Artificial Neural Network Method Research Articles

Türkiye’de 2030 Yılına Kadar Bina Sektöründeki Enerji Tüketim Tahmini

As in the rest of the world, energy consumption is increasing in Turkey as a result of population growth and technological developments. Energy efficiency studies are becoming more and more important for Turkey, which meets more than half of its energy needs from imported sources. When the sectors where energy is consumed intensively are analyzed, it is seen that the share of building sector energy consumption in total energy consumption is approximately 20%. This high rate shows that energy efficiency studies in the building sector will reduce the amount of energy imports and the current account deficit, as well as help to ensure security of supply.
 In this study, Turkey's population, Gross Domestic Product (GDP), number of buildings and the total amount of energy consumed in buildings in the past 20 years are used to estimate how much energy will be consumed in the building sector for the future. Within the scope of the study, how much energy will be consumed in the building sector in Turkey until 2030 is estimated by using Artificial Neural Networks (ANN) method in Matlab program. It is calculated that Turkey will consume 37.868 thousand Tons of Oil Equivalent (TOE) of energy in the building sector in 2030. In addition, the importance of energy efficiency studies in the building sector is emphasized.

Estimation of zooplankton density with artificial neural networks (a new statistical approach) method, Elazığ-Türkiye

This study was carried out to predict the zooplankton density in the Cip reservoir (Elazığ) with an artificial neural network, using some water quality parameters. The plankton samples were collected monthly from Cip Reservoir in 2021- 2022, using a standard plankton net from three stations. Water temperature, dissolved oxygen, pH, electrical conductivity, secchi disk, alkalinity, total nitrogen and total phosphorus were measured. The actual values of zooplankton density and results obtained from the artificial neural networks were compared. Mean absolute percent error (MAPE) values were calculated with actual values and ANNs values. ANNs values were determined to be close to the real data. MAPE percentage value at the first station was determined as 1.143 for Rotifer, 0.118 for Cladocera, and 0.141 for Copepoda. The MAPE percentage value at the second station was determined as 0.941 for Rotifer, 0.377 for Cladocera, and 0.185 for Copepoda. The MAPE percentage value at the third station was determined as 0.342 for Rotifer, 0.557 for Cladocera, and 0.301 for Copepoda. In the present study, it has been seen that artificial neural networks with a learning feature are successful in predicting zooplankton densities in an aquatic environment. It can be concluded from the study that ANNs are a powerful tool for understanding their relationships with the environment.

Analysis of the Severity of Accidents on Rural Roads Using Statistical and Artificial Neural Network Methods

This study assesses the relationship that existed between various variables and their subvariables on rural roads in Qom, Iran, using statistical analysis and calculates the relationship between the considered factors and accident severity. A logit model was applied to determine the factors affecting the severity of accidents. In addition, two artificial neural network (ANN) models were developed using two kinds of learning methods to train neurons to select the best result. The results of modeling and analysis of accidents using various techniques revealed that each technique, depending on its purpose, examined the severity of accidents from a different point of view and represented various outcomes. Finally, the performance of the proposed models was validated utilizing other mathematical models. As a result, putting the output results together, the best measures can be suggested to increase the safety of people on rural roads. The outcomes of this study may aid these service providers in strategic planning and policy framework.

Land cover change analysis of buffer areas in New Capital City of Nusantara, Indonesia: A cellular automata approach on satellite imageries data

The proposed plan to move Indonesia's capital city to the New Capital City of Nusantara in East Kalimantan Province undoubtedly requires careful efforts to ensure food supply for the population. Population migration to the new capital may pose a food security challenge. To address this fundamental issue, one of the most crucial approaches is to establish buffer areas that can support the food needs of the new capital. The currently existing official Area Sampling Frame survey conducted by the government to assess food vulnerability faced several limitations, including weather conditions, field terrain variations, and high cost. In this study, we propose the utilization of remote sensing satellite imagery data in buffer areas to analyze changes and predict future land cover, which can provide valuable data for assessing food availability. We investigate the integration of a Cellular Automata method with the two most popular analytical methods of classical Logistic Regression and data-driven Artificial Neural Networks, known as CA-LR and CA-ANN, to identify and map land cover changes in the new capital buffer zones. Our findings reveal that both combined methods, CA-LR and CA-ANN, yield fairly promising results, with correctness and kappa statistic values exceeding 80%. Prediction results indicate that buffer areas are predominantly covered by trees, while built-up areas are still limited. The flooded vegetation cover, including rice fields, is predicted to decrease by 2024. This should be a matter of concern for stakeholders, considering the construction of the new capital city is still ongoing and the number of migrants is expected to keep rising.

Efficiency Analysis of the Photovoltaic Shading and Vertical Farming System by Employing the Artificial Neural Network (ANN) Method

Productive facades, consisting of photovoltaic shading and vertical farming systems, have been proposed as a means to improve the thermal and visual status of residential buildings while also maintaining energy performance and providing vegetables. However, how to quickly and accurately predict electricity and vegetable output during the numerous influencing architectural and environmental factors is one of the key issues in the early stages of design, and few studies have investigated the impact of such structures on both indoor environmental qualities and production performance. In this paper, we present a novel prediction method that uses experimental data to train and test an artificial neural network (ANN). The results indicated that using the Bipolar Sigmoid activation function to process the experimental data input to the artificial neuron network gives more accurate predicted results both in the yield of photovoltaic shading and vertical farming systems. In addition, this prediction method was applied to a typical high-rise residential building in Singapore to assess the self-sufficiency potential of high-rise residential buildings integrated with productive facades. The results indicated that the upper part of the building can meet 20.0–23.1% of the annual household electricity demand of a family of four in a four-room residential unit in Singapore and almost the entire year’s vegetable demand, while the middle part can meet 18.4–21.2% and 89.1%, respectively. The results demonstrated the importance of a productive facade in reducing energy demand, enhancing food security, and improving indoor visual and thermal comfort.

Comparison of two different artificial neural network models for prediction of soil penetration resistance

A time-varying, nonlinear soil-plant system contains many unknown elements that can be quantified based on analytical methodologies. Artificial Neural Networks (ANNs) are a widely used mathematical computing, modelling, and predicting method that estimates unknown values of variables from known values of others. This paper aims to simulate relationship between soil moisture, bulk density, porosity ratio, depth, and penetration resistance and to estimate soil penetration resistance with the help of ANNs. For this aim, the Generalized Regression Neural network (GRNN) and Radial Basis Function (RBF) models were developed and compared for the estimation of soil penetration resistance values in MATLAB. A dataset of 153 samples was collected from experimental field. From the 153 data, 102 data (33%) were selected for training and the remaining 51 data (67%) were used for testing. The estimation process was implemented 10 replications using randomly selected testing and training data. Mean Squared Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE) were used to evaluate estimation accuracy on the developed ANN methods. Based on MSE, RMSE, MAE and Standard Deviation (SD), statistical results showed that the GRNN modelling presented better results than the RBF model in predicting soil penetration resistance success.

Towards estimating the thermal properties of carbon allotropes and their derivatives: Hybridization between the artificial neural network method and the experimental design approach

Carbon allotropes and their derivatives have attracted the interest of many researchers owing to their ability to conduct heat. In this study, a multi-layer perception network (MLP) was developed to predict its thermal transport properties. Usually, trial and error are the appropriate solutions to find the number of hidden layers, the number of neurons, the activation, and the transfer function. To overcome this problem, we used the Surface Response Methodology (RSM), which identifies the influence of each parameter on the response and ends up finding a robust model of Artificial Neural Networks (ANN). The study demonstrated that hybridization between (ANN) and (RSM) methods is an efficient and powerful method for modeling the thermal conductivity of carbon allotropes; it allows to shorten the time to test different combinations. Notably, the study achieved exceptional results with a correlation coefficient surpassing 99.89%, a coefficient of determination exceeding 99.78%, and minimal RMSE (Root Mean Square Error).

การพยากรณ์ยอดขายและความต้องการสินค้าในอุตสาหกรรมผลิตอุปกรณ์ไฟฟ้า กรณีศึกษา บริษัทผลิตอุปกรณ์ไฟฟ้าแรงสูง

The purpose of this study is to find the product demand forecasting model from the artificial neural network method in a case study of a high-voltage equipment manufacturing company. And reduce fines for late delivery of goods by using the Matlab program. The forecasting model is established by using time series neural network method, and the nonlinear autoregressive neural network (NARX) function model is established by combining with external input data. Then the forecast deviation is the average deviation percentage error (MAPE) uses the least dynamics model. By using data that collects product sales from 2019 – 2022 total 48 set of data are divided into 2 data sets used in the study. by the first dataset is learning 36 datasets and the second dataset is learning 12 datasets. The model of this study was defined with the number of hidden layer nodes as 2, 6 and 10. The time delay was 2 and 3. Forecasting models are combined into 6 models. The best model result of each item from NARX forecast through the learning process Levenberg-Marquardt (trainlm), including models 1, 4, 5, 4, 3, 3, and 2 sequential The results show the least average absolute percentage error at 8.32, 5.85, 2.61, 3.39, 2.91, 2.19 and 7.76 and the results of the forecasting model were analyzed to reduce fines from selling all 7 items Able to reduce penalties for late delivery of 4 items showing the percentage of difference in cost reduction equal to 0.28% 0.24% 0.32% and 0.22%. It can be concluded that the prediction of the neural network inverted neural network model It can be used to forecast product sales and product demand as well. And reduce the cost of late delivery fines for better results.

Predictions on Flexible CdTe Solar Cell Performances by Artificial Neural Networks

CdTe solar cells on ultra-thin glass substrates are light and flexible. Flexible cells are widely preferred modules in technological fields. The flexibility of these cells enables them to cope with deformations. The efficiency of these has reached 19%. In this work, we used artificial neural network (ANN) method for the determination the performance of flexible CdTe solar cells despite bending and time. The performances of the solar cell before and after bending have been predicted. According to the results from the ANN calculations using the experimental data in the literature, MSE values of ANN estimates range from 0.06% to 0.28%.

Experimental study and predictive modelling of damping ratio in hybrid polymer concrete

The improved damping capacity of concrete materials is crucial for many applications, especially the ones exposed to dynamic external forces. This study explores the optimal circumstances of copolymer composite binder, ratio of fine to coarse aggregate, curing time, size and content of tyre crumb rubber on enhancing damping ratio of hybrid polymer concrete. Owing to the limitation of conventional models and experimental methods in effectively handling the multiple and complicated relationship between the input parameters of the hybrid polymer concrete and its damping capacity, two robust machine leaning models including extreme gradient boosting (XGB), and artificial neural network (ANN) algorithms were developed to predict the values of damping ratio from an experimental database of 196 samples. Multiple linear regression (MLR) was employed in this study as a benchmark for comparing with the ANN and XGB methods. The results indicate that the considered algorithms yield accurate models for predicting the damping ratio of HPC composites. However, ANN and XGB outperform MLR with an impressive R-square of 0.985 and 0.981, respectively versus 0.875. Analysing the importance of input features using the XGB algorithm also reveals that the curing time has the highest impact on predicting the damping ratio. The volume fractions of resin matrix and crumb rubber are the next most important features. The crumb rubber size, within the studied range, appears to be the least impactful one.

Artificial Neural Network (ANN) Validation Research: Free Vibration Analysis of Functionally Graded Beam via Higher-Order Shear Deformation Theory and Artificial Neural Network Method

Presented herein is the free vibration analysis of functionally graded beams (FGMs) via higher-order shear deformation theory and an artificial neural network method (ANN). The transverse displacement (w) is expressed as bending (wb) and shear (ws) components to define the deformation of the beam. The higher-order variation of the transverse shear strains is accounted for through the thickness direction of the FGM beam, and satisfies boundary conditions. The governing equations are derived with the help of Hamilton’s principle. Non-dimensional frequencies are obtained using Navier’s solution. To validate and enrich the proposed research, an artificial neural network method (ANN) was developed in order to predict the dimensionless frequencies. Material properties and previous studies were used to generate the ANN dataset. The obtained frequency values from the analytical solution and ANN method were compared and discussed with respect to the mean error. In conclusion, the solutions were demonstrated for various deformation theories, and all of the results were thereupon tabularized and visualized using 2D and 3D plots.

Optimization of Machine Learning Classification Analysis of Malnutrition Cases in Children

Malnutrition is one of the problems that occurs in children due to a lack of nutritional intake. Indonesia contributed 36%, making it the fifth country with the largest cases of malnutrition in the world. On this basis, a solution is needed to reduce the growth rate of malnutrition cases. This research aims to carry out classification analysis to determine nutritional status by optimizing machine learning (ML) performance. The ML classification analysis process will later utilize the performance of the artificial neural network (ANN) method with the Multilayer Perceptron (MLP) algorithm. ML performance can be optimized using the Pearson’s correlation (PC) method to produce optimal classification analysis patterns. This research data set uses child nutrition case data from 576 patients sourced from the M. Djamil Padang Province Regional General Hospital (RSUP). The data set is divided into 417 training data and 159 test data. On the basis of the tests that have been carried out, the performance of the PC method can provide precise and accurate analysis patterns. This analysis pattern has also been able to provide a fairly good level of accuracy, namely 95%. Not only that, this research is also able to present analysis patterns with the best ANN architectural model in classifying nutritional status. Based on the overall results, this research can be used as an alternative solution to the treatment of nutritional problems in children.

Toward digitalization strategic perspective in the European food industry: non-linear nexuses analysis

Purpose Organizations deal with digital technologies to achieve their strategic goals. The shift toward digitization is a major challenge because it requires companies to create a digital outlook that influences organizational design. As a result, investigation of institutional theory and entrepreneurial orientation theory in the European food industry has become the focus of research in recent times.Design/methodology/approach To this end, data were collected from 83 companies related to the food industry in the European context. By applying a hybrid phase of the partial least squares structural equation modeling (PLS-SEM) and artificial neural network (ANN) methods, this study captured the causal–non-linear relationships among the study constructs.Findings The findings revealed that the variables of institutional theory and entrepreneurial orientation theory affect the adoption of the digital strategy. There is also a dual interaction role for e-business capabilities and digital transformation. The results of non-linear relationships confirmed that digital strategy adoption is highly influenced by digital transformation, followed by risk-taking, digital leadership, e-business capabilities, organizational agility, proactiveness and innovativeness.Research limitations/implications The authors provided significant implications for practitioners and academics about the most influential determinants of digital strategy – businesses must move swiftly toward digitization across its various units to achieve their objectives. An organization’s leadership must realize that equipping the employees with necessary skills is the first step toward digitalization.Originality/value The current study underscores the digital strategy, which is usually an overlooked area of investigation, in the food industry. The study identifies some important predictors of digital strategy adoption with the interaction’s role of digital transformation and e-business capabilities. Such relationships have been rarely discussed. In addition, the adoption of a hybrid SEM-AAN approach makes the study an original one.

BPANN modified constitutive descriptions for flow behavior and softening mechanism in the α+β phase region of Ti-55511 alloy with equiaxed microstructure

The deformation behavior of Ti-55511 alloy was investigated by hot compression tests conducted at deformation temperatures of 993∼1113 K and strain rates of 0.001–1 s−1. To eliminate the effects of friction and adiabatic temperature rise on the flow stresses, the flow curves were corrected. Subsequently, a strain compensated Arrhenius (SCA) model was developed and further modified using the back propagation artificial neural network (BPANN) method. The modified model exhibits significant improvements compared to the SCA model. The average absolute relative error (AARE) decreases from 4.895 % to 0.756 %, while the relative error (δ) range decreases from −14.75 %–17.99 % to −4.78 %–8.01 %. Moreover, cross-validation also confirms the modified model has higher generalization ability and prediction accuracy than the SCA model. Microstructural characterization reveals dynamic recrystallization (DRX) of the α phase and dynamic recovery (DRV) of the β phase are the main mechanisms responsible for softening, and their volume fractions increase with decreasing strain rate and increasing deformation temperature. The equiaxed α phase undergoes elongation first, followed by gradual migration and transformation of grain boundaries. Finally, the β phase wedges into the α phase until the α phase separates. The continuous dynamic recrystallization (CDRX) of β phase accompanies its DRV, which is mainly characterised by progressive sub-grain rotation. The α grains aggregated by dislocations promote CDRX of the β phase.

Performance Evaluation of Artificial Neural Network Methods Based on Block Machine Learning Classification

Performance Evaluation of Artificial Neural Network Methods Based on Block Machine Learning Classification

Prediction of Mycotoxin Response of DNA-Wrapped Nanotube Sensor with Machine Learning

DNA-wrapped single-walled carbon nanotubes (SWCNTs) have demonstrated great versatility in their use as optical sensors. SWCNTs emit a near-infrared fluorescence that is responsive to even the slightest changes in the nanotube environment, enabling sensors that can respond to single-molecule fluctuations within the vicinity of their surfaces. The fluorescence response and surface interactions of these sensors are determined by the DNA wrapping sequence. However, the lack of information on the relationship between the DNA sequence and its effect on the SWCNT fluorescence remains a bottleneck for designing sequences that are specific to analytes of interest. We have recently demonstrated the use of directed volution to control the fluorescence response of SWCNTs through DNA design. Iterative cycles of DNA mutation, screening, and selection allowed us to evolve sequences that yield DNA-wrapped SWCNT sensors with a desired fluorescence response to mycotoxins.In this work, we apply the screening results of the DNA libraries used in this approach to train machine learning (ML) algorithms. Artificial neural network (ANN) and support vector machine (SVM) methods were used to predict the response of ssDNA-SWCNT sensors to a specific mycotoxin. The reliability of these models was further assessed through cross-validation. The ANN and SVM models with cross-validation were able to accurately classify the various DNA sequences as yielding either a high or low fluorescence response with an accuracy of 73 and 81%, respectively. The models were further used to predict the performance of alternative DNA sequences outside the initial training dataset, using the Hierarchy and k-means++ clustering methods to evaluate the similarity and dissimilarity of each DNA sequence. Compared to the SVM model, the ANN model showed an improved ability to predict high responses for dissimilar DNA sequences. We further applied a combinatorial approach based on SVM and ANN models to design new DNA sequences for improving sensor performance. The success of this approach was validated experimentally, demonstrating the rational design of improved sensors with 95% prediction accuracy.The application of ML algorithms to directed evolution libraries of DNA thus allows one to accurately map the performances of these sensors within a particular sequence space. The computational success of this mapping provides a framework for replacing current empirical approaches with the rational design of DNA sequences for SWCNT sensing. Keywords: Nano Biosensor, DNA-SWCNT, Mycotoxin, Machine learning, Directed evolution method.

Immediate Impact of the COVID-19 Pandemic on Heart and Kidney Transplantation andtheRecovery Trends in 30 Developed andLess-Developed Countries.

BACKGROUND The COVID-19 pandemic had multifaceted and disproportionate impacts on various countries. We investigated the decline of heart and kidney transplantation in 2020 and recovery trends in 2020, 2021, and 2022 in 30 developed and developing countries, considering COVID-19 incidence and mortality and pandemic-time economic variables. MATERIAL AND METHODS Data were obtained from reliable open databases. Nations were grouped by hierarchical cluster analysis into high-gross domestic product (GDP), mid-GDP, and low-GDP countries. Expected transplant numbers for 2020 to 2021 were estimated by the artificial neural network method using data from 2015 to 2019. Effect size and its inference were determined through the Hodges-Lemann estimate and Wilcoxon signed-rank test, respectively. The possible disproportionate effect was estimated by the Jonckheere-Tersptra test. Associations between transplantation and economic variables, COVID-19 caseload, and mortality were examined using Kendall rank correlation analysis. RESULTS All nations experienced a decline in 2020 and some real recovery in 2020 to 2022. For high-GDP countries, decline was insignificant and recovery was marginal; for mid-GDP countries, decline was significant for heart and deceased kidneys and recovery was modest; for low-GDP countries, decline was significant for heart, live kidneys, and deceased kidneys and recovery was marginal. The low-GDP countries were disproportionally negatively impacted, although the associations between the impact and economic variables, COVID-19 incidence, and COVID-19 mortality were statistically insignificant. CONCLUSIONS More inclusive studies of socioeconomic and cultural factors that affected the impact of the COVID-19 pandemic in different countries can be useful for better preparedness and reducing disruption in healthcare in future global pandemics.

Estimation reference crop evapotranspiration (ET0) using artificial intelligence model in an arid climate with external data

Water resource management and crop growth control require the calculation of reference evapotranspiration (ET0), but meteorological data can often be incomplete, necessitating models with minimal inputs. This study was conducted in Iran’s arid synoptic stations of Sirjan and Kerman, where data scarcity is severe. Penman–Monteith FAO-56 was selected as the target data for modeling artificial neural network (ANN), fuzzy neural adaptive inference system (ANFIS), and ANN-gray wolf optimization (ANN-GWO). The performance of these models was evaluated using an input dataset consisting of the current station’s minimum and maximum temperatures, ET0, and the wind speed of the nearby station (external data) in three different combinations. The models’ accuracy was assessed using two widely used criteria: root mean square error (RMSE) and coefficient of determination (R2), as well as the empirical Hargreaves equation. In the absence of climatic data, the ANFIS, ANN, and ANN-GWO methods using minimum and maximum temperatures, which are relatively easier to estimate, outperformed the empirical Hargreaves equation method in both stations. The results demonstrate that the ANFIS method performed better than ANN and ANN-GWO in all three input combinations. All three methods showed improvement when external data (wind speed and ET0 of the adjacent station) were used. Ultimately, the ANFIS method using minimum and maximum temperatures and the adjacent station’s ET0 in Kerman and Sirjan yielded the best results, with an RMSE of 0.33 and 0.36, respectively.

Comparison of Road Luminance Data Estimated by Fuzzy Logic and ANN, A Case study of Kocaeli Sakıp Sabancı Caddesi

Road lighting constitutes the most important part of city lighting. While correct road lighting solves the chaos in transportation, it also paves the way for smooth urban traffic flow. It also allows people to have a comfortable and safe journey. In order to reduce accidents in the evening hours, safe driving and visual comfort must be provided to drivers. Just like inadequate road lighting, excessive lighting also causes accidents and waste of electrical energy. Good visibility of objects on the road and visual comfort for drivers is possible by ensuring an equal brightness distribution on the road surface. In this article, a prediction model was created with Artificial Neural Network (ANN) and Fuzzy Logic (FL) methods to find the luminance values of the road in Sakıp Sabancı street in Kocaeli province. When the predictions obtained by ANN and FL were compared, the results were seen to be accurate and compatible. When the estimation results obtained as a result of the application were examined, it was observed that the estimation of luminance values in good and qualified road lighting could be achieved with these methods.

Identification of Bioethanol Quality for Motorcycle Fuel

The availability of crude oil as a raw material for vehicle fuel is dwindling and limited in nature. One of the renewable energies worth developing is bioethanol, which is one of the alternative fuels that can be used as a biofuel and can be processed from plants containing starch and glucose. In this research, the entire bioethanol identification system in a sugar cane drip distillation apparatus was examined. The distillation process using MQ3 and MQ135 sensors resulted in an alcohol percentage of 42% and 46%. The maximum temperature measured by a thermocouple during distillation was 88°C, while the minimum temperature recorded was 87°C. This study utilized a backpropagation artificial neural network method to identify the detected bioethanol. The architecture of the artificial neural network included 2 input nodes, 4 neuron nodes, and 2 output nodes. The training and testing results showed that the formed backpropagation was able to identify and differentiate the detection of bioethanol according to the given inputs with a success rate of 88.86% for detected bioethanol and 94.86% for undetected bioethanol..