Classification of Plant Types based on Leaf Image using the Artificial Neural Network Method

Detecting the different blends of diesel and biodiesel fuels using electronic nose machine coupled ANN and RSM methods

Heat rate of a combined cycle power plant (CCPP) is a parameter that is typically used to assess how efficient a power plant is. In this paper, the CCPP heat rate was predicted using an artificial neural network (ANN) method to support maintenance people in monitoring the efficiency of the CCPP. The ANN method used fuel gas heat input (P1), CO2 percentage (P2), and power output (P3) as input parameters. Approximately 4322 actual operation data are generated from the digital control system (DCS) in a year. These data were used for ANN training and prediction. Seven parameter variations were developed to find the best parameter variation to predict heat rate. The model with one input parameter predicted heat rate with regression R2 values of 0.925, 0.005, and 0.995 for P1, P2, and P3. Combining two parameters as inputs increased accuracy with regression R2 values of 0.970, 0.994, and 0.984 for P1 + P2, P1 + P3, and P2 + P3, respectively. The ANN model that utilized three parameters as input data had the best prediction heat rate data with a regression R2 value of 0.995.

In this study, seismic events in Kula district (Manisa, Turkey) and its vicinity have been investigated and then natural and artificial seismic activities are discriminated. Total of 77 digital vertical component velocity seismograms of seismic activities with M L ≤3.5 magnitude from seismic activity catalogs between 2009 to 2014 recorded by Manisa Kula (KULA) broadband station operated by Bogazici University, Kandilli Observatory and Earthquake Resarch Institute Regional Earthquake- Tsunami Monitoring Center (RETMC) were used in this study. The maximum S-wave and maximum P-wave amplitude ratio (Ratio) of vertical component velocity seismograms and power ratio for (1 and 12 sec.) (Complexity-C) and total signal duration (Duration) of the waveform were calculated. The earthquakes and the quarry blasts have been discriminated using linear discriminant function (LDF) and Back Propagation-Feed Forward Neural Networks (BPNNs) that is one of the learning algorithms at the artificial neural networks (ANNs) methods taking correlation between these parameters into consideration. 39 (51%) of the 77 seismic activities were identified as quarry blasts and 38 (49%) of them as earthquakes LDF and ANNs methods have been applied together for the first time for Ratio-C, Ratio-logS and Ratio-duration parameter pairs with the data of Manisa and surroundings, and earthquakes and quarry blasts have been distinguished from each other. LDF and ANNs methods were compared for each pair of parameters. Both of two methods are successful but the ANNs method has higher accuracy percentage values than LDF method when there is sufficient number of data. The accuracy percentages are different for a pair of Ratio versus C, for a pair of Ratio versus logS and for a pair of Ratio versus duration, respectively.

Diesel oil is known to be one of the major petroleum products that can pollute water and soil. Soil pollution caused by petroleum hydrocarbons has substantially impacted the environment, especially in the Middle East. In this study, modeling and optimization of hexadecane removal from soil was performed using two pure cultures of Acinetobacter and Acromobacter and consortium culture of both bacterial species using artificial neural network (ANN) method. Then the best ANN structure was proposed based on mean square error (MSE) as well as correlation coefficient (R) for pure cultures of Acinetobacter and Acromobacter as well as their consortium. The results showed that the correlations between the actual data and the data predicted by ANN (R2) in Acromobacter, Acinetobacter and consortium of both cultures were 0.50, 0.47 and 0.63, respectively. Despite the low correlation between the experimental data and the data predicted by the ANN, the correlation coefficient and the precision of ANN for the consortium was higher. As a result, ANN had desirable precision to predict hexadecan removal by the cobsertium culture of Ochromobater and Acintobacter.

Background: The optimal management of hospitals as the most important centers for providing health services has always been the focus of decision makers and policymakers. The present study aimed to evaluate the technical efficiency of hospitals affiliated to Ardebil University of Medical Sciences using Artificial Neural Network (ANN) method. Methods: The present descriptive, applied, and cross-sectional study was conducted in 2017. The study population consisted of 8 educational and therapeutic centers affiliated to Ardabil University of Medical Sciences and Health Services. Census sampling method was applied to select the participants and researcher-made checklists were used to collect the data. The artificial neural network method was also used to investigate the study objectives. The applied network was a multi-layer prospectron with a Levenberg Marquardt optimization function, a sigmoid tangent transfer function in the middle layer, and a linear function in the output layer. For processing and preparing data, Excel spreadsheet was used and Matlab software were used for analyzing and designing the network. Results: The results showed that the average technical efficiency of the studied hospitals was 0.75. Ardabil Bou Ali, Khalkhal Imam Khomeini, and Meshkinshahr Valiasr hospitals had the highest technical efficiency (1.00). With regard to the hospital bed occupancy rate, Germi Velayat hospital had the lowest efficiency rate (0.75). Furthermore, Parsabad Imam Khomeini hospital had the lowest efficiency in the number of discharged healthy patients (0.69) in 2017. Finally, the improvement space of each variable was calculated to achieve the total technical efficiency. Conclusion: According to the results, one of the factors that decreases the technical efficiency of the educational centers is the low rate of hospital beds' utilization and occupancy rate. In other words, lack of full-capacity operation by the hospital centers and the low rates of active hospital bed imposed costs and consequently reduced the hospitals' performance.

The relativistic mean field (RMF) model with a small number of adjusted parameters has been used successfully in the last thirty years for predictions of various ground-state nuclear properties of nuclei. In this model, Dirac and Klein–Gordon like equations obtained from application of variation principle on phenomenological Lagrangian density are solved iteratively for calculations of nuclear properties of nuclei. For this purpose, parameters such as masses of considered mesons, nucleon–meson coupling constants, and self-couplings of mesons are needed and they are fitted from experimental data. Some parameter sets for RMF model introduced to correct predictions of nuclear properties of nuclei cover nuclidic chart. Besides Artificial Neural Network (ANN) method is used successfully in many field of science as in nuclear physics. ANN is known as a very powerful tool that are used when standard techniques fail to estimate the correlation between the variables. In the present study, ANN method has been employed to check its understanding capability of relations between RMF model parameters and their predictions on the ground-state binding energies of some spherical nuclei. Understanding capability of ANN method for these relations of considered nuclei has been found well. Based on this success, new non-linear parameter set for RMF model called DEFNE by us have been produced by using ANN method. Furthermore, predictions of RMF model with DEFNE parameter set for ground-state binding energies and charge radii of nuclei cover nuclidic chart have been found as in agreement with the available experimental data.

In this paper to make use of complementary potential in the mapping of LULC spatial data is acquired from LandSat 8 OLI sensor images are taken in 2019. They have been rectified, enhanced and then classified according to Random forest (RF) and artificial neural network (ANN) methods. Optical remote sensing images have been used to get information on the status of LULC classification, and extraction details. The classification of both satellite image types is used to extract features and to analyse LULC of the study area. The results of the classification showed that the artificial neural network method outperforms the random forest method. The required image processing has been made for Optical Remote Sensing Data to be used in LULC mapping, include the geometric correction, Image Enhancements, The overall accuracy when using the ANN methods 0.91 and the kappa accuracy was found 0.89 for the training data set. While the overall accuracy and the kappa accuracy of the test dataset were found 0.89 and 0.87 respectively.

In Masamba City, the use of electrical energy is influenced by the welfare of the population. The higher the level of welfare of the population, the greater the use of electricity. So, power plants must be ready to supply electricity load demand when experiencing sudden fluctuations in electricity demand. One way that needs to be done to deal with this is structured planning and forecasting. This study aims to analyze the results of the error value of the ability to forecast electrical loads using the Artificial Neural Network method. In this study, the method used is a quantitative approach where data collection is done by literature study, observation, and direct interviews at the UP3 Palopo office. The Artificial Neural Network (ANN) method is programmed using MATLAB software. Where forecasting using ANN obtained the smallest error value of 0.83% with an estimated power generated by ANN of 35,991 MVA on day 2 and for the largest error value on day 7 with an error value of 8.33% with estimated power generated by ANN of 36.0836 MVA.

Neural network method for solving parabolic two-temperature microscale heat conduction in double-layered thin films exposed to ultrashort-pulsed lasers

The simulation of wave propagation, such as soliton propagation, based on the Rosenau-KdV-RLW equation on unbounded domains requires a bounded computational domain. Therefore, a special boundary treatment, such as an absorbing boundary condition (ABC) or a perfectly matched layer (PML), is necessary to minimize the reflections of outgoing waves at the boundary, preventing interference with the simulation’s accuracy. However, the presence of higher-order partial derivatives, such as uxxt and uxxxxt in the Rosenau-KdV-RLW equation, raises challenges in deriving accurate artificial boundary conditions. To address this issue, we propose an artificial neural network (ANN) method that enables soliton propagation through the computational domain without imposing artificial boundary conditions. This method randomly selects training points from the bounded computational space-time domain, and the loss function is designed based solely on the initial conditions and the Rosenau-KdV-RLW equation itself, without any boundary conditions. We analyze the convergence of the ANN solution theoretically. This new ANN method is tested in three examples. The results indicate that the present ANN method effectively simulates soliton propagation based on the Rosenau-KdV-RLW equation in unbounded domains or over extended periods.

Comparative studies between response surface methodology (RSM) and artificial neural network (ANN) methods to find the effects of electrospinning parameters on the porosity of nanofiber mats is described. The four important electrospinning parameters studied included solution concentration (wt.%), applied voltage (kV), spinning distance (cm) and volume flow rate (mL/h). It was found that the applied voltage and solution concentration are the two critical parameters affecting the porosity of the nanofiber mats. The two approaches were compared for their modeling and optimization capabilities with the modeling capability of RSM showing superiority over ANN, having comparatively lower values of errors. The mean relative error for the RSM and ANN models were 1.97% and 2.62% and the root mean square errors (RMSE) were 1.50 and 1.95, respectively. The superiority of the RSM-based approach is due to its high prediction accuracy and the ability to compute the combined effects of the electrospinning factors on the porosity of the nanofiber mats.

Reference crop evapotranspiration (ETo) is an important component of the hydrological cycle that is used for water resource planning, irrigation, and agricultural management, as well as in other hydrological processes. The aim of this study was to estimate the ETo based on limited meteorological data using an artificial neural network (ANN) method. The daily data of minimum temperature (Tmin), maximum temperature (Tmax), mean temperature (Tmean), solar radiation (SR), humidity (H), wind speed (WS), sunshine hours (Ssh), maximum global radiation (gradmax), minimum global radiation (gradmin), day length, and ETo data were obtained over the long-term period from 1969 to 2019. The analysed data were divided into two parts from 1969 to 2007 and from 2008 to 2019 for model training and testing, respectively. The optimal ANN for forecasting ETo included Tmax, Tmin, H, and SR at hidden layers (4, 3); gradmin, SR, and WS at (6, 4); SR, day length, Ssh, and Tmean at (3, 2); all collected parameters at hidden layer (5, 4). The results showed different alternative methods for estimation of ETo in case of a lack of climate data with high performance. Models using ANN can help promote the decision-making for water managers, designers, and development planners.

This paper employs an artificial neural network (ANN) method to achieve an approximate solution [Formula: see text] of the one-dimensional fractional Rayleigh–Stokes equation with boundary and initial conditions. ANN is a machine learning branch that aims to minimize the loss function, which closely approaches the solution of mathematics equations. By combining the ANN output with the initial and boundary conditions, we can achieve an adequate approximation for this fractional equation. The study utilizes a multi-layer feed-forward neural network algorithm and error backpropagation method with unsupervised learning to minimize the loss function and parameter alteration (weights and biases). The accuracy and effectiveness of the ANN technique were demonstrated through several examples. This method can approximate solutions of the Rayleigh–Stokes equation for more than 10,000 points with considerable accuracy in a short time. A comparison of the results achieved with other existing techniques indicates that the ANN method provides a more appropriate approximation.

ABSTRACTThis paper proposes a multilayer artificial neural network (ANN) method to predict the probability of default (PD) within a survival analysis framework. The ANN method captures hidden interconnections among covariates that influence PD, potentially leading to improved predictive performance compared to both logit and skewed logit models. To assess the impact of covariates on PD, we introduce a generalized covariate method that accounts for compositional effects among covariates and employ stochastic dominance analysis to rank the importance of covariate effects across both the ANN and logit model approaches. Applying the ANN method to a large dataset of small business loans reveals prediction gains over the logit models. These improvements are evident for short‐term prediction horizons and in reducing type I misclassification errors in the identification of loan defaults, an aspect crucial for effective credit risk management. Regarding the generalized covariate effects, our results suggest that behavior‐related covariates exert the strongest influence on PD. Moreover, we demonstrate that the ANN structure stochastically dominates the logit models for the majority of the covariates examined.

Evaluation of the relationship between the physical properties and capillary water absorption values of building stones by regression analysis and artificial neural networks