinin Uyarlamalı Sinirsel Bulanık Çıkarım Sistemi (ANFIS) İle Modellenmesi ve Optimizasyonu

Bu çalışma kapsamında dizel motorların egzoz gaz resirkülasyonu (EGR) sistemlerinde kullanılmak üzere egzoz gazlarını soğutan yenilikçi bir sistem tasarımı yapılmıştır. Tasarlanan sistemde 12 V elektrikli pompa ve fan kullanılmıştır. Bu elemanlar için ayrı ayrı PID (Oransal Integral Türev) kontrolcüler tasarlanmış ve kontrolcü parametreleri için optimizasyon yöntemi kullanılmıştır. Emme manifolduna giren gazların miktarı ve sıcaklığı için farklı çalışma koşullarında motor NOx ve fren özgül yakıt tüketimi (brake specific fuel consumption (BSFC)) üzerindeki etkileri analiz edilmiştir. Veriye dayalı yöntemler kullanılarak matematiksel model geliştirilmiştir. Elde edilen matematiksel model sayesinde tasarlanan kontrol sistemi için farklı motor çalışma koşullarında referans EGR akış miktarı ve sıcaklık değeri tanımlanmıştır. Uyarlamalı Sinirsel Bulanık Çıkarım Sistemi (Adaptive Neural Fuzzy Inference System (ANFIS)) ile tanımlanan modellerin gerçek verilerle uyumluluğu istatistiksel olarak analiz edilmiştir.

Measuring and Comparing Forces Acting on Moldboard Plow and Para-Plow with Wing to Replace Moldboard Plow with Para-Plow for Tillage and Modeling It Using Adaptive Neuro-Fuzzy Interface System (ANFIS)

The objective of this study was to measure the draft, vertical, and lateral forces acting on the moldboard plow, para-plow without a wing, para-plow with forward-bent wing, and para-plow with a backward-bent wing at three working depths and three forward speeds in clay loam soil to investigate the use of a suitable para-plow instead of the moldboard plow. Also, modeling the draft, vertical, and lateral forces acting on the implements using Adaptive Neuro-Fuzzy Interface System (ANFIS) was another objective of this research. To measure the draft, vertical, and lateral forces, a three-point hitch dynamometer was used. The results showed that, with the increment of the forward speed and working depth, the draft force required by the used implements increased. This increase was also true for vertical and lateral forces acting on the implements. Modeling of the draft, vertical, and lateral forces acting on the implements was performed using the effective parameters of the implements, working depth, and forward traveling speed using the (ANFIS) fuzzy neural system model. The root mean square error (RMSE) for the draft, vertical, and lateral forces for the above models were obtained equal to 0.121, 0.014, and 0.016, respectively.

Robot technological system of analysis of cybersecurity information systems and communication networks

The article presents the authors’ views on the organization of the modern system of cybersecurity. We considered the tasks, assigned to the monitoring subsystem and reconnaissance of cyberspace as a source of preventive hardware and software effects on the opposing side. The shortcomings of the existing structures of such subsystems are analyzed and the requirements to the intelligent automated system of cybersecurity are specified. The article offers hardware and software solutions, based on the development of a mobile robotic system for audit the stability of the network infrastructure and applications to existing and future cyber threats. A variant of the functional scheme of a robotic system with a universal housing is presented and with a set of mechanical drives, depending on the characteristics of the problems to be solved. For providing high functional flexibility and speed of the robotic system it is proposed to use fuzzy neural networks. The union of two independent theories - neural networks and fuzzy logic - allowed creating neural-fuzzy systems, having a high level of intelligence and independence in decision-making. The advantage of hybrid neural-fuzzy technology of modeling and information processing is the ability to view the generated rules and giving them a meaningful linguistic interpretation. A reasonable conclusion is made, that is the main meaning of that is the main meaning of the shift of creation concepts and using of artificial intelligence - transition from individual systems to distributed information processing and the development of multi-agent intelligent systems.

Forecast of export demand based on artificial neural network and fuzzy system theory

This paper analyses the significance and methods of foreign trade export forecasting. The index system of foreign trade export forecasting is determined based on the analysis of foreign trade export forecasting research results. The concepts and principles of artificial neural network and fuzzy system theory are expounded, and their respective advantages and disadvantages as well as their complementarities are analyzed. This paper introduces the types and training algorithms of evolutionary morphological neural network, combines the neural network with the fuzzy system theory, and establishes the prediction model. Finally, the evolutionary morphological neural network model is applied to the prediction of foreign trade export in view of the characteristics of export and considering the influence of various factors, the whole process of establishing evolutionary morphological neural network forecasting model is introduced in detail, and the change range of export is predicted, and the ideal forecasting results are obtained.

A pretreatment method of wastewater based on artificial intelligence and fuzzy neural network system

A pretreatment method of industrial saline wastewater based on Artificial Intelligence based fuzzy neural network analysis was proposed to improve the pretreatment accuracy of industrial saline wastewater. This method uses a four-layer AI fuzzy neural network model and proposes a graded fuzzy neural network model for pretreatment method of industrial saline wastewater, it includes input layer, fuzzification layer, fuzzy logical layer and output layer, and designs the framework and calculation mode of the fuzzy function block and the neural network module. Finally, the dynamic simulation experiments of dissolved oxygen control in the fifth zone and nitrate nitrogen control in the second zone are carried out based on the simulation benchmark model (BSM1) platform. The experimental results show that this approach can effectively raise the adaptive control accuracy of the system compared with PID, feed forward neural network and conventional recurrent neural network.

Collaborative Fuzzy Clustering Mechanism & Data Driven Fuzzy Neural System

In this Research, Fuzzy Logic guideline move system for self-developing Fuzzy neural deduction systems is anticipated. Highlights of proposed strategy, named (CFCM-DDNFS)Collaborative Fuzzy Clustering Mechanism & Data Driven Fuzzy Neural System mechanismwere; (1) Fuzzy guidelines are produced simply by Fuzzy c-means (FCM) and afterward adjusted by the (PCFC)preprocessed synergistic Fuzzy clustering method, and (2) Boundary & Structured learning are accomplished at the same time without choosing the underlying boundaries. The CFCM-DDNFS could implemented to manage enormous information issues by the goodness of the PCFC method, which is fit for managing colossal data-sets while saving the protection and security of data-sets. At first, the whole data-set is composed into two individual data-sets for the PCFC system, where each of the data-set is bunched independently. The information on model factors (bunch focuses) and the lattice of only one divide of data-set across synergistic strategy are sent. CFCM-DDNFS can accomplish consistency within the sight of aggregate information on the PCFC and lift the framework demonstrating procedure by boundary learning capacity of oneself developing neural Fuzzy induction systems (SONFIN). Proposed strategy beats existing strategies for time arrangement forecast issues. Keyword : Fuzzy system,Neural network, Big Data, On-line learning framework,Privacy & security, Time arrangement expectation, Collaborative strategy.

Adaptive type-2 neural fuzzy sliding mode control of a class of nonlinear systems

The objective of this study is to design an optimal control scheme for the control of a class of nonlinear flexible multi-body systems with extremely coupled dynamics and infinite dimensions. The assumed mode method (AMM) acquires a finite-dimensional model, but there are uncertainties in the truncated model that make the system a difficult control problem. The proposed control scheme is a hybrid of the sliding mode control (SMC) and the type-2 neural fuzzy system (NFS). A new modified conjugate gradient (CG) algorithm is used to optimize the NFS parameters. The control law of the proposed control scheme requires the estimation of the uncertain system functions, which is provided by the NFS. Moreover, NFS plays an essential role in avoiding the chattering phenomenon commonly observed in conventional SMC. The control scheme stability is assured by the Lyapunov stability theorem. Various other intelligent control schemes have also been tested to make a comparison with the proposed control scheme. The simulation results evidently indicate that the proposed control scheme has enhanced tracking efficiency while managing the inherent deflections of the system and is therefore concluded as a reliable control technique for the nonlinear, flexible multi-body class of systems.

Finite-time and fixed-time synchronization control of discontinuous fuzzy Cohen-Grossberg neural networks with uncertain external perturbations and mixed time delays

This paper aims to investigate the synchronization control of a class of discontinuous fuzzy Cohen-Grossberg neural networks (DFCGNNs) with uncertain external perturbations and mixed time delays (discrete and distributed time-delays). By using functional differential inclusions theory, inequality technique and the non-smooth analysis of Lyapunov-Krasovskii functional, a simple switching adaptive control is designed and some new criteria are obtained to achieve the finite-time synchronization of the proposed drive-response systems. Besides, the upper bound of the settling time is estimated. Based on the finite-time synchronization, we further present a novel discontinuous state-feedback controller to the response neural system in order to derive the fixed-time synchronization criteria and estimate the settling time. Compared with the previous results which can only fixed-timely synchronize or finite-timely synchronize some simple delayed fuzzy neural network systems with continuous activation functions and without uncertain external perturbations, the DFCGNNs with uncertain external perturbations and mixed time delays addressed and the theoretical results of this paper are more generalized and inclusive. Finally, several simulation examples and remarks are provided to verify the correctness and advantages of the main results.

IT2-GSETSK: An evolving interval Type-II TSK fuzzy neural system for online modeling of noisy data

As a core part of a fuzzy neural system, the rule base antecedents and consequents may carry uncertainties because they are trained using noisy data. So, handling the uncertain rule base is an important need in some specific problems such as noisy non-dynamic problems which leads a better data modeling. As a solution, Interval Type-II (IT2) version of GSETSK (Generic Self-Evolving Takagi-Sugeno-Kang), namely IT2-GSETSK, is presented in this paper. This solution uses IT2 membership functions for handling uncertainties, plus having Type-I (GSETSK) capabilities. In this way IT2-GSETSK is a fully-online model able to handle data streams and cope with time-variant data. It also provides up-to-date, relevant and compact rule base that is easily interpretable by human. The IT2-GSETSK is tested over several applications including medical, environmental and financial predictions, which show satisfactory performance of IT2-GSETSK. Moreover, it is observed that while GSETSK performs well enough for dynamic problems with less noise, noisy non-dynamic problems benefit significantly from IT2-GSETSK.

Event-triggered synchronization control for T–S fuzzy neural networked systems with time delay

This paper investigates the event-triggered drive-response synchronization control for T–S fuzzy neural networked systems with time delay. In order to save network resources and improve transmission efficiency, an error output-dependent event-triggering scheme is proposed, which can determine whether the measurement errors should be transmitted to the controller or not. For the purpose of studying the synchronization problem under the given event-triggering scheme, a novel T–S model of synchronous control based on the drive-response mode is proposed. By using the Lyapunov–krasovskii functional method, sufficient conditions are obtained to guarantee the asymptotic stability of the synchronization error systems. Moreover, by solving some linear matrix inequalities (LMIs), a set of controllers for the response systems can be achieved explicitly. Finally, two numerical examples are provided to illustrate effectiveness of the proposed method.

Wavelet Energy Fuzzy Neural Network-Based Fault Protection System for Microgrid

To perform the fault protection for the microgrid in grid-connected mode, the wavelet energy fuzzy neural network-based technique (WEFNNBT) is proposed in this paper. Through the accurate activation of protective relay, the microgrid can be effectively isolated from the utility power system to prevent serious voltage fluctuation when the power quality of power system is disturbed. The proposed WEFNNBT can be divided into three stages—feature extraction (FE), feature condensation (FC), and disturbance identification (DI). In the FE stage, the feature of power signal at the point of common coupling (PCC) between microgrid and utility power system would be extracted with discrete wavelet transform (DWT). Then, the wavelet energy and variation of singular power signal can be obtained according to Parseval Theorem. To determine the dominant wavelet energy and enhance the robustness to the noise, the feature information is integrated in the FC stage. The feature information then would be processed in the DI stage to perform the fault identification and activate the protective relay if necessary. From the experimental results, it is realized that the proposed WEFNNBT can effectively perform the fault protection of microgrid.

Adaptive tracking control of an unmanned aerial system based on a dynamic neural-fuzzy disturbance estimator

The main goal of this study is developing an adaptive controller which can solve the trajectory tracking for a class of quadcopter unmanned aerial system (UAS), namely a quadrotor. The control design introduces a new paradigm for adaptive controllers based on the implementation of a set of differential neural networks (DNNs) in the consequence section of a Takagi–Sugeno (T–S) fuzzy inference system. This dynamic fuzzy inference structure was used to approximate the UAS description. The particular form of interaction between neural networks and fuzzy inference systems proposed in the present work received the name of dynamic neural fuzzy system (DNFS). An adaptive controller based on this DNFS form was the main solution attained in this study. This DNFS controller was focused on the estimation and compensation of the uncertain section of the Quadrotor dynamics and then, forced the UAS to perform a hover flight while the tracking of desired angular positions succeeded, which results in tracking a desired trajectory in the X-Y plane. The control design methodology supported on the Lyapunov stability theory guaranteed ultimate boundedness of the estimation and tracking errors simultaneously. Several experimental tests in an outdoor environment by using a real Quadrotor platform was performed by using an RTK-GPS (Real Time Kinematic) system to determine the position of the vehicle in the X-Y plane. The experimental results confirmed the superior performance of the proposed algorithm based on the combination of DNNs and T–S techniques with respect to classical robust controllers.

Research on Noise Diagnosis for Sweeping Vehicle Based on ANFIS

Aiming at the lack of method on noise diagnosis for sweeping vehicle, this paper proposes a vehicle noise diagnosis scheme based on adaptive fuzzy neural inference system (ANFIS). The paper explains the principle of ANFIS in the field of noise diagnosis, and performs a noise test on a sweeping vehicle. The spectrum analysis is utilized to obtain the data for ANFIS network training, inference and verification. Combined with ANFIS theory, the noise source distribution of the sweeping vehicle is inferred and verified. The results show that the noise source diagnosis method based on ANFIS is feasible and effective for the sweeping vehicle. Furthermore, the method can provide a reference for the similar situation where only parts of the noise sources are known.