Feed-forward Artificial Neural Network Research Articles

Daily scale streamflow forecasting in multiple stream orders of Cauvery River, India: Application of advanced ensemble and deep learning models

Accurate forecasts of streamflow (Qflow) are crucial for optimal management of water reservoir systems and preparing for catastrophic events such as floods. Although several machine learning (ML) techniques have been documented to forecast Qflow with notable accuracy, no model could do so with the utmost certainty since Qflow data can be noisy and sometimes incomplete. Despite these challenges, there has been significant progress in the field of Qflow forecasting. Hence, novel techniques need to be explored or applied to forecast Qflow with improved accuracy and reduced data dependency. Recently, optimized input selection techniques were shown to increase the performance of forecast models, although they have yet to be thoroughly investigated in Qflow forecasting applications. This study aims at the application of deep-learning and advanced ML techniques, namely the (i) Convolutional Neural Network (CNN), (ii) Random Forest (RF), and (iii) Gradient Tree Boosting (GTB), to forecast daily Qflow in multiple stream orders of the Cauvery River, India. The performance of these models was compared with that of conventional multilayer perceptron (MLP), a class of feedforward artificial neural network (ANN). Antecedent daily Qflow time series were used as model inputs, with input selection guided by Akaike’s Information Criteria (AIC) and Bayesian Information Criteria (BIC). The models were also calibrated using inputs selected from commonly used cross-correlation function for comparative evaluation with those models developed using AIC and BIC. The results show that applying AIC and BIC to select the ideal model input parameters marginally increase the model’s performance. The deep-learning and advanced ML techniques clearly outperformed the conventional ANN models in forecasting one- and three-day lead daily Qflow at multiple gauging stations considered along the Cauvery River. It was found that the deep learning (CNN) model had superior forecasting accuracy with respect to all 3 stations for the 1st Input-Output (I/O) combination. An NNSE, WI and KGE indices of 0.85 or higher indicates that the forecast is highly efficient.

AI-assisted study of auxetic structures

In this study, the viability of using machine learning models to predict stress-strain curves of auxetic structures based on geometry-describing parameters is explored. Given the computational cost and time associated with generating these curves through numerical simulations, a machine learning-based approach promises a more efficient alternative. A range of machine learning models, including Artificial Neural Networks, k-Nearest Neighbors Regression, Support Vector Regression, and XGBoost, is implemented and compared regarding the aptitude to predict stress-strain curves under quasi-static compressive loading. Training data is generated using validated finite element simulations. The performance of these models is rigorously tested on data not seen during training. The Feed-Forward Artificial Neural Network emerged as the most proficient model, achieving a Mean Absolute Percentage Error of 0.367 ± 0.230.

Ionospheric TEC prediction using FFNN during five different X Class solar flares of 2021 and 2022 and comparison with COKSM and IRI PLAS 2017

Abstract The Ionospheric Total Electron Content (TEC) measured in the ray path of the signals directly contributes to the Range Error (RE) of the satellite signals, which affects positioning and navigation. Employing the Co-Kriging-based Surrogate Model (COKSM) to predict TEC and RE correction has proven prolific. This research attempted to test and compare the prediction capability of COKSM with an Artificial Intelligence-based Feed Forward Neural Network model (FFNN) during five X-Class Solar Flares of 2021–22. Also, the results are validated by comparing them with the IRI PLAS 2017 model. TEC, solar, and geomagnetic parameters data for Hyderabad GPS station located at 17.31° N latitude and 78.55° E longitude were collected from IONOLAB & OMNIWEB servers. The COKSM uses six days of input data to predict the 7th day TEC, whereas prediction using the FFNN model is done using 45 days of data before the prediction date. The performance evaluation is done using RMSE, NRMSE, Correlation Coefficient, and sMAPE. The average RMSE for COKSM varied from 1.9 to 9.05, for FFNN it varied from 2.72 to 7.69, and for IRI PLAS 2017 it varied from 7.39 to 11.24. Likewise, evaluation done for three different models over five different X-class solar flare events showed that the COKSM performed well during the high-intensity solar flare conditions. On the other hand, the FFNN model performed well during high-resolution input data conditions. Also, it is notable that both models performed better than the IRI PLAS 2017 model and are suitable for navigational applications.

An Intelligent System to Improve Diagnostic Support for Oral Squamous Cell Carcinoma.

Oral squamous cell carcinoma (OSCC) is one of the most-prevalent cancer types worldwide, and it poses a serious threat to public health due to its high mortality and morbidity rates. OSCC typically has a poor prognosis, significantly reducing the chances of patient survival. Therefore, early detection is crucial to achieving a favorable prognosis by providing prompt treatment and increasing the chances of remission. Salivary biomarkers have been established in numerous studies to be a trustworthy and non-invasive alternative for early cancer detection. In this sense, we propose an intelligent system that utilizes feed-forward artificial neural networks to classify carcinoma with salivary biomarkers extracted from control and OSCC patient samples. We conducted experiments using various salivary biomarkers, ranging from 1 to 51, to train the model, and we achieved excellent results with precision, sensitivity, and specificity values of 98.53%, 96.30%, and 97.56%, respectively. Our system effectively classified the initial cases of OSCC with different amounts of biomarkers, aiding medical professionals in decision-making and providing a more-accurate diagnosis. This could contribute to a higher chance of treatment success and patient survival. Furthermore, the minimalist configuration of our model presents the potential for incorporation into resource-limited devices or environments.

A Radial Basis Function Neural Network Algorithm for the Simultaneous Retrieval of Two Meteorological Parameters from Solar Radiation

Local meteorological parameters are key in understanding the frequency of occurrence of extreme weather conditions such as floods, and droughts, among others. In this study, we present a method for simultaneous retrieval of two weather parameters. The method is based on already measured monthly average values of weather parameters from 2011 to 2016, which were used to train a Feed-forward radial basis function neural network (RBFNN) to obtain a fast and accurate method to compute global solar radiation for specified weather parameters pair. In inverse modelling, a multidimensional unconstrained non-linear optimization was employed to retrieve the weather parameters pair. The new approach was validated using weather parameter data measured at the Department of Physics, Makerere University (0.31° N, 32.58° E, 1200 m). Statistical tools were used to evaluate the method's performance. In the Feed-forward artificial neural network (ANN), the correlation coefficient (R), mean bias error (MnB), root mean square error (RMSE), and mean percentage error (MAPE) were in the ranges 0.80-0.95, -0.0011-0.0077, 0.55-1.04 and 2.49%-5.82%, respectively. The pairs (sunshine hours, relative humidity) and (sunshine hours, minimum temperature) had the highest correlation coefficient of 0.95. In the inverse artificial neural network (ANNi), the R, MnB, RMSE and MAPE were in the ranges 0.3-0.9, 0.01-0.08, 0.51-11.14 and 2.1%-14.5%, respectively. The pair (sunshine hours, relative humidity) had the highest correlation coefficients of 0.92 and 0.62, respectively. The method helps in obtaining weather parameter data sets in places where measuring equipment is lacking or during days when measuring equipment malfunctions.

Hybrid Modeling of a Simply-Supported Bridge Under Opposite Moving Loads via Multi-Layered Feed-Forward ANN

Keeping in view the design of short to medium-span bridges in critical locations such as road/river crossings under the action of oppositely moving high-speed loads, this paper presents a hybrid model for the prediction of dynamic displacement employing a theoretical non-dimensional framework aided by an artificial neural network (ANN). The introduction of such a hybrid model reduces parametric space and time. The feed-forward ANN is implemented to train the comprehensive data set obtained from the non-dimensional theoretical scheme. The dimensionless input to ANN comprises the bridge and load parameters, whereas maximum dynamic displacement is the output parameter. The robustness and efficiency of the proposed surrogate ANN framework in terms of computational efficiency are compared with the conventional mode-superposition method formulated in dimensionless form. The training of test and validated data sets results in best-fit models with optimistic statistical metrics and accuracy [Formula: see text] 6%. Moreover, the sensitivity of different parameters to the displacement response is highlighted using Pearson’s correlation and Quantile–Quantile plots. To incorporate the best-fit models into the design by bypassing the high-fidelity model, a user-friendly interface is developed, confirming the standards of the high-speed rating code (TZY2014-232 2014). Such interfaces have the potential for wide application in the preliminary design of short-to medium-span bridges under opposite moving loads in critical locations such as crossing valleys, rivers, and railroads.

Klasifikasi Hama Serangga pada Pertanian Menggunakan Metode Convolutional Neural Network

Insect pest attacks pose a serious threat that can potentially cause significant losses in agricultural production. Therefore, the effective recognition and control of insect pests are crucial for maintaining agricultural productivity and quality of yields. With the advancement of computer technology and artificial intelligence, computer technology can be utilized to automatically recognize images in object recognition, particularly for insect pest classification using the Convolutional Neural Network (CNN) method with the Xception architecture. CNN is one of the types of deep feed-forward artificial neural networks widely used in digital image analysis and can process data in the form of grid patterns. CNN consists of three types of layers: convolutional layer, pooling layer, and fully connected layer. The use of CNN in this research aims to facilitate the classification of insect pests. The CNN process involves stages of training, testing, and validation on insect pests to determine the classification of images of various insect pest species. This research utilizes 1363 image samples with 13 classes of insect pests. The training process of CNN involves several parameters such as batch size, number of epochs, learning rate, and optimizer. The experiment's results indicate that the best accuracy achieved by this model is 93.81% during the training phase and 81.75% during the validation phase. This demonstrates that the model successfully performs insect pest classification using the CNN method.

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.

A deep learning feed-forward neural network framework for the solutions to singularly perturbed delay differential equations

In this paper, we explore a deep learning feedforward artificial neural network (ANN) framework as a numerical tool for approximating the solutions to singularly perturbed delay differential equations (SPDDE). Our approach trains the network with fewer uniform data points along with linear interpolation as a dependent variable. More importantly, the exact solution is not used for training the network. A mean square or Euclidean norm type total loss function is utilized to assess the deep learning network’s performance. In the training and testing phase, our network adjusts itself to fit the tangent and the curvature by minimizing the total loss function and fine-tuning the network’s hyperparameters during the backpropagation stage. Our focus in this contribution is to investigate an answer to the question- “whether neural networks can learn to solve the differential equations with both delay and boundary layer behavior”? Traditional numerical methods are known to perform poorly in approximating the solutions to SPDDE due to boundary layer behavior, which is the sharp change in the gradient of the solution inside a region of very small width. Our proposed network architecture is used to investigate the above question for three varieties of SPDDE. The numerical results demonstrate that the fine-tuned adaptive deep learning architecture can effectively approximate the solution to SPDDE for varieties of delay and perturbation parameters. It is expected that the current method can be extendable to approximate 2D convection-diffusion partial differential equations with a boundary layer.

Application of GIS and feedforward back-propagated ANN models for predicting the ecological and health risk of potentially toxic elements in soils in Northwestern Nigeria.

Potentially toxic elements (PTEs) occur naturally in most geologic materials. However, recent anthropogenic disturbances such as ore mining have contributed significantly to their enrichment in soils. Their occurrence in soil may portend a myriad of related risks to the environment and biota. Most traditional soil quality evaluation methods involve comparing the background values of the elements to the established guideline values, which is often time-consuming and fraught with computational errors. As a result, to conduct a comprehensive and unbiased evaluation of soil quality and its effects on the ecosystem and human health, this research combined geochemical, numerical, and GIS data for a composite health risk zonation of the entire study area. Furthermore, the multilayer perceptron artificial neural network (MLP-NN) was used to forecast the most important toxic components influencing soil quality. Geochemical, statistical, and quantitative soil pollution evaluation (pollution index and ecological risk index) showed that apart from mining, the spread and association of trace elements and oxides occur as a consequence of surface environmental conditions (e.g., leaching, weathering, and organo-metallic complexation). The hazard quotients (HQs) and hazard index (HI) of all PTEs were greater than one. This indicates that residents (particularly children) are more susceptible to risks from toxic element ingestion than dermal exposure and inhalation. Ingestion of As and Cr resulted in higher cancer risks and lifetime cancer risk levels (> 1.0E 04), with risk levels increasing toward the northeastern, western, and southeastern directions of the study area. The low modeling errors observed from the sum of square errors, relative errors, and coefficient of determination confirmed the efficiency of the MLP-NN in pollution load prediction. Based on the sensitivity analysis, Hg, Sr, Zn, Ba, As, and Zr showed the greatest influence on soil quality. Focus on remediation should therefore be placed on the removal of these elements from the soil.

Non-exercise-based racing time prediction of cross-country skiers using machine learning methods combined with Relief-F feature selection

This study proposes new non-exercise models for estimating the racing time of cross-country skiers. Machine learning methods employed to build the prediction models include General Regression Neural Network (GRNN), Support Vector Machine (SVM), Multilayer Feed-Forward Artificial Neural Network (MFANN), and Radial Basis Function Neural Network (RBFNN), whereas the Relief-F algorithm combined with a ranker search has been utilized as the feature selector. The self-created data set contains samples collected from 370 cross-country skiers with inhomogeneous capabilities. Each sample in the data set contains physiological variables such as sex, age, height, weight, and body mass index (BMI) combined with an immersive set of survey data. The outcomes suggest that generally, the GRNN-based models exhibit the best prediction performance and can be used as a feasible tool for the prediction of the racing time of cross-country skiers with tolerable root mean square errors (RMSEs). It is seen that inclusion of age and assigned starting wave of cross-country skiers in models leads to much lower RMSEs, suggesting that the racing time of cross-country skiers is highly correlated to these two predictor variables. When compared with the exercise-based models, the proposed non-exercise-based models produce consistently comparable prediction performance for all evaluated machine learning methods. The non-exercise-based models have the relevant benefit of practical feasibility, as the models do not require the skiers to complete physical exercises and are also applicable to a wide range of cross-country skiers.

Artificial neural network model of non-Darcy MHD Sutterby hybrid nanofluid flow over a curved permeable surface: Solar energy applications

The conversion of solar radiation to thermal energy has recently much interest as the requirement for renewable heat and power grows. Due to their enhanced ability to promote heat transmission, nanofluids can significantly improve solar-thermal systems' efficiency. This section aims to study the heat transfer behavior of the Sutterby hybrid nanofluid flow of magnetohydrodynamics in the presence of a non-uniform heat source/sink and linear thermal radiation over a non-Darcy curved permeable surface. A novel implementation of an intelligent numerical computing solver based on multi-layer perceptron (MLP) feed-forward back-propagation artificial neural network (ANN) with the Levenberg-Marquard algorithm is provided in the current study. Data were gathered for the ANN model's testing, certification, and training. Established mathematical equations are nonlinear, which are resolved for velocity, the temperature in addition to the skin friction coefficient, and the rate of heat transfer by using the bvp4c with MATLAB solver. The ANN model selects data, constructs and trains a network, then evaluates its efficacy via mean square error. Graphs illustrate the impact of a wide range of physical factors on variables, including pressure, velocity, and temperature. In the entire study, the thermal energy improved by the SiO2 (silicon dioxide) - Au (gold) hybrid nanofluid than the SiO2-TiO2 (titanium dioxide) hybrid nanofluid. The higher internal heat generation/absorption parameter values increase the temperature.

Hybrid Deep Neural Network for Data Driven Missile Guidance with Maneuvering Target

Missile guidance, owing to highly complex and non-linear relative movement between the missile and its target, is a challenging problem. This is further aggravated in case of a maneuvering target which changes its own flight path while attempting to escape the incoming missile. In this study, to achieve computationally superior and accurate missile guidance, a deep learning is employed to propose a self-tuning technique for a fractional-order proportional integral derivative (FOPID) controller of a radar-guided missile chasing an intelligently maneuvering target. A multi-layer two-dimensional architecture is proposed for a deep neural network that combines the prediction feature of recurrent neural networks and estimation feature of feed-forward artificial neural networks. The proposed deep learning based missile guidance scheme is non-intrusive, data-based, and model-free wherein the parameters are optimized on-the-run while predicting the target’s maneuvering tactics to correct for processing time and loop delays of the system. Using deep learning for online optimization with minimal computational burden is the core feature of the proposed technique. Dual-core parallel simulations of missile-target dynamics and the control system were performed to demonstrate superiority of the proposed scheme in feasibility, adaptability, and the ability to effectively minimize the miss-distance in comparison with traditional and neural offline-tuned PID and FOPID based techniques. Compared to state-of-the-art offline-tuned neural control, the miss-distance was reduced by 68.42% for randomly maneuvering targets. Furthermore, a minimum miss-distance of 0.97 m was achieved for intelligently maneuvering targets for which the state-of-the-art method failed to hit the target. Overall, the proposed technique offers a novel approach for addressing the challenges of missile guidance in a computationally efficient and effective manner.

Exploring the Influence of Induced Magnetic Fields and Double-Diffusive Convection on Carreau Nanofluid Flow through Diverse Geometries: A Comparative Study Using Numerical and ANN Approaches

This current investigation aims to explore the significance of induced magnetic fields and double-diffusive convection in the radiative flow of Carreau nanofluid through three distinct geometries. To simplify the fluid transport equations, appropriate self-similarity variables were employed, converting them into ordinary differential equations. These equations were subsequently solved using the Runge–Kutta–Fehlberg (RKF) method. Through graphical representations like graphs and tables, the study demonstrates how various dynamic factors influence the fluid’s transport characteristics. Additionally, the artificial neural network (ANN) approach is considered an alternative method to handle fluid flow issues, significantly reducing processing time. In this study, a novel intelligent numerical computing approach was adopted, implementing a Levenberg–Marquardt algorithm-based MLP feed-forward back-propagation ANN. Data collection was conducted to evaluate, validate, and guide the artificial neural network model. Throughout all the investigated geometries, both velocity and induced magnetic profiles exhibit a declining trend for higher values of the magnetic parameter. An increase in the Dufour number corresponds to a rise in the nanofluid temperature. The concentration of nanofluid increases with higher values of the Soret number. Similarly, the nanofluid velocity increases with higher velocity slip parameter values, while the fluid temperature exhibits opposite behavior, decreasing with increasing velocity slip parameter values.

On Performance of Marine Predators Algorithm in Training of Feed-Forward Neural Network for Identification of Nonlinear Systems

Artificial neural networks (ANNs) are used to solve many problems, such as modeling, identification, prediction, and classification. The success of ANN is directly related to the training process. Meta-heuristic algorithms are used extensively for ANN training. Within the scope of this study, a feed-forward artificial neural network (FFNN) is trained using the marine predators algorithm (MPA), one of the current meta-heuristic algorithms. Namely, this study is aimed to evaluate the performance of MPA in ANN training in detail. Identification/modeling of nonlinear systems is chosen as the problem. Six nonlinear systems are used in the applications. Some of them are static, and some are dynamic. Mean squared error (MSE) is utilized as the error metric. Effective training and testing results were obtained using MPA. The best mean error values obtained for six nonlinear systems are 2.3 × 10−4, 1.8 × 10−3, 1.0 × 10−4, 1.0 × 10−4, 1.2 × 10−5, and 2.5 × 10−4. The performance of MPA is compared with 16 meta-heuristic algorithms. The results have shown that the performance of MPA is better than other algorithms in ANN training for the identification of nonlinear systems.

ANN-based energy management system for PV-powered EV charging station with battery backup and vehicle to grid support

ABSTRACT This paper proposes an Artificial Neural Network (ANN) based Energy Management System (EMS) for the economic operation of a Photovoltaic (PV) powered Electric Vehicle Charging Station (EVCS) with battery backup and Vehicle to Grid (V2G) support. The Charging Station Battery (CSB) and Parking Lot EVs (PLV) support the EVCS to operate economically by charging during off-peak hours and discharging during peak hours and also augment the reliability of the EVCS by serving as a backup supply during Stand-Alone mode. The proposed method reduces the operating cost of the PV powered EVCS while ensuring that proper power balance is maintained between the various components: PV, Electric Vehicles (EV), CSB, the PLV ready for V2G operation and the grid. This method relies on the information from real time data such as the connection status of the EV and PLV, State of Charge (SoC) of the EV, CSB and PLV, the available PV power and the Time of Use (ToU) tariff based on which the resources are allocated intelligently. For implementing this control action, a Supervised Feed Forward ANN model is developed which is trained by means of a set of data using various algorithms and it is observed that the Bayesian Regularisation algorithm outperformed the other algorithms in terms of prediction accuracy. With the proposed ANN-based EMS the energy utilized by the EVCS from the grid exhibited a reduction of 28% as compared to the system without EMS throughout the course of a simulation lasting 24 hours.

Detection of Credit Card Fraud with Artificial Neural Networks

Along with the Internet, digital technologies are frequently used in every moment of our lives. Many transactions that we carry out in monetary terms such as shopping in our daily life are now done digitally. With the developing digitalization in the world, people's lives become easier and people can access different products in a short time. In particular, people can spend and shop quickly and easily without carrying cash in their pockets with a credit card. However, with the increase in the use of credit cards, there are also some security vulnerabilities. Fraudsters can gain unfair advantage by obtaining certain credit card information such as passwords. They can shop with someone else's credit card without permission. These transactions cause substantial financial damage to individuals and institutions. With the increase in the use of credit cards with the developing technology, such credit card fraud is also increasing rapidly. Taking precautions against credit card fraud is a very important issue in order to ensure the safety of people. For this reason, in order to ensure the security of both banks and financial institutions that provide credit card services, it is necessary to prevent credit card fraud and to detect fraud that may occur in credit cards within the scope of combating fraud. In our study, Artificial Neural Networks were used to detect credit card fraud transactions. A prediction model has been developed to detect fraud in credit card transactions with ANN. Using the Credit Card data set obtained from the Kaggle database, modeling was done with the Feed Forward Artificial Neural Network method. The aim of this study is to automatically detect abnormal behaviors made with credit cards. 98.44% success was achieved with feedforward artificial neural network.

A machine learning tool for future prediction of heat release capacity of in-situ flame retardant hybrid Mg(OH)2-Epoxy nanocomposites

In this work, the fire behavior of a sol-gel in-situ hybrid Mg(OH)2-epoxy nanocomposite was investigated and an artificial neural network-based system built on a fully connected feed-forward artificial neural network was developed to predict its heat release capacity. The nanocomposite containing only ∼5 wt% loading of Mg(OH)2 promoted a remarkable decrease in heat release capacity (∼34%) measured by pyrolysis combustion flow calorimetry and in peak of heat release rate (∼37%), and heat release rate (∼29%), as assessed by cone calorimetry, as well as a significant decrease of total smoke release and smoke extinction area about 22 and 5%, respectively, indicating the suitability of Mg(OH)2 as an effective smoke suppressant. The proposed machine learning approach may be used as a promising alternative for a cost- and time-saving prediction of the fire performances of novel flame retardant polymer-based nanocomposites.

Feed-forward ANN and traditional machine learning-based prediction of biogas generation rate from meteorological and organic waste parameters

Feed-forward ANN and traditional machine learning-based prediction of biogas generation rate from meteorological and organic waste parameters

Machine learning instructed microfluidic synthesis of curcumin-loaded liposomes

The association of machine learning (ML) tools with the synthesis of nanoparticles has the potential to streamline the development of more efficient and effective nanomedicines. The continuous-flow synthesis of nanoparticles via microfluidics represents an ideal playground for ML tools, where multiple engineering parameters – flow rates and mixing configurations, type and concentrations of the reagents – contribute in a non-trivial fashion to determine the resultant morphological and pharmacological attributes of nanomedicines. Here we present the application of ML models towards the microfluidic-based synthesis of liposomes loaded with a model hydrophobic therapeutic agent, curcumin. After generating over 200 different liposome configurations by systematically modulating flow rates, lipid concentrations, organic:water mixing volume ratios, support-vector machine models and feed-forward artificial neural networks were trained to predict, respectively, the liposome dispersity/stability and size. This work presents an initial step towards the application and cultivation of ML models to instruct the microfluidic formulation of nanoparticles.