Artificial Neural Network System Research Articles

Micro-ECDM Performances Analysis Using Fuzzy Logic and ANFIS During Micro-Channel Fabrication on Silica Glass

Higher accuracy and meticulousness are highly demandable in modern industrial field during micro-machining performances by electrochemical discharge machining (ECDM) process. The paper deals with the experimental fuzzy logic control (FLC) analysis as well as artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) analysis during micro-channel fabrication on silica glass. A comparative analysis of FLC, ANN as well as ANFIS has been performed and experimental error prediction has been propounded to estimate minimum error possibilities such as mean absolute error (MAE), root mean square error (RMSE) and regression value (R). Computational time training dataset, minimum sample size and prediction time are also illustrated in this paper. In this paper, ANN, FLC and ANFIS models are analyzed for tool wear rate (TWR) and heat-affected zone (HAZ). 3D Rule Viewer and regression analysis as well as validation of test results and characteristics graph between performances with number of epochs of ANN model also are included in this paper for TWR and HAZ. Influence of process parameters like voltage, duty ratio, pulse frequency and electrolyte concentration on Surface Viewer of TWR and HAZ is also illustrated. It is found that ANFIS has great effectiveness of prediction of error during micro-ECDM process.

A Hybrid Optimization-Based Artificial Neural Network Model for Wireless Power Transfer in Electric Vehicles

Electric Vehicles (EVs) are powered by a battery mounted in the vehicle, which powers the motor and drives the wheels. Most commercial EVs can be charged by plugging them into a charging station. Such conductive recharging has various drawbacks, including physical plugging of the cable, safety concerns, and charging time. Manually charging EVs might be dangerous due to the chance of an electric spark or disaster. Advances in Wireless Power Transfer (WPT) demonstrate the capacity to transfer significant amounts of electricity over short and medium-range distances. The ultimate purpose of this paper is to improve the efficacy of electric car wireless charging systems. Here, a hybrid optimization-based Artificial Neural Network (ANN) model is applied to improve the efficacy of the WPT model in EVs. To optimize the weights of the ANN classifier, a hybrid approach termed as Grasshopper-Assisted Elephant Herd Optimization (GA-EHO) method is proposed. The GA-EHO is derived through the hybridization of Elephant Herd Optimization (EHO) Algorithm and the Grasshopper Optimization Algorithm (GOA) techniques. Finally, the experimental study reveals that at 70% learning rate, the proposed ANN system achieves a minimal MSE value of 0.0528, which is lower than other current classifiers, such as SVM, LSTM, and CNN.

Nanowire-Enhanced Fully Transparent and Flexible Indium Gallium Zinc Oxide Transistors with Chitosan Hydrogel Gate Dielectric: A Pathway to Improved Synaptic Properties.

In this study, a transparent and flexible synaptic transistor was fabricated based on a random-network nanowire (NW) channel made of indium gallium zinc oxide. This device employs a biocompatible chitosan-based hydrogel as an electrolytic gate dielectric. The NW structure, with its high surface-to-volume ratio, facilitated a more effective modulation of the channel conductance induced by protonic-ion polarization. A comparative analysis of the synaptic properties of NW- and film-type devices revealed the distinctive features of the NW-type configuration. In particular, the NW-type synaptic transistors exhibited a significantly larger hysteresis window under identical gate-bias conditions. Notably, these transistors demonstrated enhanced paired-pulse facilitation properties, synaptic weight modulation, and transition from short- to long-term memory. The NW-type devices displayed gradual potentiation and depression of the channel conductance and thus achieved a broader dynamic range, improved linearity, and reduced power consumption compared with their film-type counterparts. Remarkably, the NW-type synaptic transistors exhibited impressive recognition accuracy outcomes in Modified National Institute of Standards and Technology pattern-recognition simulations. This characteristic enhances the efficiency of practical artificial intelligence (AI) processes. Consequently, the proposed NW-type synaptic transistor is expected to emerge as a superior candidate for use in high-efficiency artificial neural network systems, thus making it a promising technology for next-generation AI semiconductor applications.

Physics-based and machine-learning models for accurate scour depth prediction.

Scour phenomena remain a significant cause of instability in offshore structures. The present study estimates scour depths using physics-based numerical modelling and machine-learning (ML) algorithms. For the ML prediction, datasets were collected from previous studies, and the trained models checked against the statistical measures and reported outcomes. The numerical assessment of the scour depth has been also carried out for the current and coupled wave-current environment within a computational fluid dynamics framework with the aid of the open-source platform REEF3D. The outcomes are validated against the previously reported experimental studies. The results obtained from ML schemes demonstrated that the artificial neural network and adaptive neuro-fuzzy interface system models have an elevated level of effectiveness compared with the other models. Whereas the numerical analysis results show a good agreement against the reported values. For the current only conditions, the normalized scour depth (S/D) at the front and rear end of the pier is 0.65 and 0.81. For the wave-current conditions, the normalized scour depth (S/D) is 0.26. The study highlights the importance of machine learning and physics-based numerical modelling to assess scour depth within a reasonable time frame without compromising accuracy. This article is part of the theme issue 'Physics-informed machine learning and its structural integrity applications (Part 2)'.

Prediction of Colour Characteristics of Microwave-Dried Tomato Slices Using Artificial Neural Network and Adaptive Neuro-Fuzzy Inference Systems

Summary Variation in the colour of dried tomatoes is frequently a problem for both consumers and processors. This study investigated digital imaging and applied soft-computational modelling using the Artificial Neural Network (ANN) and Adaptive Neuro-fuzzy Inference System (ANFIS) to evaluate the surface colour of microwave-dried tomato slices. The tomatoes were pretreated with water blanching, ascorbic acid, and sodium metabisulphite, then cut into slices of 4, 6, and 8 mm thickness. The slices were then dried in a microwave oven at power levels of 90, 180, and 360 W. The colour characteristics of the dried tomato slices (L*, a*, b*, colour change, browning index, hue, and chroma) were determined. The response variables were modelled and optimised using ANN and ANFIS. The efficiency and performance of the model were assessed using the coefficient of determination (R2), the root means square error (RMSE), and the mean absolute error (MAE). The results revealed the ranges of 36.70 – 48.83, 36.81 – 44.56, 31.03 – 40.34, 8.43 – 21.24, 11.78 – 39.82, 48.15 – 60.11, and 0.82 – 0.87 for the colour characteristics of L*, a*, b*, colour change, browning index, hue, and chroma, respectively. The outcomes showed that ANN and ANFIS models could make more accurate predictions. The predictive models were experimentally validated and agreed with the experimentally obtained values. However, the ANFIS model gave better performance, with higher values for R2 (1.000) and lower values for RMSE (0.02952) and MAE (0.02209). These findings will be helpful to processors and can be scaled up and adjusted for the bulk colour characteristics of microwave-dried tomatoes.

Estimating residential buildings’ energy usage utilising a combination of Teaching–Learning–Based Optimization (TLBO) method with conventional prediction techniques

Among the most significant solutions suggested for estimating energy consumption and cooling load, one can refer to enhancing energy efficiency in non-residential and residential buildings. A structure's characteristics must be considered when estimating how much heating and cooling is required. To design and develop energy-efficient buildings, it can be helpful to research the characteristics of connected structures, such as the kinds of cooling and heating systems needed to ensure sui interior air quality. As an important part of energy consumption and demand of buildings, the assessment of cooling load conditions from the envelope of large buildings has not been comprehensively understood yet. In the present paper, a new conceptual system has been developed to anticipate cooling load in the sector of residential buildings. Also, the paper briefly describes the major models of the developed system to maintain continuity and concentrate on the prediction model of the cooling load. To predict cooling load, authors have modelled two methods of artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) in conjunction with teaching-learning-based optimization (TLBO). This article aims to illustrate how artificial intelligence (AI) approaches play an essential role in addressing the mentioned necessity and help estimate the optimal design parameters for various stations. The value of the multiple determination coefficient is also determined. The values of the training R2 (coefficient of multiple determination) are 0.96446 and 0.97585 for TLBO-MLP and TLBO-ANFIS in the training stage and 0.95855 and 0.9721 in the testing stage, respectively, with an unknown dataset which is acceptable. The training RMSE values for TLBO-MLP and TLBO-ANFIS are 0.0685 and 0.11176 for training and 0.07074 and 0.12035 for testing, respectively, for the unknown dataset, which is acceptable. The lowest RMSE value and the higher R2 value indicate the favourable accuracy of the TLBO-MLP technique. According to the high value of R2 (97%) and the low value of RMSE, TLBO-MLP can predict residential buildings’ cooling load.

P2.05-06 An Artificial Neural Network System to Predict the Immune-related Adverse Events Based on the RNA Data of ORIENT-3 Study

P2.05-06 An Artificial Neural Network System to Predict the Immune-related Adverse Events Based on the RNA Data of ORIENT-3 Study

UV-photodegradation of R6G dye in three-phase fluidized bed reactor: Modeling and optimization using adaptive neuro-fuzzy inference system and artificial neural network

The development of an efficient photoreactor is still the most challenging task in the photocatalytic process. The main challenges include the determination of the optimal hydrodynamic conditions required for the efficient photodegradation of biorecalcitrant pollutants. In this study, a TiO2-ZnO/BAC composite catalyst was applied to investigate the hydrodynamic characteristics of a three-phase fluidized bed reactor in the UV-photodegradation of rhodamine 6G dye. Artificial neural network (ANN) and adaptive neuro-fuzzy inference systems (ANFIS) were used for modeling and optimization of the photodegradation process. From the preliminary study, 70° column inclination angle and 0.028 ms−1 superficial gas velocity, showed good axial solid distribution and average gas holdup. Subsequently, 0.019 ms−1 gas flow rate at 70° column inclination angle was found to be the optimum condition for the removal of rhodamine 6G (97.0 %). Moreover, solid distribution was found to be a dominant limiting factor in the photodegradation process as compared to gas holdup. Meanwhile, sensitivity analysis showed that all the input parameters (lamp position, inclination angle, and superficial gas velocity) were above 10%, confirming a strong influence on the process. The ANN-trainlm and ANFIS-hybrid with R-values of 0.9911 and 0.9546, respectively confirmed that the predicted model fits well with the experimental data. Furthermore, the inclination angle of 70° can be important in solar photoreactors to attain a relatively efficient tilt angle when using solar energy.

Empirical analysis of power quality using UPQC with hybrid control techniques

The Unified Power Quality Conditioner (UPQC), composed of both parallel active filters (PAF) and series active filters (SAF), recognized as a crucial device for enhancing the overall power quality of the system. In context, different control methods are employed to evaluate the performance of UPQC. This study aims to explore the performance of four distinct control techniques: dq, pq, Artificial Neural Network (ANN) and Adaptive Neuro Fuzzy Inference System (ANFIS) in hybrid combinations to manage both PAF and SAF in UPQC device. Four different hybrid cases are developed for addressing three power quality problems: harmonic distortion, voltage sag, and swell to evaluate the performance of these control techniques. The results revealed a significant reduction in Total Harmonic Distortion (THD) for hybrid ANFIS-ANN control method, achieving a THD value of 4.26 % for load voltage and 0.08 % for source current after compensation in comparison with other methods. Furthermore, this paper concludes by recommending the most suitable combination of control techniques for specific applications.

Optimization Studies on the Coagulation–Flocculation Process for PWW Treatment Using Cactus oputia Extract: Comparative Studies for Performance Evaluation

The treatment of Paint Waste Water, PWW is an important global issue for the minimization of water pollution especially in a developing country like Nigeria. The coagulation–flocculation process using Cactus Oputia Extract, COE as natural coagulant is reported for PWW treatment in the present study. The important process parameters pH, settling time, coagulant dosage and initial concentration were optimized using design of experiments (DOE). A full factorial composite experimental design and response surface methodology (RSM) were used to obtain the optimum values of the parameters. Also utilized in this study were Artificial Neural Network, ANN and Adaptive Neuro Fuzzy Inference System, ANFIS for performance evaluation. Using the numerical optimization technique of the design expert software, a combination of factors which concurrently fulfil the requirements placed on the response variable were determined. The best effluent removal efficiency was predicted to be 89.78% at the optimum values of the process parameters pH 6, settling time 30 mins, coagulant dosage 3 g and initial concentration 1.2 mg/l in this study.

Adaptive neuro‐fuzzy inference systems controller design on Buck converter

AbstractAdaptive neuro‐fuzzy inference system (ANFIS) approach is designed for a Buck converter. Because DC–DC converters are under the negative impact of different disturbances, a need for a well‐behaved technique is felt to provide higher robustness in various scenarios, including parametric variations, load uncertainty, supply voltage variation, and noise. Therefore, the fuzzy logic‐based controller is adopted for this structure that provides better error detection and correction, more comprehensive range of operating conditions, and is more readily customizable. However, the fuzzy technique suffers from slow dynamics, lack of reliability against broader range of disturbances, and has a huge computational burden. To overcome the weaknesses addressed before, this technique combined with an artificial neural network (ANN) system that can tune the fuzzy part resulting in an adaptive and robust structure. ANFIS method is a promising approach that has two soft‐computing control structures, including a fuzzy logic‐based part consisting of ANN. This combination has provided many significant benefits over fuzzy logic, such as low computational burden with faster dynamics, higher flexibility with adaptable rules, and a simple structure providing ease of practical implantation; also, it does not need a mathematical moulding of the system since the whole system has been considered as a Black‐box system. To better show the superiority of this method, two other control schemes are designed as fuzzy‐based PID technique and PID controller optimized by PSO algorithm. Finally, the ANFIS control strategy is tested in various working cases through simulation and experiment results as a beneficial alternative for practical applications.

2259P An artificial neural network system to predict the fraction of type I polarized macrophage

2259P An artificial neural network system to predict the fraction of type I polarized macrophage

Comparative Analysis of PWM AC Choppers with Different Loads with and Without Neural Network Application

In this paper, we focus on the "Artificial Neural Network (ANN) based PWM-AC chopper". This system is based on the PWM AC chopper-encouraged single-phase induction motor. The main purpose of this paper is to design and implement an ideal technique regarding speed control. Here analyzed PWM-based AC-AC converter with resistive load, R-L load and finally, the PWM AC chopper is fed to single phase induction for speed control. Using other soft computing and optimization techniques such as Artificial Neural Networks, Fuzzy Logic, Convolution algorithm, PSO, and Neuro Fuzzy can control the Speed. We used Artificial Neural Network to control the Speed of the PWM-AC Single phase induction motor drive. The Neural Network toolbox has been further used for getting desired responses. Neural system computer programs are executed in MATLAB. The performance of the proposed method of ANN system of PWM AC Chopper fed single phase induction motor drive is better than other traditional and base methods for controlling the Speed, based on the MOSFET.

Prediction and optimization of liquid dispersion of monoethanolamine in a rotating packed bed for CO2 absorption

Rotating packed bed (RPB) has recently gained traction as a preferred technology for the absorptive removal of CO2. In this device, solvent dispersion is the most crucial hydrodynamic parameter that directly affects the liquid-side mass transfer coefficient. The current study focuses on the optimization of the dispersion of monoethanolamine (MEA), a commonly used CO2 absorption solvent, in the RPB. Computational fluid dynamics (CFD) model was developed and validated against the published experimental data. The validated model was used to investigate the effect of operational parameters such as solvent concentration, rotational speed of the packed bed, inlet velocity, and the contact angle of packing material on MEA dispersion. The liquid dispersion of MEA was modeled using response surface methodology (RSM), artificial neural network (ANN) and adaptive neuro-fuzzy inference systems (ANFIS). The comparison of the modeled data revealed that ANN has the least mean square error (0.16) and is, therefore, the most capable model to predict the liquid dispersion of MEA in the RPB. Based on the ANOVA analysis, the inlet velocity of MEA is found to be an insignificant parameter and, therefore, does not contribute significantly to the liquid dispersion of MEA in the RPB. At optimized operating conditions, the MEA dispersion increased from 5346 to 10485 m−1, thus enhancing the physical absorption of CO2 up to eight times. CFD coupled with ANN for the prediction and RSM for the optimization of liquid dispersion can be employed to optimize industrial-scale RPB for the effective absorptive removal of CO2.

Intelligent Medical Velostat Pressure Sensor Mat Based on Artificial Neural Network and Arduino Embedded System

The promising research on flexible and tactile sensors requires conducting polymer materials and an accurate system for the transduction of pressure into electrical signals. In this paper, the intelligent sensitive mat, based on Velostat, which is a polymeric material impregnated with carbon black, is investigated. Various designs and geometries for home-made sensor mats have been proposed, and their electrical and mechanical properties, including reproducibility, have been studied through the tests performed. The mat pressure sensors have been interfaced with an Arduino microcontroller in order to monitor, read with high precision, and control the variation of the resistance under applied pressure. An approximation method was then developed based on a neural network algorithm to explore the relationship between different mat shapes, the pressure and stresses applied on the mat, the resistance of the conductive Velostat material, and the number of active sensing cells in order to control system input signal management.

Damping Undammed Low Frequency Oscillations in Power Systems: An ANN-Based Approach Using Pre-Disturbance Data

The paper examines undammed low frequency oscillations (LFOs) that can lead to system collapse, citing the Jordan power network incident on May 21, 2021. Traditional model-based methods for studying LFOs' small-signal stability have limitations. To address this, an online damping controller based on an artificial neural network (ANN) is proposed. Unlike existing ANN-based methods relying on offline controllers, this novel approach utilizes pre-disturbance data from phasor measurement units (PMUs) to dampen oscillations effectively. The paper addresses challenges of partially observable systems in online eigenvalue prediction using ANN. MATLAB is used to implement a feedforward ANN system trained on PMU data. The study involves a three-area test system with various operational scenarios, training the ANN across 406 scenarios to predict eigenvalues and damp LFOs.

Application of machine learning to predict of energy use efficiency and damage assessment of almond and walnut production

A study was conducted in Shahrekord city, Iran, focusing on improving the production of almond and walnut crops on rural agricultural lands. The gardeners selected for the study shared similar characteristics and production histories. One of the major challenges in producing these crops was the manual harvesting process, which required a significant amount of human labor in the region. To collect data, questionnaires and face-to-face interviews were conducted. The study used machine learning models, specifically artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) models, to predict energy use efficiency and environmental impacts in almond and walnut production. Among the models used, the ANFIS model with a three-level topology was found to be the most accurate in predicting output energy generation and environmental impacts in both almond and walnut production. The R2 values for the testing stage ranged from 0.969 to 0.996 for output energy generation and 0.994 to 0.997 for environmental impacts. The study demonstrated the effectiveness of using machine learning models like ANN and ANFIS in predicting energy use efficiency and environmental impacts in almond and walnut production, which can aid in planning and managing these crops more sustainably and efficiently in the future.

Quantitative and Qualitative Analyses of Mass Spectra of OEL Materials by Artificial Neural Network and Interface Evaluation: Results from a VAMAS Interlaboratory Study.

Quantitative analysis of binary mixtures of tris(2-phenylpyridinato)iridium(III) (Ir(ppy)3) and tris(8-hydroxyquinolinato)aluminum (Alq3) by using an artificial neural network (ANN) system to mass spectra was attempted based on the results of a VAMAS (Versailles Project on Advanced Materials and Standards) interlaboratory study (TW2 A31) to evaluate matrix-effect correction and to investigate interface determination. Monolayers of binary mixtures having different Ir(ppy)3 ratios (0, 0.25, 0.50, 0.75, and 1.00), and the multilayers containing these mixtures and pure samples were measured using time-of-flight secondary ion mass spectrometry (ToF-SIMS) with different primary ion beams, OrbiSIMS (SIMS with both Orbitrap and ToF mass spectrometers), laser desorption ionization (LDI), desorption/ionization induced by neutral clusters (DINeC), and X-ray photoelectron spectroscopy (XPS). The mass spectra were analyzed using a simple ANN with one hidden layer. The Ir(ppy)3 ratios of the unknown samples and the interfaces of the multilayers were predicted using the simple ANN system, even though the mass spectra of binary mixtures exhibited matrix effects. The Ir(ppy)3 ratios at the interfaces indicated by the simple ANN were consistent with the XPS results and the ToF-SIMS depth profiles. The simple ANN system not only provided quantitative information on unknown samples, but also indicated important mass peaks related to each molecule in the samples without a priori information. The important mass peaks indicated by the simple ANN depended on the ionization process. The simple ANN results of the spectra sets obtained by a softer ionization method, such as LDI and DINeC, suggested large ions such as trimers. From the first step of the investigation to build an ANN model for evaluating mixture samples influenced by matrix effects, it was indicated that the simple ANN method is useful for obtaining candidate mass peaks for identification and for assuming mixture conditions that are helpful for further analysis.

An artificial intelligence approach to model and optimize biodiesel production from used cooking oil using CaO incorporated zeolite catalyst

The current work investigated the possibility of employing chicken eggshell-zeolite composite as a cheap and recyclable heterogeneous catalyst for used cooking oil (UCO) conversion into its corresponding methyl ester (UCOME) via methanolysis process. Various catalysts were formulated by loading eggshell on the zeolite and calcined at different temperatures to obtain CaO incorporated zeolite (CaO-Zel) catalyst. The catalyst sample calcined at 800 °C for 4 h (CaO-Zel-800) exhibited best activity for methanolysis process and was analyzed using various techniques, including BET, surface basicity, TPD-CO2, TGA/DTA, SEM, XRD and FTIR. The transesterification process was modeled using artificial intelligence approach viz. artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) while optimization of the operating variables (temperature, catalyst loading, time and molar ratio) was performed by interfacing the developed models with ant colony optimization (ACO) algorithm. The closeness of coefficient of determination (R2) to unity and low mean square error (MSE) indicated that the methanolysis of UCO was adequately described by the developed models with ANFIS model (R2 = 0.9997 and MSE = 0.1271) superior to ANN model (R2 = 0.9953 and MSE = 2.0762). The highest UCOME yield of 99.45 ± 0.51 wt% was achieved with ANFIS-ACO under the condition; reaction temperature (72.97 °C), methanol/UCO molar ratio (7.14:1), reaction time (131.97 min) and catalyst dosage (5.39 wt%). The results of the sensitivity analysis revealed that all the operating variables influence UCOME yield, and none could be discarded. The CaO-Zel-800 catalyst was reused for five consecutive reaction cycles, and its activity only decreased by about 12%.

Development of Artificial Neural Network Model for Medical Specialty Recommendation

Timely diagnosis is crucial for a patient’s future care and treatment. However, inadequate medical service or a global pandemic can limit physical contact between patients and healthcare providers. Combining the available healthcare data and artificial intelligence methods might offer solutions that can support both patients and healthcare providers. This study developed one of the artificial intelligence methods, artificial neural network (ANN), the multilayer perceptron (MLP), for medical specialist recommendation systems. The input of the system is symptoms and comorbidities. Meanwhile, the output is the medical specialist. Leave one out cross-validation technique was used. As a result, this study’s F1 score of the model was about 0.84. In conclusion, the ANN system can be an alternative to the medical specialist recommendation system.