Neuro-fuzzy System Research Articles

A review on soft computing techniques used in induction motor drive application

<div align="center"><table width="590" border="1" cellspacing="0" cellpadding="0"><tbody><tr><td valign="top" width="387"><p>In this paper, hybrid models based on fuzzy systems and neural networks are reviewed. A fuzzy inference system is explicitly represented by expertise for induction motor drives, incorporating the learning capability of artificial neural networks. Researchers have been attracted to neuro-fuzzy techniques for training and inference in induction motor drives due to their efficiency. According to the classification of research articles from 2000 to 2020, this article presents a review of different artificial neural network techniques, fuzzy and neuro-fuzzy systems. The main objective is to provide a concise overview of current neuro-fuzzy research and to enable readers to identify appropriate methods according to their research interests.</p><p> </p></td></tr></tbody></table></div>

Real-Time Load Forecasting and Adaptive Control in Smart Grids Using a Hybrid Neuro-Fuzzy Approach

The transition to smart grids is revolutionizing the management and distribution of electrical energy. Nowadays, power systems must precisely estimate real-time loads and use adaptive regulation to operate in the era of sustainable energy. To address these issues, this paper presents a new approach—a hybrid neuro-fuzzy system—that combines neural networks with fuzzy logic. We use neural networks’ adaptability to describe complex load patterns and fuzzy logic’s interpretability to fine-tune control techniques in our approach. Our improved load forecasting system can now respond to changes in real-time due to the combination of these two powerful methodologies. Developing, training, and implementing the forecasting and control system are detailed in this article, which also explores the theoretical underpinnings of our hybrid neuro-fuzzy approach. We demonstrate how the technology improves grid stability and the accuracy of load forecasts by using adaptive control methods. Furthermore, comprehensive simulations confirm the proposed technology, showcasing its smooth integration with smart grid infrastructure. Better energy management is just the beginning of what our method can accomplish; it also paves the way for a more sustainable energy future that is easier on the planet and its inhabitants. In conclusion, this study’s innovative approach to adaptive control and real-time load forecasting advances smart grid technology, which, in turn, improves sustainability and energy efficiency.

Multi-functional Inconel 625 micro-machining and process optimization using ANFIS

The primary objective of this study is to develop an adaptive neuro-fuzzy inference-based system (ANFIS) that assesses the ideal parameters and impact of the micro-electro-discharge machining (Micro-EDM) technique on surface roughness (SR). In this work multi-functional Inconel alloy specimens were machined using micro-elector-discharge machining (Micro-EDM) with zinc-coated copper electrodes in a graphite nano abrasive submersion medium. An ANFIS and Taguchi design of experiments (DOE) was employed to evaluate the Micro-EDM method. The study utilized Taguchi's L9 mixed orthogonal arrays (OA) to consider three process variables, namely open voltage (V), capacitance (C), and powder concentration (CP). ANOVA analysis was conducted to identify important process variables and evaluate their influence on surface properties. The experimental results showed that all approaches were effective in representing the process. Based on the results of the response surface methodology (RSM) and ANFIS predictions, the correlation coefficients (R2) of the experimental roughness were 0.995649 and the root mean square error (RMSE) value was 1.01226e-06, respectively. The results obtained indicate a strong alignment between the predicted and experimental values, effectively fulfilling the proposed objective. These findings hold significant potential for practical applications in the Micro-EDM industry, particularly in the machining of Inconel alloy.

EFNN-Nul0- a trustworthy knowledge extraction about stress identification through evolving fuzzy neural networks

This paper presents a novel hybrid architecture, denoted as EFNN-Nul0 (evolving neuro-fuzzy system based on null-unineurons), meticulously crafted for stress identification within the realm of pattern classification. The model seamlessly integrates neural networks, fuzzy systems, and n-uninorms to cultivate knowledge within an adaptive, real-time learning context. Neuro-fuzzy rules, encapsulating interdependencies among input features through IF-THEN type relationships, are formulated utilizing null-unineurons derived from n-uninorms. This construction enables the articulation of both AND- and OR-connections, thereby augmenting the interpretability of the generated rules. The evolution of neurons is facilitated by an extended adaptation of the autonomous data partition method (ADPA). To dynamically interpret the evolution of rules, the paper introduces (i) a mechanism that tracks changes in rules over data stream samples, providing insights into the process dynamics as a structural active learning component, and (ii) a concept for incrementally updating feature weights. These weights encapsulate the varying impact levels of features on stress identification, enabling the reduction of rule length by effectively masking out less significant features. The outcomes of the rules are represented by certainty vectors, recursively updated through an indicator-based recursive weighted least squares (I-RWLS) approach. Neuron activation levels play a pivotal role in determining the weights, ensuring stable local learning. The effectiveness of the proposed model is substantiated through a comprehensive comparison with existing hybrid and evolving approaches in the literature, focusing on binary and multi-class pattern classification. Aspects that identify special characteristics of neuro-fuzzy models related to interpretability will also be demonstrated in this work. The results demonstrate the model's superior performance, characterized by consistently higher accuracy trend lines over time. Furthermore, the model's interpretability is underscored by the coherent neuro-fuzzy rules, contributing to its remarkable accuracy (the EFNN-Nul0 achieved approximately 99% accuracy in stress identification), establishing its efficacy in addressing stress identification within classification problems.

Speed Control of Induction Motors using Neuro Fuzzy Expert Systems

Induction motors and their speed control is critical aspect in several real life applications. Often, the proportional integral derivative (PID) controlling mechanism is found in several applications in industries where speed of motors or induction motors is to be controlled. The PID controller is specifically useful since it tries to minimize the steady state error as well as increase the response or speed of the system, thereby incorporating the benefits of proportional derivative and proportional integral control. However, the real time operation of PID controllers is challenging due to its tuning. The controlling mechanism is critically important for the application of the PID. Previously, manual tuning was used which needed experts and was also prone to errors. With the advent of sophisticated optimization tools, automatic tuning has gained momentum. In this work, a combination of neural networks and fuzzy logic often called neuro fuzzy expert systems has been proposed for the speed control of induction motors. It has been shown that proposed system is capable to attain better results compared to conventional techniques. The system has been designed on Matlab/Simulink. Key Words: Induction Motors, Speed Control, PID Controllers, Fuzzy Logic, Neuro Fuzzy Expert Systems.

Multi-objective optimization and performance assessment of response surface methodology (RSM), artificial neural network (ANN) and adaptive neuro-fuzzy interfence system (ANFIS) for estimation of fouling in phosphoric acid/steam heat exchanger

Fouling is a common occurrence in industrial heat exchangers, leading to a decrease of the thermal efficiency. This research addresses the challenge of applying various machine learning algorithms including Response Surface Methodology (RSM), Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) to effectively model the fouling resistance in a heat exchanger in phosphoric acid concentration loop. Subsequently, a multi-objective optimization approach was employed to minimize the fouling resistance. Confirmatory experiments are then conducted in the phosphoric acid concentration plant using optimized variables.The findings of this study indicate that the three models accurately align with one year of operational data, achieving a remarkably high coefficient of correlation (R). Among the models utilized, RSM demonstrates the highest level of prediction accuracy, with an R of 0.9998, accompanied by the lowest mean square error (MSE) of 5.9388 10−13 and root mean squared error (RMSE) of 7.7064 10−7. The RSM optimization process identifies the optimal conditions for variables such as time, acid inlet and outlet temperature, steam temperature, acid density, and volume flow rate, which are determined to be 114.805 h; 72.214 °C; 80.407 °C; 116.784 °C; 1642.47 kg/m3; and 2308.1 m3/h, respectively. The predicted fouling resistance demonstrates a strong correlation with the actual data, with a negligible percentage difference of 2.19 %, further validating the accuracy and reliability of the RSM model.

A robust genetic algorithm-based optimal feature predictor model for brain tumour classification from MRI data

ABSTRACT Brain tumour can be cured if it is initially screened and given timely treatment to the patients. This proposed idea suggests a transform- and windowing-based optimization strategy for exposing and segmenting the tumour region in brain pictures. The processes of image processing that are included in the proposed idea include preprocessing, transformation, feature extraction, feature optimization, classification, and segmentation. In order to convert the pixels connected to the spatial domain into a multi-resolution domain, the Gabor transform is first applied to the brain test image. The Gabor converted brain image is then used to extract the parameters of the multi-level features. After that, the Genetic Algorithm (GA) is used to optimize the extracted features, and Neuro Fuzzy System (NFS) is used to classify the optimistic prominent section. Finally, the tumour region in brain images is found and segmented using the normalized segmentation algorithm. The effective detection and classification of brain tumours by the characteristics of sensitivity, specificity, and accuracy are described by the suggested GA-based NFS classification approach. The trial findings are displayed with an average of 99.37% sensitivity, 98.9% specificity, 99.21% accuracy, 97.8% PPV, 91.8% NPV, 96.8% FPR, and 90.4% FNR.

Performance analysis of a gas turbine engine via intercooling and regeneration- Part 2

Abstract The current study aims to amplify the predictive ability of the numerical model developed for a gas turbine engine-based power plants by process of regeneration and intercooling. Artificial neural networks (ANN) and adaptive neuro-fuzzy interface systems (ANFIS) are the two techniques mainly concentrated in this study which were not properly implemented previously. The performance parameters namely, specific power (SP), thermal efficiency (η), and enthalpy based specific fuel consumption (EBSFC) of a Turboprop engine were predicted using thermodynamic parameters namely, pressure ratio (PR), nozzle pressure ratio (NPR), turbine inlet temperature (TIT), for constant regeneration (R), and intercooling (E) efficiencies. The results showed that a high regression result R 2 of 0.9831 and 0.9899 was found for the ANFIS model for η for training and testing, respectively. Also, the ANFIS model resulted in best performance of the performance characteristics when compared to ANN.

Neuro-fuzzy-SVPWM switched-inductor-capacitor-based boost inverter for grid-tied fuel cell power generators, design and implementation

Hydrogen energy shows promise as a renewable energy source for various applications like battery and electric vehicle charging stations, as well as grid connections. However, high current ripple from fuel cells (FCs) and inadequate voltages for grid use pose challenges. This study presents a novel solution using neural fuzzy network control in a high-gain DC-DC boost converter to address these issues. The suggested converter charges in parallel and discharges in series, minimizing the current ripple range in the fuel cell network. Additionally, the switch-capacitor cell efficiently increases the output voltage.In this study, a Neuro-fuzzy system with 9 rules is trained meticulously over 50 epochs using hybrid optimization and grid partition methods, achieving a low training error of 0.045 with 522,064 samples. The neural fuzzy network, employing the weighted average method for Defuzzification, produces duty cycle values from 0.02 to 0.5 in response to input signals. Additionally, an innovative Space Vector Pulse Width Modulation (SVPWM) approach within the inverter circuit enhances voltage generation precision and power quality for grid delivery, notably reducing current ripple and ensuring stable power supply. This combined with the neural fuzzy network in the converter efficiently converts hydrogen energy into AC voltage for seamless grid integration, revolutionizing boost converter efficiency and advancing hydrogen energy utilization across various energy sectors.

Prediction of the discharge capacity of piano key weirs using artificial neural networks

ABSTRACT The discharge capacity of the piano key weir (PKW) is an important flow feature which ultimately decides the design of PKWs. In the present research work, the different architecture of artificial neural networks (ANNs) was employed to predict the discharge capacity of the trapezoidal piano key weir (TPKW) by varying geometric parameters. Furthermore, adaptive neuro-fuzzy interference system (ANFIS), support vector machines (SVMs) and non-linear regression (RM) techniques were also applied to compare the performance of best ANN models. The performance of each model was evaluated using statistical indices including scatter-index (SI); coefficient of determination (R2) and mean square error (MSE). The prediction capability of all the models was found to be satisfactory. However, results predicted by ANN-22(H-15) [R2 = 0.998, MSE= 0.0024, SI = 0.0177] were more accurate than ANFIS (R2 = 0.995, MSE = 0.00039, SI=0.0256), SVM (R2 = 0.982, MSE = 0.000864, SI =0.0395) and RM (R2 = 0.978, MSE = 0.001, SI = 0.0411). It was observed that Si/So, Wi/Wo and L/W ratios have the greatest effect on the discharge performance of TPKW. Furthermore, sensitivity analysis confirmed that h/P is the most influencing ratio which may considerably affect the discharge efficiency of the TPKW and ANN architecture having a single hidden layer and keeping neurons three times of input parameters in hidden layers generated better results.

METHOD OF HIGH-PRECISION NEURO-FUZZY SYSTEMS DEFUZZIFIERS’ AUTOMATIC CONSTRUCTION

Increasing the precision of neuro-fuzzy approximation systems is considered. Modification of defuzzifier, which allows producing multiple types of approximating functions is presented. RMSE (Root Mean Square Error), maximal error, training time and models’s size are used to evaluate the efficiency of proposed method. Post-processing functions were also considered as a tool to increase approximation precision. Common solution to this problem is a modification of model’s fuzzy sets (membership functions and their number), which significantly increases model’s size and complexity. Given approach uses search algorithms to find the configuration of defuzzification layer, which produces the best approximation precision for a given function and fuzzy sets configuration, since all possible configurations are checked. It is possible to model classic approximation methods and combine them. Possible modifications of presented search algorithms application are described. Search depth can be limited in order to reduce the size of produced model. The computing experiment, aimed at finding best approximation precision was carried out. Presented method provided lowering of RMSE without significant increasing of model’s size. Using Fourier series components in resulting expression can be used for precise periodic functions approximation or for general precision improvement. Using post-processing functions provided significant improvement in precision for multiplication approximation, which proves, that it can be used as another tool to be used to reduce errors in some specific cases.

Enhanced ship engine vibration reduction using magnetorheological dampers and Adaptive Neuro-Fuzzy Control System

ABSTRACT In this paper, an innovative approach is proposed to mitigate ship engine vibrations by integrating a semi-active control system with magnetorheological (MR) dampers and an Adaptive Neuro-Fuzzy Control System (ANFIS). The ANFIS controller adjusts MR damper damping force to effectively adapt to engine vibrations. A comprehensive mathematical model simulates the complex dynamics of the ship engine vibration system, validated with real-world data. The model incorporates engine vibration measurements and MR damper properties to train the ANFIS controller. Compared to passive control systems, the proposed semi-active system reduces torsional vibrations by 22.606% and longitudinal vibrations by 29.840%. The ANFIS controller dynamically adjusts damping force in response to engine conditions, ensuring optimal performance. This system effectively mitigates ship engine vibrations, improving overall ship performance and passenger comfort. Consistency between modeling and experimental results confirms the system's viability and efficacy.

Modeling the weld bead penetration in the presence of Cr2O3 nanoparticles in the submerged arc welding process using a modified neuro-fuzzy system

Modeling the weld bead penetration in the presence of Cr2O3 nanoparticles in the submerged arc welding process using a modified neuro-fuzzy system

Stochastic neuro-fuzzy system implemented in memristor crossbar arrays.

Neuro-symbolic artificial intelligence has garnered considerable attention amid increasing industry demands for high-performance neural networks that are interpretable and adaptable to previously unknown problem domains with minimal reconfiguration. However, implementing neuro-symbolic hardware is challenging due to the complexity in symbolic knowledge representation and calculation. We experimentally demonstrated a memristor-based neuro-fuzzy hardware based on TiN/TaOx/HfOx/TiN chips that is superior to its silicon-based counterpart in terms of throughput and energy efficiency by using array topological structure for knowledge representation and physical laws for computing. Intrinsic memristor variability is fully exploited to increase robustness in knowledge representation. A hybrid in situ training strategy is proposed for error minimizing in training. The hardware adapts easier to a previously unknown environment, achieving ~6.6 times faster convergence and ~6 times lower error than deep learning. The hardware energy efficiency is over two orders of magnitude greater than field-programmable gate arrays. This research greatly extends the capability of memristor-based neuromorphic computing systems in artificial intelligence.

Predictive modeling of copper (II) adsorption from aqueous solutions by sawdust: a comparative analysis of adaptive neuro-fuzzy interference system (ANFIS) and artificial neural network (ANN) approaches

Heavy metal ions are considered to be the most prevalent and toxic water contaminants. The objective of thois work was to investigate the effectiveness of employing the adsorption technique in a laboratory-size reactor to remove copper (II) ions from an aqueous medium. An adaptive neuro-fuzzy inference system (ANFIS) and a feed-forward artificial neural network (ANN) were used in this study. Four operational factors were chosen to examine their influence on the adsorption study: pH, contact duration, initial Cu (II) ions concentration, and adsorbent dosage. Using sawdust from wood, prediction models of copper (II) ions adsorption were optimized, created, and developed using the ANN and ANFIS models for tests. The result indicates that the determination coefficient for copper (II) metal ions in the training dataset was 0.987. Additionally, the ANFIS model’s R 2 value for both pollutants was 0.992. The findings demonstrate that the models presented a promising predictive approach that can be applied to successfully and accurately anticipate the simultaneous elimination of copper (II) and dye from the aqueous solution.

Enhancing river flow predictions: Comparative analysis of machine learning approaches in modeling stage-discharge relationship

Streamflow, a pivotal variable in water resources management, holds profound significance in shaping the decision-making processes of hydrologic projects. This paper tries to delve into the exploration of the stage-discharge relationship using three machine learning methods (MLMs) namely multi-layer neural networks (MLNN), radial basis neural networks (RBNN), and neuro-fuzzy systems (ANFIS) to predict and simulate mean daily stage-discharge data derived from two monitoring stations, Bulakbasi and Karaozü, Kizilirmak River, Turkey. Root mean square error (RMSE), Mean absolute percentage error (MAPE), coefficient of determination (R2), and the Developed Discrepancy Ratio (DDR) metrics were utilized to MLMs' performance assessment. The performance evaluation indices (RMSE, MAEP, R2, DDR) for the preeminent MLNN model applied to Bulakhbashi and Karasu stations were determined as (0.29, 1.57, 0.9998, 17.62) and (1.71, 6.56, 0.9980, 6.65), respectively. The MLNN model contributed to a notable enhancement in the RMSE performance index for the aforementioned stations, exhibiting improvements of 87% and 56%, respectively. These results affirm the MLNN's proficiency in accurately capturing the stage-discharge at both monitoring stations.

Enhancing Analytical Precision in Company Earnings Reports through Neurofuzzy System Development: A Comprehensive Investigation

The object of research is the fundamental and technical indicators of companies after the release of the earnings report. This study attempts to address the issue of understanding the impact of fundamental and technical analysis indicator dynamics on profits and loss news releases. This research provides an in-depth analysis of stock price forecasting models, focusing on the influence of earning report seasons as catalysts for stock price growth. The study explores the relationship between key financial indicators, including earnings per share (EPS), revenue, and the maximum price observed in the 52-week period of the previous year (MaxW52). A trading algorithm is developed based on the adaptive neurofuzzy inference system (ANFIS). Through a comprehensive analysis of the neural network’s training sample, it is concluded that abnormally large negative indicators have a profound impact on traders’ emotional reactions. This results leads to a hypothesis for further research, suggesting that report indicators may be processed by computational algorithms, potentially including artificial intelligence (AI). Consequently, the emergence of emotional trading robots managed by investment funds becomes a crucial area for investigation. Understanding the behavior of these algorithms enables proactive decision-making, allowing traders to leverage their knowledge and sell-purchased securities to these algorithms before their transactions occur. The implications of this research shed light on the evolving landscape of trading strategies and the role of emotionality in financial markets.

Prediction of colour strength in environmentally‐friendly dyeing of polyester fabric with madder using supercritical carbon dioxide

AbstractThe textile industry is one of the significant reasons for global water pollution, with dyeing processes being particularly environmentally detrimental. Researchers have explored alternative approaches to address this issue, such as using natural dyes, supercritical fluids and so forth. In addition to environment‐friendly approaches, reducing the number of experiments in studies, accurate production straightaway and using artificial intelligence (AI), one of the technologies of the present and the future that will provide significant support. Reaching clearer results with AI technology will not necessarily contribute to environment‐friendly technologies. However, AI techniques, including artificial neural networks (ANNs) and adaptive neuro fuzzy interface system (ANFIS) were employed to predict the colour strength (K/S) of the dyed fabric based on process parameters. A comprehensive experimental design involving pressure, temperature, and time variations was conducted, and the results were analysed using multi‐factor analysis of variance (MANOVA). The study demonstrates that supercritical carbon dioxide (scCO2) dyeing with madder on polyester fabric is a promising and environmentally friendly approach. Additionally, the optimised ANN and ANFIS models, aided by genetic algorithms (GAs), exhibit high predictive accuracy (less than 3%), providing insights into the impact of process parameters on colour strength. This research underscores the potential of AI‐driven automation in textile dyeing, offering solutions for dye formula prediction, colour matching, and defect detection, reducing the need for human intervention in these processes.

A new deep neuro-fuzzy system for Lyme disease detection and classification using UNet, Inception, and XGBoost model from medical images

A new deep neuro-fuzzy system for Lyme disease detection and classification using UNet, Inception, and XGBoost model from medical images

One to twelve-month-ahead forecasting of MODIS-derived Qinghai Lake area, using neuro-fuzzy system hybridized by firefly optimization.

Lakes, as the main sources of surface water, are of great environmental and ecological importance and largely affect the climatic conditions of the surrounding areas. Lake area fluctuations are very effective on plant and animal biodiversity in the areas covered. Hence, accurate and reliable forecasts of ​lake area might provide the awareness of water and climate resources and the survival of various species dependent on area fluctuations. Using machine learning methods, the current study numerically predicted area fluctuations of ​China's largest lake, Qinghai, over 1 to 12months ahead of lead time. To this end, Moderate Resolution Imaging Spectroradiometer (MODIS) sensor images were used to monitor the monthly changes in the area of ​the lake from 2000 to 2021. Predictive inputs included the MODIS-derived lake area time latency specified by the autocorrelation function. The data was divided into two periods of the train (initial 75%) and test (final 25%), and the input combinations were arranged so that the model in the test period could be used to predict 12 scenarios, including forecast horizons for the next 1 to 12months. The adaptive neuro-fuzzy inference system (ANFIS) was utilized as a predictive model. The firefly algorithm (FA) was also used to optimize ANFIS and improve its accuracy, as a hybrid model ANFIS-FA. Based on evaluation criteria such as root mean square error (RMSE) (477-594 km2) and R2 (88-92%), the results confirmed the acceptable accuracy of the models in all forecast horizons, even long-term horizons (10months, 11months, and 12months). Based on the normalized RMSE criterion (0.095-0.125), the models' performance was reported to be appropriate. Furthermore, the firefly algorithm improved the prediction accuracy of the ANFIS model by an average of 16.9%. In the inter-month survey, the models had fewer forecast errors in the dry months (February-March) than in the wet months (October-November). Using the current method can provide remarkable information about the future state of lakes, which is very important for managers and planners of water resources, environment, and natural ecosystems. According to the results, the current approach is satisfactory in predicting MODIS-derived fluctuations of Qinghai Lake area and has research value for other lakes.