Resilient Propagation Research Articles

Prediction of Critical Flashover Voltage of High Voltage Insulators Leveraging Bootstrap Neural Network

Understanding the flashover performance of the outdoor high voltage insulator has been in the interest of many researchers recently. Various studies have been performed to investigate the critical flashover voltage of outdoor high voltage insulators analytically and in the laboratory. However, laboratory experiments are expensive and time-consuming. On the other hand, mathematical models are based on certain assumptions which compromise on the accuracy of results. This paper presents an intelligent system based on Artificial Neural Networks (ANN) to predict the critical flashover voltage of High-Temperature Vulcanized (HTV) silicone rubber in polluted and humid conditions. Various types of learning algorithms are used, such as Gradient Descent (GD), Levenberg-Marquardt (LM), Conjugate Gradient (CG), Quasi-Newton (QN), Resilient Backpropagation (RBP), and Bayesian Regularization Backpropagation (BRBP) to train the ANN. The number of neurons in the hidden layers along with the learning rate was varied to understand the effect of these parameters on the performance of ANN. The proposed ANN was trained using experimental data obtained from extensive experimentation in the laboratory under controlled environmental conditions. The proposed model demonstrates promising results and can be used to monitor outdoor high voltage insulators. It was observed from obtained results that changing of the number of neurons, learning rates, and learning algorithms of ANN significantly change the performance of the proposed algorithm.

Automated Prediction of Surgical Site Infection Coronary Artery Bypass (CABG) Grafting Surgery

Background: In 5 hospitals located in Belo Horizonte city (>3,000,000 inhabitants) a focused survey on surgical site infection (SSI) was performed in patients undergoing CABG surgery. We statistically evaluated such incidences to enable study of the prediction power of SSI through pattern recognition algorithms, in this case the multilayer perceptron (MLP) artificial neural networks. Methods: Data were collected between July 2016 and June 2018 on SSI by the hospital infection control committees (CCIHs) of the hospitals involved in the research. We collected all data used in the analysis during their routine SSI surveillance procedures. The information was forwarded to the NOIS (Nosocomial Infection Study) Project, which uses the SACIH (Automated Hospital Infection Control System) software to collect data from a sample of hospitals participating voluntarily in the project. After data collection, 3 procedures were performed: (1) a treatment of the collected database for use of intact samples; (2) a statistical analysis on the profile of the hospitals collected; and (3) an assessment of the predictive power of 5 types of MLP (ie, backpropagation standard, momentum, resilient propagation, weight decay, and quick propagation) for SSI prediction. MLPs were tested with 3, 5, 7, and 10 hidden layer neurons and a database split for the resampling process (65% or 75% for testing and 35% or 25% for validation). They were compared by measuring the AUC (area under the curve; range, 0–1) presented for each of the configurations. Results: From 666 initial data, only 278 were able for analysis. We obtained the following statistics: 9.35% manifested SSIs; length of stay varied from 1 to 119 days, with ~40% staying between 10 and 19 days; 15.1% of the patients died. Regarding the prediction power of SSI, the experiments have a maximum value of 0.713. Conclusions: Despite the considerable loss rate of >50% of the database samples due to the presence of noise, it was possible to have a relevant sampling to evaluate the profile of hospitals in Belo Horizonte. In addition, for the predictive process, although some configurations had results equal to 0.5, others reached 0.713, which indicates that the automated SSI monitoring framework for patients undergoing coronary artery bypass grafting surgery is promising. To optimize data collection and to enable other hospitals to use the SSI prediction tool (available at www.sacihweb.com), a mobile application was developed.Funding: NoneDisclosures: None

Artificial Neural Networks Applied to Prediction to Assess the Likelihood of Surgical Site Infection in Different Surgeries

Background: Based on data obtained from hospitals in the city of Belo Horizonte (population ~3,000,000), we evaluated relevant factors such as death, age, duration of surgery, potential for contamination and surgical site infection, plastic surgery, and craniotomy. The possibility of predicting surgical site infection (SSI) was then analyzed using pattern recognition algorithms based on MLP (multilayer perceptron). Methods: Data were collected by the hospital infection control committees (CCIHs) in hospitals in Belo Horizonte between 2016 and 2018. The noisy records were filtered, and the occurrences were analyzed. Finally, the predictive power of SSI of 5 types MLP was evaluated experimentally: momentum, backpropagation standard, weight decay, resilient propagation, and quick propagation. The model used 3, 5, 7, and 10 neurons in the occult layer and with resamples varied the number of records for testing (65% and 75%) and for validation (35% and 25%). Comparisons were made by measuring the AUC (area under the curve (range, 0–1). Results: From 1,096 records of craniotomy, 289 were usable for analysis. Moreover, 16% died; averaged age was 56 years (range, 40–65); mean time of surgery was 186 minutes (range, 95–250 minutes); the number of hospitalizations ranged from 1 (90.6%) to 8 (0.3%). Contamination among these cases was rated as follows: 2.7% contaminated, 23.5% potentially contaminated, 72.3% clean. The SSI rate reached 4%. The prediction process in AUCs ranged from 0.7 to 0.994. In plastic surgery, from 3,693 records, 1,099 were intact, with only 1 case of SSI and no deaths. The average age for plastic surgery was 41 years (range, 16–91); the average time of surgery was 218.5 minutes (range, 19–580 minutes); the number of hospitalizations ranged from 1 (77.4%) to 6 times (0.001%). Contamination among these cases was rated as follows: 27.90% potential contamination, 1.67% contaminated, and 0.84% infected. The prediction process ranged in AUCs from 0.2 to 0.4. Conclusions: We identified a high noise index in both surgeries due to subjectivity at the time of data collection. The profiles of each surgery in the statistical analyses were different, which was reflected in the analyzed structures. The MLP for craniotomy surgery demonstrated relevant predictive power and can guide intelligent monitoring software (available in www.sacihweb.com). However, for plastic surgeries, MLPs need more SSI samples to optimize outcomes. To optimize data collection and to enable other hospitals to use the SSI prediction tool, a mobile application was developed.Disclosures: NoneFunding: None

Pattern Recognition Algorithms for Predicting Surgical Site Infection in Abdominal Hysterectomy

Background: This research represents an experiment based in surgical site infection (SSI) to patients undergoing abdominal hysterectomy surgery procedures in hospitals in Belo Horizonte, (population, 3 million). We statistically evaluated such incidences and studied the SSI prediction power of pattern recognition algorithms, the artificial neural networks based in multilayer perceptron (MLP). Methods: Between July 2016 and June 2018, data on SSI were collected by the hospital infection control committees (CCIH) of the 3 hospitals involved in the research. They collected all data used in the analysis during their routine SSI surveillance procedures. The information was forwarded to the NOIS (Nosocomial Infection Study) Project, which used SACIH (ie, automated hospital infection control system software) to collect data from a sample of hospitals participating voluntarily in the project. After data collection, 3 procedures were performed for SSI prediction: (1) a treatment of the database collected for the use of intact samples; (2) a statistical analysis on the profile of the hospitals collected; and (3) an assessment of the predictive power of 5 types of MLP (ie, backpropagation standard, momentum, resilient propagation, weight decay, and quick propagation). MLPs were tested with 3, 5, 7, and 10 hidden-layer neurons and a database split for the resampling process (65% or 75% for testing, 35% or 25% for validation). They were compared by measuring area under the curve (AUC; range, 0–1) presented for each of the configurations. Results: From 1,166 records collected, only 665 records were enabled for analysis. Regarding statistical data: the average duration of surgery was 100 minutes (range, 31–180); patients were aged 41–49 years; the SSI rate was low (only 10 cases); the average length of stay was 2 days; and there were no deaths among the cases. Moreover, 29% of the operative sites were contaminated and 57% were potentially contaminated, revealing a high rate of potential contamination in the operative sites. The prediction process achieved 0.995. Conclusions: Despite the noise in the database, it was possible to obtain a relevant sampling to evaluate the profile of hospitals in Belo Horizonte. In addition, for the predictive process, although some settings achieved AUC results of 0.5, others achieved and AUC of 0.995, indicating the promise of the automated SSI monitoring framework for abdominal hysterectomy surgery (available in www.sacihweb.com). To optimize data collection and to enable other hospitals to use the SSI prediction tool, a mobile application was developed.Funding: NoneDisclosures: None

Automated Risk Analysis of Surgical Site Infection in Hip Arthroplasty Surgeries

Background: In 7 hospitals in Belo Horizonte, a city with >3,000,000 inhabitants, a survey was conducted between July 2016 and June 2018, focused on surgical site infection (SSI) in patients undergoing arthroplasty surgery procedures. The main objective is to statistically evaluate such incidences and enable a study of the prediction power of SSI through pattern recognition algorithms, the MLPs (multilayer perceptron). Methods: Data were collected on SSI by the hospital infection control committees (CCIHs) of the hospitals involved in the research. All data used in the analysis during their routine SSI surveillance procedures were collected. The information was forwarded to the NOIS (Nosocomial Infection Study) Project, which used SACIH automated hospital infection control system software to collect data from a sample of hospitals participating voluntarily in the project. After data collection, 3 procedures were performed: (1) a treatment of the database collected for the use of intact samples; (2) a statistical analysis on the profile of the hospitals collected; and (3) an assessment of the predictive power of 5 types of MLP (backpropagation standard, momentum, resilient propagation, weight decay, and quick propagation) for SSI prediction. MLPs were tested with 3, 5, 7, and 10 hidden layer neurons and a database split for the resampling process (65% or 75% for testing and 35% or 25% for validation). The results were compared by measuring AUC (area under the curve; range, 0–1) presented for each of the configurations. Results: Of 1,246 records, 535 were intact for analysis. We obtained the following statistics: the average surgery time was 190 minutes (range, 145–217 minutes); the average age of the patients was 67 years (range, 9–103); the prosthetic implant index was 98.13%; the SSI rate was 1.49%, and the death rate was 1.21%. Regarding the prediction power, the maximum prediction power was 0.744. Conclusions: Despite the considerable loss rate of almost 60% of the database samples due to the presence of noise, it was possible to perform relevant sampling for the profile evaluation of hospitals in Belo Horizonte. For the predictive process, some configurations have results that reached 0.744, which indicates the usefulness of the structure for automated SSI monitoring for patients undergoing hip arthroplasty surgery. To optimize data collection and to enable other hospitals to use the SSI prediction tool (available in www.sacihweb.com ), a mobile application was developed.Funding: NoneDisclosures: None

Resilient Backpropagation Neural Network on Prediction of Poverty Levels in South Sulawesi

Poverty is a topic that continues and is always discussed up to this time, as a benchmark indicator of how the level of welfare and prosperity in the lives of people in a country. Several attempts have been made by the central and regional governments to reduce poverty levels, including “Bantuan Langsung Tunai” (BLT) and the “Program Keluarga Harapan” (PKH). However, poverty reduction in Indonesia is still slowing down, including in South Sulawesi. Based on this, this study aims to predict poverty levels in South Sulawesi. Factors thought to influence poverty levels are the Human Development Index (HDI), the Open Unemployment Rate (TPT), and the Gross Regional Domestic Product (GRDP). The data used are data from 2010 to 2014. The method used is a backpropagation neural network with a resilient algorithm or better known as a resilient backpropagation neural network (RBNN). The results of the prediction of poverty levels using predictors of HDI, TPT, and GRDP showed that the analysis of the RBNN reached its optimum using architecture [3- 9 - 1] and reached convergence at the 81th iteration with an accuracy rate of 95.34%.

A Bayesian Regularized Artificial Neural Network for Simultaneous Determination of Loratadine, Naproxen and Diclofenac in Wastewaters

Background:: Simultaneous determination of medication components in pharmaceutical samples using ordinary methods have some difficulties and therefore these determinations usually were made by expensive methods and instruments. Chemometric methods are an effective way to analyze several components simultaneously. Objective:: In this paper, a novel approach based on Bayesian regularized artificial neural network is developed for the determination of Loratadine, Naproxen, and Diclofenac in water using UV-Vis spectroscopy. Methods: A dataset is collected by performing several chemical experiments and recording the UV-Vis spectra and actual constituent values. The effect of a different number of neurons in the hidden layer was analyzed based on final mean square error, and the optimum number was selected. Principle Component Analysis (PCA) was also applied to the data. Other back-propagation methods, such as Levenberg-Marquardt, scaled conjugate gradient, and resilient backpropagation, were tested. Results:: In order to see the proposed network performance, it was performed on two crossvalidation methods, namely partitioning data into train and test parts, and leave-one-out technique. Mean square errors between expected results and predicted ones implied that the proposed method has a strong ability in predicting the expected values. Conclusion:: he results showed that the Bayesian regularization algorithm has the best performance among other methods for simultaneous determination of Loratadine, Naproxen, and Diclofenac in water samples.

Analyzing the Contribution of Training Algorithms on Deep Neural Networks for Hyperspectral Image Classification

Autoencoder (<small>AE</small>)-based deep neural networks learn complex problems by generating feature-space conjugates of input data. The learning success of an AE is too sensitive for a training algorithm. The problem of hyperspectral image (<small>HSI</small>) classification by using spectral features of pixels is a highly complex problem due to its multi-dimensional and excessive data nature. In this paper, the contribution of three gradient-based training algorithms (i.e., scaled conjugate gradient (<small>SCG</small>), gradient descent (<small>GD</small>), and resilient backpropagation algorithms (<small>RP</small>)) on the solution of the HSI classification problem by using AE was analyzed. Also, it was investigated how neighborhood component analysis affects classification performance for training algorithms on HSIs. Two hyperspectral image classification benchmark data sets were used in the experimental analysis. Wilcoxon signed-rank test indicates that RB is statistically better than SCG and GD in solving the related image classification problem.

Comprehensive chemometric classification of snack products based on their near infrared spectra

The authentication and quality assurance of snack products have become important, since these convenience foods are popular in the modern lifestyle. Near infrared spectroscopy with machine learning algorithms can be used for the determination of qualitative properties of these products. Our study focuses on the determination of four important qualitative aspects of snacks: a) frying oil, b) raw material, c) place of origin, d) production technology. Our aim was to classify snack samples from various countries and producers successfully based on the aforementioned criteria. Three well-known machine learning algorithms, namely partial least squares discriminant analysis (PLS-DA), multilayer feed-forward of resilient backpropagation network (RPropMLP) and random forest (RF) were applied for the task, with custom validation protocols, optimized for the number of samples. Cross-validation and test validation were used to verify the robustness of the models. Accuracy of the models was above 0.80 in each case. The results showed that the neural network-based algorithm outperformed the other algorithms in every case based on the accuracy and area under the ROC curve values, which reveals a potential advantage of neural network-based algorithms in the case of smaller datasets. The models can be easily implemented to the quality control of snack products.

Breast Cancer Diagnosis Using Multi-Stage Weight Adjustment In The MLP Neural Network

Abstract Breast cancer is the most common kind of cancer, which is the cause of death among the women worldwide. There is evidence that shows that the early detection and treatment can increase the survival rate of patients who suffered this disease. Therefore, this paper proposes an automatic breast cancer diagnosis technique using a genetic algorithm for simultaneous feature selection and parameter optimization of an Multi Layer Perceptron (MLP) neural network. The aim of this paper is to propose a hybrid classification algorithm based on Multi-stage Weights Adjustment in the MLP (MWAMLP) neural network in two parts to improve the breast cancer diagnosis. In the first part, the three classifiers are trained simultaneously on the learning dataset. The output of the first part classifier together with the learning dataset is placed in a new dataset. This dataset uses a hybrid classifier method to model the mapping between the outputs of each ordinary classifier of the first part with real output labels. The proposed algorithm is implemented with three different variations of the backpropagation (BP) technique, namely the Levenberg–Marquardt, resilient BP and gradient descent with momentum for fine tuning of the weight of MLP neural network and their performances are compared. Interestingly, one of the proposed algorithms titled MWAMLP-RP produces the best and on average, 99.35% and 98.74% correct classification, respectively, on the Wisconsin Breast Cancer Database dataset, which is comparable with the obtained results from the methods titled GP-DLNN, GAANN and CAFS and other works found in the literature.

Machine learning-based left ventricular hypertrophy detection using multi-lead ECG signal

This work proposes a novel method for the detection of Left Ventricular Hypertrophy (LVH) from a multi-lead ECG signal. Left Ventricle walls become thick due to prolonged hypertension which may fail to pump heart effectively. The imaging techniques can be used as an alternative diagnose LVH; however, they are more expensive and time-consuming than proposed LVH. To overcome this issue, an algorithm to the diagnosis of LVH using ECG signal based on machine learning techniques were designed. In LVH detection, the pathological attributes such as R wave, S wave, inversion of QRS complex, changes in ST segment noticed in the ECG signal. This clinical information extracted as a feature by applying continuous wavelet transform. The signals were reconstructed with the frequency between 10 and 50 Hz from the wavelet. This followed by the detection of R wave and S wave peaks to obtain the relevant LVH diagnostic features. The Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Ensemble of Bagged Tree, AdaBoost classifiers were employed and the results are compared with four neural network classifiers including Multilayer Perceptron (MLP), Scaled Conjugate Gradient Backpropagation Neural Network (SCG NN), Levenberg–Marquardt Neural Network (LMNN) and Resilient Backpropagation Neural network (RPROP). The data source includes Left Ventricular Hypertrophy and healthy ECG signal from PTB diagnostic ECG database and St Petersburg INCART 12-Lead Arrhythmia Database. The results revealed that the proposed work can diagnose LVH successfully using neural network classifiers. The accuracy in detecting LVH is 86.6%, 84.4%, 93.3%,75.6%, 95.6%, 97.8%, 97.8%, 88.9% using SVM, KNN, Ensemble of Bagged Tree, AdaBoost, MLP, SCG NN, LMNN and RPROP classifiers, respectively.

Productivity estimate using regression and artificial neural networks in small familiar areas with agrosilvopastoral systems

This study aimed to model the height of trees and volumetric production of eucalypts trees on the agrosilvopastoral systems (AGP) of Zona da Mata Mineira region, Brazil, using artificial neural network (ANN) and regression models to determine the best alternative. The data was obtained from five systems with different spatial arrangements (8 × 3 m, 10 × 3 m, 11 × 3 m, 12 × 3 m, 12 × 2 m, and 12 × 4 m), ages (5.5, 6.5 and 8 years) and genotypes, of which 122 sample trees were scaled. Hypsometric and volumetric models were adjusted considering no stratification or stratification by the AGP, spatial arrangement, and genotype. A multilayer perceptron ANN was trained using resilient propagation and skip layer training algorithms. The stratification variables used in the regression were used in the ANN as categorical variables. To estimate height of trees were used as continuous variables: diameter at breast height (dbh), dominant height (Dh), and age. To estimate volume were used as continuous variables: dbh, total height, and age. The AGPs’s mean annual increment at 5.5, 6.5 and 8 years of age ranged from 21 to 62 m3 ha−1 year−1. ANN was proven to be an efficient methodology for hypsometric and volumetric estimates of eucalypt in AGP in the study region.

Smart modeling of viscosity of viscoelastic surfactant self-diverting acids

The Viscoelastic Surfactant (VES) based self-diverting acids, the recent generation of the acid diverters, have been widely applied to alleviate the non-uniform distribution of the treatment fluids into the heterogeneous porous medium. Measuring the rheological characteristics of the aforesaid acids is an unavoidable step in designing an acid stimulation project. The purpose of this study is to generate smart soft-computing models for prognosticating the viscosity of VES-based acids as a function of five independent variables including VES concentration, temperature, shear rate, pH value and the concentration of the Ca2+ (or CaCl2). Robust types of artificial intelligence methodologies including Multilayer Perceptron (MLP), Radial Basis Function (RBF), and Group Method of Data Handling (GMDH) were trained to achieve both computer-based and user-accessible models for the scope of this research. The MLP approach was optimized with four diverse algorithms including Resilient Backpropagation (RB), Bayesian Regularization (BR), Levenberg-Marquardt (LM), and Scaled Conjugate Gradient (SCG). The Differential Evolution (DE) and Independent Component Analysis (ICA) techniques were utilized to precisely tune the control parameters of the RBF network. Accordingly, several models were generated over three experimental data sets collected from the literature. The truthfulness and reliability of the developed models were assessed by employing several statistical parameters and graphical analyses. According to the results, all the proposed smart models were thoroughly victorious for predicting the VES-based acid viscosity. The MLP-SCG strategy outperformed other algorithms with Root Mean Square Error (RMSE), Average Absolute Relative Deviation (AARD), and R2 values of 0.0025, 0.6265%, and 0.9998, respectively. The so-called outliers analysis reveals that all datapoints used for modeling are located in the applicability domain leading to the database validity and model reliability. Furthermore, a comparison between the smart models proposed here against literature traditional correlations demonstrates the superiority of the intelligent models for the scope of this study. At last, it should be expressed that the created strong tools here could be of high weight for the successful design and implementation of matrix acidizing.

Finite-emittance Wigner crystals in the bubble regime

AbstractWe study the influence of finite-emittance electron bunches in the bubble regime of laser-driven wakefield acceleration onto the microscopic structure of the bunch itself. Using resilient backpropagation (Rprop) to find the equilibrium structure, we observe that for realistic and already observed emittances the previously found crystalline structures remain intact and are only widened marginally. Higher emittances lead to larger electron displacements within the crystal and finally its breaking.

Automated fast computational adaptive optics for optical coherence tomography based on a stochastic parallel gradient descent algorithm.

The transverse resolution of optical coherence tomography is decreased by aberrations introduced from optical components and the tested samples. In this paper, an automated fast computational aberration correction method based on a stochastic parallel gradient descent (SPGD) algorithm is proposed for aberration-corrected imaging without adopting extra adaptive optics hardware components. A virtual phase filter constructed through combination of Zernike polynomials is adopted to eliminate the wavefront aberration, and their coefficients are stochastically estimated in parallel through the optimization of the image metrics. The feasibility of the proposed method is validated by a simulated resolution target image, in which the introduced aberration wavefront is estimated accurately and with fast convergence. The computation time for the aberration correction of a 512 × 512 pixel image from 7 terms to 12 terms requires little change, from 2.13 s to 2.35 s. The proposed method is then applied for samples with different scattering properties including a particle-based phantom, ex-vivo rabbit adipose tissue, and in-vivo human retina photoreceptors, respectively. Results indicate that diffraction-limited optical performance is recovered, and the maximum intensity increased nearly 3-fold for out-of-focus plane in particle-based tissue phantom. The SPGD algorithm shows great potential for aberration correction and improved run-time performance compared to our previous Resilient backpropagation (Rprop) algorithm when correcting for complex wavefront distortions. The fast computational aberration correction suggests that after further optimization our method can be integrated for future applications in real-time clinical imaging.

Resilient back-propagation approach in small-world feed-forward neural network topology based on Newman–Watts algorithm

The scientific researches are focused on network topologies and training algorithms fields because they reduce overfitting problem in artificial neural networks. In this context, we showed in our previous work that Newman–Watts small-world feed-forward artificial neural networks present better classification and prediction performance than conventional feed-forward artificial neural networks. In this study, we investigate the effects of the Resilient back-propagation algorithm on SW network topology and propose a Resilient Newman–Watts small-world feed-forward artificial neural network model by assuming fixed initial topological conditions. We find that Resilient small-world network further reduces overfitting and further increases the network performance when compared to the conventional feed-forward artificial neural networks. Furthermore, it is shown that the proposed network model does not increase the algorithmic complexity as per other models. The obtained results imply that the proposed model can contribute to the solving of overfitting problem encountered in both deep neural networks and conventional artificial neural networks.

Performance Evaluation of Training Algorithms in Backpropagation Neural Network Approach to Blast-Induced Ground Vibration Prediction

Abstract
 Backpropagation Neural Network (BPNN) is an artificial intelligence technique that has seen several applications in many fields of science and engineering. It is well-known that, the critical task in developing an effective and accurate BPNN model depends on an appropriate training algorithm, transfer function, number of hidden layers and number of hidden neurons. Despite the numerous contributing factors for the development of a BPNN model, training algorithm is key in achieving optimum BPNN model performance. This study is focused on evaluating and comparing the performance of 13 training algorithms in BPNN for the prediction of blast-induced ground vibration. The training algorithms considered include: Levenberg-Marquardt, Bayesian Regularisation, Broyden–Fletcher–Goldfarb–Shanno (BFGS) Quasi-Newton, Resilient Backpropagation, Scaled Conjugate Gradient, Conjugate Gradient with Powell/Beale Restarts, Fletcher-Powell Conjugate Gradient, Polak-Ribiére Conjugate Gradient, One Step Secant, Gradient Descent with Adaptive Learning Rate, Gradient Descent with Momentum, Gradient Descent, and Gradient Descent with Momentum and Adaptive Learning Rate. Using ranking values for the performance indicators of Mean Squared Error (MSE), correlation coefficient (R), number of training epoch (iteration) and the duration for convergence, the performance of the various training algorithms used to build the BPNN models were evaluated. The obtained overall ranking results showed that the BFGS Quasi-Newton algorithm outperformed the other training algorithms even though the Levenberg Marquardt algorithm was found to have the best computational speed and utilised the smallest number of epochs.
 
 Keywords: Artificial Intelligence, Blast-induced Ground Vibration, Backpropagation Training Algorithms

빅데이터를 이용한 인공신경망 기반 시간별 전력수요 예측

Short-term power load prediction is crucial for scheduling generators, securing transmission capacity, and determining economic market price. This paper proposes a Bigdata-based Artificial Neural Network (B-ANN) model for a short-term load forecasting by utilizing decision-tree method. The bigdata has been composed of 2018 time index, meteorological data, exchange rates, oil price, and past power demand in Seoul. The decision-tree method is applied to classify the data in accordance with the value of decreasing entropy. In addition, the resilient back propagation algorithm (RPROP) is applied for ANN that derives fast results and result in reduced error through variable learning rates. The applicability and effectiveness of the proposed model are verified by conducting simulations for forecasting 1-day hourly power load. The results confirm that the proposed model has decreased the mean absolute percentage error (MAPE) dramatically in comparison with other forecasting methods such as multiple regression analysis, time series analysis, and auto-regressive integrated moving average (ARIMA) model.

Polypropylene nanofibre behaviour modelling for uranium adsorption by forcespinning method: artificial neural network modelling

ABSTRACT In this study, a handmade device was developed to prepare polymeric polypropylene (PP) nanofibres by forcespinning method. The characterisations of the prepared adsorbent were investigated by Fourier transform infrared spectroscopy (FTIR) and Scanning electron microscopy (SEM). Then, the performance of the prepared nanofibres was investigated as an adsorbent for the removal of uranium (VI) ion from the aqueous solution. Experimental parameters such as contact time and pH were optimised. The maximum adsorption was obtained with the adsorbent weight of 0.03 g, contact time of 60 min, and pH of 4. This study showed that PP nanofibre is an efficient adsorbent for uranium removal from aqueous solution. Also, the uranium removal process was modelled by artificial neural network (ANN) for predicting capacity removal. In this method, feed-forward back-propagation learning with Levenberg-Marquardt (LM), adaptive learning rate back-propagation (GDX) and resilient back-propagation (RP) algorithms were used.

Mathematical Modeling of Photovoltaic Properties of Nipc/P-Si (Organic/Inorganic) Heterojunction by Using Artificial Neural Networks Model

In this research, the artificial neural network (ANN) and resilient back propagation (R-prop) training algorithm are utilized to model the photovoltaic properties of Nickel–phthalocyanine (NiPc/p-Si) heterojunction. The experimental data are extracted from experimental studies. Experimental data are utilized as inputs in the ANN model. Training of different structures of the ANN is processed to approach the minimum value of error. Eight artificial neural networks are trained to get a better mean square error (MSE) and best execution for the networks. The ANN performances are also investigated and their values are very small (MSE < 10-3). The simulation results of the current-voltage characteristics of NiPc films are produced and provided excellent matching with the corresponding experimental data. Utilization of ANN model for predictions is also processed and gives accurate results. The equation which describes the relation between the inputs and outputs is obtained. The high accuracy of the ANN model has appeared in the major guessing power and the ability of generalization depending on the obtained equations.