Neuro-fuzzy Research Articles

Self-organizing interval type-2 function-link fuzzy neural network control for uncertain manipulators under saturation: A predefined-time sliding-mode approach

Robotic manipulators have extensive applications in academic and industrial fields such as grabbing, welding, and machining. Nowadays, better tracking performance and faster transient response have been the fundamental demands in the field of manipulator control. However, robotic manipulators suffer from uncertainties due to various factors in real-world environments. Therefore, achieving satisfactory tracking performance for the manipulator in the presence of uncertainties is a pressing problem. Moreover, most existing control schemes struggle to determine the convergence time of the whole system and rely on specifically designed parameters via experience and prior knowledge, which leads to poor transferability and generalization. To address the abovementioned issues, a predefined-time controller based on a self-organizing interval type-2 function-link fuzzy neural network (IT2FLFNN) is proposed in this paper. Different from most existing fuzzy neural network (FNN) based control schemes, the proposed network starts from an empty rule base and it is constructed by a self-organizing mechanism during the online control process, which provides greater flexibility in terms of network structure and input partition. Also, a function-link network (FLN) is integrated into IT2FLFNN to enhance its convergence speed and rejection of uncertainties. Furthermore, to ensure the predefined-time convergence, a predefined-time sliding-mode control part is merged into the control framework and online learning laws for IT2FLFNN are established via the Lyapunov stability analysis. Finally, a comparative simulation and an experiment demonstrate the superiority of the proposed IT2FLFNN control scheme.

Enhanced fingerprint classification through modified PCA with SVD and invariant moments.

This research introduces a novel MOMENTS-SVD vector for fingerprint identification, combining invariant moments and SVD (Singular Value Decomposition), enhanced by a modified PCA (Principal Component Analysis). Our method extracts unique fingerprint features using SVD and invariant moments, followed by classification with Euclidean distance and neural networks. The MOMENTS-SVD vector reduces computational complexity by outperforming current models. Using the Equal Error Rate (EER) and ROC curve, a comparative study across databases (CASIA V5, FVC 2002, 2004, 2006) assesses our method against ResNet, VGG19, Neuro Fuzzy, DCT Features, and Invariant Moments, proving enhanced accuracy and robustness.

Discrete-Time Event-Triggered Type-2 fuzzy wavelet neural network control for Multi-Motor servo system

This paper investigates a discrete-time event-triggered adaptive neural network control scheme for a multi-motor servo system (MMSS) to realize a desired position tracking performance. First, a discrete-time model of the MMSS, including a dead-zone and a nonlinear friction, is established based on the Euler’s discretization. Then, a discrete-time adaptive neural network controller is designed by integrating a type-2 fuzzy wavelet neural network (T2FWNN) and the backstepping technology. The neural network is not only used to estimate uncertain nonlinearities, but also can handle the non-causal problem caused by the conventional backstepping method. Meanwhile, a fixed threshold event-triggered mechanism along with the incorporation of a dead-zone operator is superimposed into the actual controller thus saving communicational resources. Besides, stability analysis proves that all the signals in the closed MMSS are bounded, and the position tracking error converges to a small neighborhood of the origin. Finally, abundant simulation experience results demonstrate the effectiveness and robustness of the proposed scheme.

Natural gas consumption forecasting using a novel two-stage model based on improved sparrow search algorithm

The foundation of natural gas intelligent scheduling is the accurate prediction of natural gas consumption (NGC). However, because of its volatility, this brings difficulties and challenges in accurately predicting NGC. To address this problem, an improved model is developed combining improved sparrow search algorithm (ISSA), long short-term memory (LSTM), and wavelet transform (WT). First, the performance of ISSA is tested. Second, the NGC is divided into several high- and low-frequency components applying different layers of Coilfets’, Fejer-Korovkins’, Symletss’, Haars’, and Discretes’ orders. In addition, the LSTM is applied to forecast the decomposed components in view of the one- and multi-step, and its hyper-parameters are optimized by ISSA. At last, the final prediction results are reconstructed. The research results indicate that: (1) Comparing to other machine algorithms (e.g. fuzzy neural network), the convergence speed and stability of ISSA are stronger in view of standard deviation and mean; (2) The prediction performance of the developed model is better than that of other forecasting models; (3) The forecasting performance of the single-step forecasting is superior to that of the two-, three-, and four- step; (4) The computational load of the proposed prediction model is the highest compared to other models, and the prediction accuracy is still excellent on the extended time series.

Text document clustering using mayfly optimization algorithm with k-means technique

Text clustering is a subfield of machine learning (ML) and natural language processing (NLP) that consists of grouping similar sentences or documents based on their content. However, insignificant features in the documents minimize the accuracy of information retrieval which makes it challenging for the clustering approach to efficiently cluster similar documents. In this research, the mayfly optimization algorithm (MOA) with a k-means approach is proposed for text document clustering (TDC) to effectively cluster similar documents. Initially, the data is obtained from Reuters-21678, 20-Newsgroup, and BBC sports datasets, and then pre-processing is established by stemming and stop word removal to remove unwanted phrases or words. The data imbalance approach is established using an adaptive synthetic sampling algorithm (ADASYN), then term frequency-inverse document frequency (TD-IDF) and WordNet features are employed for extracting features. Finally, MOA with the K-means technique is utilized for TDC. The proposed approach achieves better accuracy of 99.75%, 99.54%, and 98.24% when compared to the existing techniques like fuzzy rough set-based robust nearest neighbor-convolutional neural network (FRS-RNN-CNN), TopicStriker, Modsup-based frequent itemset, and rider optimization-based moth search algorithm (Modsup-Rn-MSA), hierarchical dirichlet-multinomial mixture, and multi-view clustering via consistent and specific non-negative matrix (MCCS).

A novel cascaded H-bridge photovoltaic inverter with flexible arc suppression function

This paper presents a novel approach that simultaneously enables photovoltaic (PV) inversion and flexible arc suppression during single-phase grounding faults. Inverters compensate for ground currents through an arc-elimination function, while outputting a PV direct current (DC) power supply. This method effectively reduces the residual grounding current. To reduce the dependence of the arc-suppression performance on accurate compensation current-injection models, an adaptive fuzzy neural network imitating a sliding mode controller was designed. An online adaptive adjustment law for network parameters was developed, based on the Lyapunov stability theorem, to improve the robustness of the inverter to fault and connection locations. Furthermore, a new arc-suppression control exit strategy is proposed to allow a zero- sequence voltage amplitude to quickly and smoothly track a target value by controlling the nonlinear decrease in current and reducing the regulation time. Simulation results showed that the proposed method can effectively achieve fast arc suppression and reduce the fault impact current in single-phase grounding faults. Compared to other methods, the proposed method can generate a lower residual grounding current and maintain good arc-suppression performance under different transition resistances and fault locations.

Time-Aware Fuzzy Neural Network Based on Frequency-Enhanced Modulation Mechanism

Time-Aware Fuzzy Neural Network Based on Frequency-Enhanced Modulation Mechanism

Inverse compensation mechanism-based adaptive fuzzy-neural fault-tolerant control for an uncertain quadrotor UAV

Actuator faults, system uncertainties, and external disturbances can affect tracking control performance of the quadrotor UAV, and even cause the instability of the system. To address this issue, an adaptive fuzzy-neural fault-tolerant control (AFNFTC) scheme is proposed for the quadrotor UAV integrated with inverse compensation strategy and fuzzy neural networks (FNNs) technology. Firstly, we design an inverse compensation mechanism to address actuator faults and utilize the approximation capabilities of the FNNs to compensate for system uncertainties. Then, based on the Lyapunov direct method, the AFNFTC scheme is designed to ensure the stability of the quadrotor UAV. Under the proposed scheme, the closed-loop system for the quadrotor UAV is proven to be uniformly bounded, and the tracking error of the quadrotor UAV is demonstrated to converge to a small compact set ultimately by appropriately selecting design parameters. Finally, the feasibility and effectiveness of the proposed scheme is validated through simulation results.

Improved fixed-time stability analysis and applications to synchronization of discontinuous complex-valued fuzzy cellular neural networks

This article mainly centers on proposing new fixed-time (FXT) stability lemmas of discontinuous systems, in which novel optimization approaches are utilized and more relaxed conditions are required. The conventional discussions about Vt>1 and 0<Vt⩽1 are no longer required. For the purpose of verifying the new lemmas, complex-valued fuzzy cellular neural networks (CVFCNNs) with discontinuous activation functions are studied with a nontraditional non-separation method, which could reduce the conservatism of the obtained results greatly. Besides, FXT synchronization is discussed simultaneously. When contrasted with the results of other similar prominent pioneering works nowadays, the accuracy of settling times (STs) is quite enhanced. At last, numerical simulations are conducted to demonstrate the validity and superiority of our established theoretical results.

Study on Talent Cultivation Management Model of Universities Based on Fuzzy Neural Network Algorithm

The Talent Cultivation Management Model for Universities represents a strategic framework designed to optimize the development and oversight of academic programs. This model focuses on identifying, nurturing, and assessing talents within the university ecosystem. It incorporates innovative methodologies to align educational offerings with individual needs and career goals. This study presents an innovative approach to developing a talent cultivation management model for universities, leveraging the integration of the Fuzzy Neural Network Algorithm with the proposed Optimized Spider Monkey Fuzzy Neural Network (OSMF-NN). Recognizing the critical importance of talent development in higher education, this research seeks to enhance the efficacy and adaptability of existing management models. The OSMF-NN algorithm, inspired by the optimization capabilities of spider monkey behavior, enhances the traditional fuzzy neural network algorithm, enabling more precise and efficient talent management. By harnessing the synergies between fuzzy logic and neural networks, the proposed model offers a robust framework for identifying, nurturing, and evaluating talents within the university ecosystem. Through comprehensive experimentation and validation, this study demonstrates the effectiveness of the OSMF-NN algorithm in optimizing talent cultivation strategies, promoting personalized learning experiences, and fostering student success in higher education institutions.

Using a fuzzy credibility neural network to select nanomaterials for nanosensors

Nanomaterials are the most important component of nanosensors, and the selection of the most desirable nanomaterial for nanosensors is a challenge for companies. The classical decision making procedure is very difficult and uncertain to select the desirable nanomaterial. Therefore, we develop a decision making model based on a fuzzy credibility neural network. In this article, we introduce a novel fuzzy credibility neural network using Dombi t-norms and co-norms and also using the score function of fuzzy credibility numbers. Further, the fuzzy credibility neural network applies to the decision making model for the selection of the best nanomaterial for nanosensors. In this approach or decision making model, we first collect data from three experts in the form of fuzzy credibility numbers and then use a neural network to aggregate the data with the help of Dombi t-norm and t-conorm. We consider the expert decision making criteria, which correspond to the input signals of the fuzzy credibility neural network, and calculate the weight of the input signal using distance measure techniques. Next, we compute the hidden layer information and the output layer information by using the fuzzy credibility neural network with the Dombi aggregation operator. The proposed approach is compared with other existing models of decision making and the results of the comparison show that the proposed technique is applicable and reliable for the decision support model.

Design and application of fuzzy neural network systems optimized with hybrid algorithms

Design and application of fuzzy neural network systems optimized with hybrid algorithms

The improvement of field-oriented control performance for double stator IPMSM based on NFC powered by dual photovoltaic solar panels

This paper presents a proposed method Neuro Fuzzy Control (NFC)-Field Oriente Control (FOC)-Space Vector Modulation (SVM) using IP controller for a five-level NPC inverter-Double Stator Interior Permanent Magnet Synchronous Motor (DSIPMSM) supplied by two photovoltaic solar panels Maximum Power Point Tracking (MPPT) Fuzzy Sliding Mode Control (FSMC). The torque generated and the magnetic flux of the DSIPMSM may be independently controlled thanks to FOC, while IP controllers are used to minimize dq axis current errors to zero. A boost converter is made up of an inductor, a capacitor, and a separate input-output connection. A MPPT approach based on the FSMC algorithm is used to extract the maximum amount of energy from the solar panel while controlling the inverter duty cycle. The NFC tool is an extremely potent instrument that bridges the gaps created by the fuzzy logic controller. The proposed method’s worth was demonstrated by the simulation results in terms of durability, decoupling between the d-q axis, and dynamic performance.

An adaptive approximation‐free control for leader‐follower tracking of multi‐agent systems with unknown nonlinearity and disturbance

AbstractDue to variable and complex work environments, nonlinearities, uncertainty and disturbances are inevitable in multi‐agent systems. Approximation‐free control can address these issues without involving approximators, such as fuzzy logic systems and neural networks. However, some issues like the singularity problem caused by the signals exceeding the preset boundary in changing work conditions still remain. This paper proposes an adaptive and reliable approximation‐free control, which comprises a novel singularity compensator and a modified transforming function. The proposed control scheme performs better in terms of convergence rate and overshoot, avoids issues relating to singularity, and has added flexibility in terms of parameter choice. The proposed control law adapts to changes in operating conditions and nonlinearities—the efficacy of which is demonstrated using simulations.

Enhancing effort estimation in global software development using a unique combination of Neuro Fuzzy Logic and Deep Learning Neural Networks (NFDLNN)

ABSTRACT Effective project planning and management in the global software development landscape relies on addressing major issues like cost estimation and effort allocation. Timely estimation of software development is a critical focus in software engineering research. With the industry increasingly relying on diverse teams worldwide, accurate estimation becomes vital. Software size serves as a common measure for costs and schedules, but advanced estimation methods consider various variables, such as project purpose, personnel expertise, time and efficiency constraints, and technology requirements. Estimating software costs involve significant financial and strategic commitments, making it crucial to address complexity and versatility related to cost drivers. To achieve enhanced accuracy and convergence, we employ the cuckoo algorithm in our proposed NFDLNN (Neuro Fuzzy Logic and Deep Learning Neural Networks) model. Through extensive validation with industrial project data, using Function Point Analysis as the algorithmic models, our NFA model demonstrates high accuracy in software cost approximation, outperforming existing methods insights of MRE of 3.33, BRE of 0.13, and PI of 74.48. Our research contributes to improved project planning and decision-making processes in global software development endeavours.

Klasifikasi Kebutuhan Dokter untuk Kesejahteraan Masyarakat Menggunakan ANFIS

In social life, the need for medical workers is different in each region, because the population varies. If doctors cannot treat a large enough number of patients in an area, it can have various impacts on society, such as contracting dangerous diseases if not treated as soon as possible. The effect will be a decline in people's standard of living. By classifying the need for the number of health workers (doctors) relative to the population, the level of welfare in an area can be obtained. To assist in optimizing health workers (doctors) they can use fuzzy logic and ANFIS (Adaptive Neuro Fuzzy Inference System). By using ANFIS, it is hoped that we can find the optimal value for the classification of health workers that will be needed in each region. In the ANFIS test, the RMSE error value was 0.2698 and the first accuracy value was 73%. Then by adding a membership function, an RMSE error value of 0.17698 was obtained, this second accuracy value increased by 10% to 83%. By using the ANFIS method to classify health workers according to different population sizes in each region, you can measure the level of community welfare well.

Fixed-time formation tracking for unmanned surface vehicles: A multi-layer neural networks approach

This paper investigates the distributed fixed-time formation problem of unmanned surface vehicles (USVs) in the presence of external disturbances, model uncertainties, input saturation and quantization constraints. To deal with the problem, a fixed-time sliding-mode control algorithm is proposed, where multi-layer neural networks (MNNs) are designed to approximate the unknown dynamics and composite disturbances of the system. The proposed MNNs combine the advantages of fuzzy neural networks (FNNs) and radial basis function neural networks (RBFNNs), exhibiting robust dynamic characteristics. Furthermore, the non-singular fast terminal sliding mode (NFTSM) is integrated into the fixed-time control framework to improve the robustness and speed of convergence for uncertain USV systems. Comparative simulations conducted with USVs demonstrate the superiority and effectiveness of the proposed algorithm.

Development and Application of Modern Intelligent Control Theory

Modern intelligent control theory has experienced multi-stage development from classical control to fuzzy control, neural networks, and model predictive control, which not only improves the flexibility and adaptability of control systems but also promotes interdisciplinary research. With the advancement of science and technology and social development, intelligent control theory will be further combined with machine learning, big data, and other fields in the future to promote industrial automation and technological innovation. This article reviews the development history of modern intelligent control theory, its main theories, and its applications in robot manufacturing, power electronics, and other fields. It also discusses the current challenges and future development trends in the field of intelligent control.

Improving the Detectionand Warning Fire System on the Smart Campus Area using ANFIS

The adoption of IoT technology in universities has significantly improved campus security, emergency response, and control systems, particularly through IoT-based surveillance and fire detection systems. Traditional fire detection systems suffer from high false alarm rates, leading to unnecessary resource allocation. This paperproposes an intelligent fire detection system leveraging IoT and adaptive fuzzy systems to enhance accuracy and reduce false alarms.The system uses sensors to collect real-time data, which is analyzed by an Adaptive Neuro-Fuzzy Inference System (ANFIS) to determine alarm levels based on input severity. ANFIS combines fuzzy logic and neural networks, enabling learning and adaptation. Data analyzed by ANFIS is sent to ThingSpeak channels for real-time monitoring. When a fire is confirmed, alerts with fire location, timestamp, and severity are sent via SMS through GSM to the fire management system. The system utilizes cost-effective, small-sized sensors, ensuring repeatable solutions.After training, the ANFIS system achieved 99.59% accuracy, significantly reducing false alarms. It demonstrated faster detection times by reducing the fire detection rate by 28% and maintaining high detection accuracy with fewer sensors, training inputs, and epochs. Utilizing IoT and ANFIS technologies, the system integrates efficiency, speed, and reliability, protecting lives and property and highlighting the importance of safety andtechnology in serving humanity.

Fixed-time control and estimation of discontinuous fuzzy leakage-delayed networks: Indefinite and economical conditions of Filippov systems

This paper considers the fixed-time stability of Filippov systems and discontinuous networks described by the Filippov systems. Firstly, the more generalized and economical inequality condition on Lyapunov function is proposed, which can not only improve the previous negative definite conditions, but also the indefinite conditions. Based on the definition of the Lyapunov stability and finite-time convergence, new fixed-time stability lemmas are proposed and the settling-time is also estimated. In order to apply the new fixed-time stability lemmas to the neural networks, fixed-time stabilization of a class of fuzzy leakage-delayed neural networks with discontinuous activations is considered which is rarely studied since the delay is leakage and activations are discontinuous and they are not easy to deal with simultaneously. By means of the Filippov regularization and differential inclusions theory, new criteria on the stabilization of the considered networks are derived via the new modified non-chattering controller, which can improve the previous ones with parameter and symbolic function. Finally, example and numerical simulations further examine the correctness of the main results.