Artificial Neural Network Modeling Technique Research Articles

Kinetic and artificial neural network modeling techniques to predict the drying kinetics of Mentha spicata L.

This study presented both the empirical and artificial neural network (ANN) approaches to estimate the moisture content of Mentha spicata. Two different types of drying methods (in shade and in oven (35 and 50°C)) were used to investigate the drying kinetics of the Mentha spicata samples. The effects of drying methods on effective diffusion coefficient, moisture ratio (MR), drying rate, and activation energy were investigated. Moreover, six different thin layer drying models (Page, Diffusion approach, Newton, Modified Henderson, Henderson and Pabis and Pabis and Midilli) and an ANN with feed forward structure were used to define the drying kinetics of these samples. In order to estimate the kinetic model parameters, sequential quadratic programming (SQP) was used. Model performances were evaluated based on the coefficient of determination (R²), root mean square error (RMSE) and mean absolute percentage error (MAPE%) values. In the kinetic part of the modeling study, the Midilli model provided better results than the others. However, the ANN had the best results when a total assessment was made. The effective diffusion coefficient values were found in the range between 1.31 × 10–¹² and 4.43 × 10–¹² m²/s. The activation energy was obtained as 44.31 kJ/kmol. The R², MAPE%, and RMSE values for the ANN test data were 1.00, 0.2257, and 5.9447 × 10⁻⁴, respectively. In the future, different modeling approaches will be applied to describe this drying process. PRACTICAL APPLICATIONS: Drying is a process where heat transfer and mass transfer take place together. Modeling is an innovative approach used in evaluation of experimental data and has increasing popularity in recent years. ANNs are a powerful data‐driven method, and they have a very broad area of usage from medicine to engineering issues. Empirical models are another approach for describing experimental data. In this study, these two modeling approaches were used to obtain the MR. Humidity is a condition that needs to be checked in food safety and protection. Therefore, it is very important to ensure control with robust modeling techniques. In this study, the developed ANN model had a high R² value (R² = 1.00). This indicated that it may be used successfully in real applications.

Evaluation of potential modeling approaches for Scots pine stem diameter prediction in north-eastern Turkey

Critical forest management decisions are strongly supported by the knowledge of forest ecosystems and their features. There is a growing trend towards using models to predict the most accurate yield projections for forest management. In this research area, features of standing trees, such as diameters along the tree bole, are of much interest. This paper is focused on the construction and evaluation of potential modeling approaches for Scots pine stem diameter prediction in north-eastern Turkey, including (1) fixed-effects, (2) mixed-effects, (3) three and five quantile regression, and (4) an artificial neural network modeling techniques. Nonlinear fixed-effects, mixed-effects, and both quantile regression models were based on a variable exponent taper equation, while the Levenberg-Marquardt algorithm was investigated for the artificial neural network (LMANN) models construction. Parameters of the tested models were calibrated by the use of several stem diameter measurements obtained at relative height positions, with models’ superior performances in diameter predictions when the information of the additional diameter measured at 60% of total height, was included. Evaluation statistics show that both quantile regression and mixed-effects models improved results as compared to fixed-effects model. Overall, LMANN models achieved the best performance with higher accuracy for the Scots pine stem diameter prediction for the whole tree as well as for sections within the tree based on the ten relative height classes. Similarly to previous findings, the results of this study support that the use of mixed-effects modeling increases flexibility and efficiency of taper equations for stem diameter prediction. Moreover, the LMANN modeling approach was evaluated as superior according to its adequacy in predicting diameters along the tree stem. Therefore, the usage of the LMANNs for the tree stem diameter prediction can be a very useful tool in forest management practice and thus worth consideration.

Prediction of benzo[a]pyrene content of smoked sausage using back-propagation artificial neural network.

Benzo[a]pyrene (BaP), a potent mutagen and carcinogen, is reported to be present in processed meat products and, in particular, in smoked meat. However, few methods exist for predictive determination of the BaP content of smoked meats such as sausage. In this study, an artificial neural network (ANN) model based on the back-propagation (BP) algorithm was used to predict the BaP content of smoked sausage. The results showed that the BP network based on the Levenberg-Marquardt algorithm was the best suited for creating a nonlinear map between the input and output parameters. The optimal network structure was 3-7-1 and the learning rate was 0.6. This BP-ANN model allowed for accurate predictions, with the correlation coefficients (R) for the experimentally determined training, validation, test and global data sets being 0.94, 0.96, 0.95 and 0.95 respectively. The validation performance was 0.013, suggesting that the proposed BP-ANN may be used to predictively detect the BaP content of smoked meat products. An effective predictive model was constructed for estimation of the BaP content of smoked sausage using ANN modeling techniques, which shows potential to predict the BaP content in smoked sausage. © 2017 Society of Chemical Industry.

Experimental and ANN modeling investigations of energy traits for rough rice drying

Abstract In the present work, the kinetics of convective deep-bed drying of rough rice was investigated at different conditions. The energy performance of the process was investigated in terms of specific energy consumption as well as energy, drying and thermal efficiencies. Different well-known artificial neural networks (ANNs) were used to predict the traits and the best topologies, transfer functions and training algorithms were determined. Increasing drying air temperature and velocity decreased drying duration while higher relative humidity increased the process time. The results showed that applying higher temperatures together with lower levels of flow rate and relative humidity of drying air improved the energy indices. Artificial neural network modeling technique can be used, as a powerful tool, to predict and determine the energy efficient drying conditions.

Quantification of active ingredients in pharmaceutical suspensions by FT Raman spectroscopy

A method for quantitative determination of nifuroxazide, acetaminophen, ibuprofen and naproxen in commercial oral suspensions containing active compounds in the 1.7–4.0% (w/w) range is described. Partial least squares regression and counter-propagation artificial neural network models were built and validated. The lowest relative standard error of prediction values was found for the strongest scattering nifuroxazide suspensions. Commercial suspensions were quantified with recovery in the 99.3–101.2% and the 98.2–100.2% range, applying partial least squares and artificial neural network modeling techniques, respectively.

Thermal performance analysis of an inclined passive solar still using agricultural drainage water and artificial neural network in arid climate

In this study, a model based on artificial neural network (ANN) was developed in order to predict the thermal performance of an inclined passive solar still in an arid climate, in which the thermal performance of the still was expressed as instantaneous thermal efficiency (ITE). Agricultural drainage water (AWD) was used as a feed for the desalination process, and this is considered a non-conventional water source. Appropriate meteorological variables, viz., ambient air temperature, relative humidity, wind speed, and solar radiation were used alongside the key operational variables, viz., flow rate, temperature, and total dissolved solids of feed water were used as input variables. The results revealed that an ANN with six neurons and a hyperbolic tangent transfer function was the most appropriate model for ITE prediction. Consequently, the optimal ANN model had a 7–6–1 architecture. The results also indicated that the optimal ANN model forecast the ITE accurately, with a mean root mean square error (RMSE) of just 1.933% and a mean coefficient of determination (CD) of 0.949. To create a sensible comparison, a multiple linear regression (MLR) model was also developed. It was found that the ANN model performed better than the MLR model, which displayed a mean RMSE of 4.345% and a mean CD of 0.739. The mean relative errors of forecasted ITE values within the ANN model were mostly in the region of +8% to −6%. One major output of this research is a comprehensive assessment of the ANN modeling technique for the ITE of a solar still, which adds a new perspective to system analysis, design and modeling of the potential productivity of solar stills in terms of the AWD desalination process.

Comparative analysis of various modelling techniques for emission prediction of diesel engine fueled by diesel fuel with nanoparticle additives

In this study, emissions of compression ignition engine fueled by diesel fuel with nanoparticle additives was modeled by regression analysis, artificial neural network (ANN) and adaptive neuro fuzzy inference system (ANFIS) methods. Cetane number (CN) and engine speed (rpm) were selected as input parameters for estimation of carbon monoxide (CO), oxides of nitrogen (NOx), and carbon dioxide (CO2) emissions. The results of estimation techniques were compared with each other and they showed that regression analysis was not accurate enough for prediction. On the other hand, ANN and ANFIS modelling techniques gave more accurate results with respect to regression analysis; linear and non-linear. Especially ANFIS models can be suggested as estimation method with minimum error compared to experimental results.

Neural network approach for predicting solar still production using agricultural drainage as a feedwater source

This study investigates the application of artificial neural network (ANN) for predicting solar still production (MD). Agricultural drainage water (ADW) was desalinated using a solar still. Important meteorological variables: ambient air temperature, relative humidity, wind speed, and solar radiation, together with the operational variables of flow rate, temperature, and total dissolved solids of feedwater, were considered as input parameters for ANN modeling. The output parameter was MD. The results revealed that the ANN model with five neurons and hyperbolic tangent transfer function was the most appropriate for MD prediction based on the minimum measures of error. The optimal ANN model had a 7–5–1 architecture. The ANN model was also compared to multiple linear regression (MLR). The results indicated that, compared to the MLR model, the ANN model provided better prediction results in all modeling stages. The average of the coefficient of determination between the ANN results and the experimental data was more than 0.96. Consequently, the ANN model was shown to have acceptable generalization capability and accuracy. The relative errors of forecasted MD values for the ANN model were mostly in the vicinity of ±10%. These results indicate that ANN can be successfully used in the MD prediction of a solar still desalinating ADW. One major output/contribution of this research involves assessment of the ANN modeling technique during ADW solar desalination, which adds a new perspective to the system analysis, design, and modeling for the potential productivity of a solar still to produce water during the ADW desalination process.

ESTIMATION OF SHEAR STRENGTH PARAMETERS OF LATERITIC SOILS USING ARTIFICIAL NEURAL NETWORK

This research work seeks to develop models for predicting the shear strength parameters (cohesion and angle of friction) of lateritic soils in central and southern areas of Delta State using artificial neural network modeling technique. The application of these models will help reduce cost and time in acquiring geotechnical data needed for both design and construction in the study area. A total of eighty-three (83) soil samples were collected from various locations in Delta State of Nigeria. The geotechnical soil properties were determined in accordance with British Standards. The range of the angle of internal friction and cohesion obtained from the tests are 2 to 43 degrees and 3 to 82 kN/m 2 respectively. The optimum artificial neural network architecture network was found to be 3-9-1, that is three inputs, nine hidden layer nodes, and one output node for cohesion. While, the angle of friction had an optimal network geometry of 3-11-1, that is three inputs, eleven hidden layer nodes, and one output node. The results of the coefficient of determination and root mean square showed that the artificial neural network method outperforms some selected empirical formulae in the prediction of shear strength parameters. http://dx.doi.org/10.4314/njt.v35i2.5

Wind turbine power curve modelling using artificial neural network

Technical improvements over the past decade have increased the size and power output capacity of wind power plants. Small increases in power performance are now financially attractive to owners. For this reason, the need for more accurate evaluations of wind turbine power curves is increasing. New investigations are underway with the main objective of improving the precision of power curve modeling. Due to the non-linear relationship between the power output of a turbine and its primary and derived parameters, Artificial Neural Network (ANN) has proven to be well suited for power curve modelling. It has been shown that a multi-stage modelling techniques using multilayer perceptron with two layers of neurons was able to reduce the level of both the absolute and random error in comparison with IEC methods and other newly developed modelling techniques. This newly developed ANN modeling technique also demonstrated its ability to simultaneously handle more than two parameters. Wind turbine power curves with six parameters have been modelled successfully. The choice of the six parameters is crucial and has been selected amongst more than fifty parameters tested in term of variability in differences between observed and predicted power output. Further input parameters could be added as needed.

Mathematical, Fuzzy Logic and Artificial Neural Network Modeling Techniques to Predict Drying Kinetics of Onion

In this study, the influence of drying temperature (40, 50, 60C) and airflow velocity (2 and 3 m/s) on drying onion was evaluated by a custom designed fluidized bed dryer equipped with a heat pump dehumidifier. A comparative study was performed among nonlinear regression techniques, fuzzy logic and artificial neural networks to estimate the dynamic drying behavior of onion. Among nine mathematical models, approximation of diffusion with R2 = 0.9999 and root mean square error = 0.004157 showed the best fit with experimental data. Fuzzy logic tool in MATLAB with Mamdani model in the form of If–Then rules along with triangular membership function used for simulation, interpolation and obtaining a theoretical increase in experimental moisture ratios were used. Feedforward–backpropagation neural system with application of Levenberg–Marquardt training algorithm, hyperbolic tangent sigmoid transfer function, training cycle of 1,000 epoch and 2-5-1 topology was determined as the best neural model in terms of statistical indices. Practical Applications Forecasting kinetics of food drying by application of precise mathematical and dynamic modeling techniques as well as preparing particular patterns for describing such drying behaviors of diverse food products seem unavoidable for processing factories to optimize the quality of their dried food products and reduce operational costs. For this purpose, in this study, we applied different temperatures and airflow velocities to dry onion by a custom designed fluidized bed dryer equipped with a heat pump dehumidifier, predicting its drying behavior by regression, fuzzy logic and artificial neural network techniques and comparing accuracy of those models. Also, this paper compares prediction patterns of drying onion by fluidized bed drying with those of other food products dried by various drying methods. The results of this study will be helpful for all researchers and producers who want to know more about the nuances of impacts of different fluidized bed drying variables on the patterns of heat and mass transfer in fruits and vegetables.

Models for pedestrian gap acceptance behaviour analysis at unprotected mid-block crosswalks under mixed traffic conditions

Pedestrian safety is an important aspect while crossing the road and it can be explained by pedestrian gap acceptance behaviour. The statistical models such as multiple linear regression (MLR) is often used to model linear relationships between dependent variable (viz., pedestrian gap acceptance behaviour) and independent variables, due to their ability to quantitatively predict the effect of various factors on the dependent variable. However such linear models cannot consider the effect of several variables on the output variable, due to primary assumptions of normality, linear, homoscedasticity and multicollinearity. In this regard, the non-linear models based on the artificial neural network (ANN), which are free from assumptions of linear models, can be easily employed for obtaining the effect of several input variables on the pedestrian accepted gap size. However, researchers have rarely applied ANN modelling technique for predicting the pedestrian gap acceptance behaviour, as the pedestrian gap acceptance behaviour depends on several pedestrian, traffic and vehicular characteristics. The ANN based models would be quite useful in establishing relationship between these factors on the pedestrian gap acceptance behaviour at midblock crosswalks under mixed traffic conditions. In this direction, the present study adopts both MLR as well as ANN with different pedestrian, traffic and vehicular characteristics to assess the significant contributing factors for pedestrians’ gap acceptance behaviour at unprotected mid-block crosswalks under mixed traffic conditions. For this purpose, a video graphic survey was conducted at a six lane divided road at unprotected mid-block crossing in Mumbai, India. The data such as pedestrian (gender and age), vehicular, traffic and pedestrian behavioural characteristics were extracted to model pedestrian accepted gaps. The model results show that pedestrian rolling behaviour has a significant effect on pedestrian accepted gap size. The model results concluded that ANN has a better prediction with possibility to consider the effect of more number of variables on the pedestrian gap acceptance behaviour as compared to the MLR model under mixed traffic conditions. However, the quantification of significant contributing variables on pedestrian accepted gap size is easy by MLR model as compared to the ANN technique. So, both models have their own significant role in pedestrian gap acceptance analysis. The developed models may be useful to enhance the existing mid-block crosswalk facilities or planning new facilities by more accurate prediction of the pedestrian gap acceptance behaviour considering the influence of various factors under mixed traffic conditions.

An off-grid hybrid PV/diesel model as a planning and design tool, incorporating dynamic and ANN modelling techniques

Remote towns and communities are normally without access to the main electrical grid and electricity is normally generated through diesel generators. Diesel fuel costs represent a significant portion of the utilities' expenditure. Solar photovoltaic (PV) integration is an attractive solution reduces fossil fuel dependency for such communities. This study presents an off-grid hybrid PV/diesel model developed using dynamic modelling and artificial neural network (ANN) techniques. Dynamic subsystem models were developed in Simulink and ANN methods were employed for predictive modelling. Utilizing simple climate data (humidity, rain fall, ambient temperature and wind speed) and load profile as model inputs, generator and PV output powers and fuel consumption can be accurately predicted. Experimental data were used for ANN training and model validation. A comparative analysis was conducted between the Simulink model and an existing industrial design tool for a remote community in the Northern Australia. Simulation results showed that the developed model is a viable planning and analytical tool for aiding future off-grid PV-to-diesel system integration applications, with R2 values ranging from 0.92 to 0.99 and mean relative errors below 5%. Lastly, the incorporation of both dynamic and ANN modelling techniques in a single model reduces modelling complexity whilst maintaining its accuracy and ease-of-use.

Development of an ANN–based air pollution forecasting system with explicit knowledge through sensitivity analysis

Little attention is given to applying the artificial neural network (ANN) modeling technique to understand site–specific air pollution dispersion mechanisms, the order of importance of meteorological variables in determining concentrations as well as the important time scales that influence emission patterns. In this paper, we propose a methodology for extracting the key information from routinely–available meteorological parameters and the emission pattern of sources present throughout the year (e.g. traffic emissions) to build a reliable and physically–based ANN air pollution forecasting tool. The methodology is tested by modeling NO2 concentrations at a site near a major highway in Auckland, New Zealand. The basic model consists of an ANN model for predicting NO2 concentrations using eight predictor variables: wind speed, wind direction, solar radiation, temperature, relative humidity, as well as “hour of the day”, “day of the week” and “month of the year” representing the time variations in emissions according to their corresponding time scales. Of the three input optimization techniques explored in this study, namely a genetic algorithm, forward selection, and backward elimination, the genetic algorithm technique gave predictions resulting in the smallest mean absolute error. The nature of the internal nonlinear function of the trained genetically–optimized neural network model was then extracted based on the response of the model to perturbations to individual predictor variables through sensitivity analyses. A simplified model, based on the successive removal of the least significant meteorological predictor variables, was then developed until subsequent removal resulted in a significant decrease in model performance. The developed ANN model was found to outperform a linear regression model based on the same input parameters. The proposed approach illustrates how the ANN modeling technique can be used to identify the key meteorological variables required to adequately capture the temporal variability in air pollution concentrations for a specific scenario.

Intermittent Streamflow Forecasting and Extreme Event Modelling using Wavelet based Artificial Neural Networks

Forecasting of intermittent stream flow is necessary for water resource planning and management at catchment scale. Forecasting of extreme events and events outside the range of training data used for artificial neural network (ANN) model development has been a major bottleneck in their generalization capabilities till date. Despite of several studies using wavelet analysis in water resource modelling, no study has yet been conducted to explore capabilities of hybrid ANN modelling techniques for extreme events outside the training range. In this study a wavelet based ANN model (WANN) is proposed for intermittent streamflow forecasting and extreme event modelling. This study is carried out in a watershed in semi arid middle region of Gujarat, India. 6 years of hydro-climatic data are used in this study. 4 years of data are used for model training, 1 year for cross-validation and remaining 1 year data are used to evaluate the effectiveness of the WANN model. Two different approaches of data arrangement are considered in this study, in one approach testing data are within the range of training dataset, whereas in another approach testing data are outside the training dataset range. Performance of four different training algorithms and different types of wavelet functions are also evaluated for WANN model development. In this study it is found that WANN model performed significantly better than standard ANN models. It is observed in this study that different wavelet functions have different role in modelling complexities of normal and extreme events. WANN model simulated peak values very well and it shows that WANN model has the potential to be applied successfully for intermittent streamflow forecasting even for the data outside the training range and for extreme events.

Predicting the Impact of Multiwalled Carbon Nanotubes on the Cement Hydration Products and Durability of Cementitious Matrix Using Artificial Neural Network Modeling Technique

In this study the feasibility of using the artificial neural networks modeling in predicting the effect of MWCNT on amount of cement hydration products and improving the quality of cement hydration products microstructures of cement paste was investigated. To determine the amount of cement hydration products thermogravimetric analysis was used. Two critical parameters of TGA test are PHPloss and CHloss. In order to model the TGA test results, the ANN modeling was performed on these parameters separately. In this study, 60% of data are used for model calibration and the remaining 40% are used for model verification. Based on the highest efficiency coefficient and the lowest root mean square error, the best ANN model was chosen. The results of TGA test implied that the cement hydration is enhanced in the presence of the optimum percentage (0.3 wt%) of MWCNT. Moreover, since the efficiency coefficient of the modeling results of CH and PHP loss in both the calibration and verification stages was more than 0.96, it was concluded that the ANN could be used as an accurate tool for modeling the TGA results. Another finding of this study was that the ANN prediction in higher ages was more precise.

Classification of Erodible Soils Using Artificial Neural Networks and Biogeochemical Indicators

Past research has highlighted the need to classify non-point–pollution sources in a watershed in order to work toward environmental remediation of the streams and rivers. The present work couples field measurements of carbon and nitrogen isotopes and the carbon to nitrogen atomic ratio with artificial neural network, multivariate logistic regression, and multivariate discriminant analysis modeling to perform the classification of erodible soils from forest and agricultural land uses. Results show the importance of all three biogeochemical indicators in aiding classification and it is proposed that all three indicators be measured and evaluated in future studies. Multivariate logistic regression performed adequately in classification and at the same time represented the most parsimonious nonnormal model, which highlights the need to avoid creating nonparsimonious models by application of advanced artificial neural network modeling techniques without underlying rationale. Model misclassification occurs due to a shift in biogeochemical cycles in a subset of the agricultural erodible soil data that is reflective of conversion of soil from tilled winter-wheat/barley rotation to conservation, and misclassification of forest soils is attributed to the high variability of soil organic matter cycling. Quantification of samples that can be tagged as potentially misclassified is performed in order that these samples can be further analyzed prior to watershed remediation.

Artificial neural network modeling techniques applied to the hydrodesulfurization process

Reduction of harmful emissions in the combustion of fossil fuels imposes tighter specifications limiting the sulfur content of fuels. Hydrodesulfurization (HDS) is a key process in most petroleum refineries in which the sulfur is mostly eliminated. The modeling and simulation of the HDS process are necessary for a better understanding of the process operation; it is also a requirement to optimize process operation. The objective of this work is to explore the use of different artificial neural network (ANN) architectures in creating various models of the HDS process for the prediction of sulfur removal from naphtha. A database was build using daily records of the HDS process from a Mexican refinery. Accuracy of the predictions was quantified by the root of the mean squared difference between the measured and the predicted sulfur content in the desulfurized naphtha, along with the coefficient of correlation as a measure of the goodness of fit. Results show that the ANN models can be used as practical tools for predictive purposes. One particular example is the ability to anticipate such situations, in the process, that could increase alertness because some variables are deviating from acceptable limits.

Optimization of composition and technology for phosphate graphite mold

In present work, the characteristics of three methods such as the orthogonal design, Fuzzy optimum method and artificial neural network modeling technique were made on the basis of the optimization and evaluation of the performance of the phosphate graphite mold. The variance analysis indicates that the phosphoric acid has greatest influence on both compression strength and tension strength of phosphate graphite mold, both drying temperature and drying time greater, and Al 2O 3 minor, respectively. The Fuzzy multi-objective comprehensive evaluation shows that the optimum technology for phosphate graphite mold designed by us is phosphoric acid 30%, Al 2O 3 30%, drying temperature 400 °C and drying time 60 min. In addition, the ANN can be used to establish mono- and multi-objective models for the prediction of other tests outside orthogonal test with rather high accuracy. However, the predicted results are worse for the linear regressive equations by the orthogonal analysis.

Millimeter-wave Micromachined Filter Design by Artificial Neural Network Modeling Technique

This paper presents a design approach for a 34 GHz λ/2 resonator micormachined bandpass filter by using the artificial neural network (ANN) modeling technique. Three important dimensions of the filter layout are used to capture critical input-output relationships in the ANN model. Once fully developed, the ANN model has been shown to be as accurate as an EM simulator and much more efficient computationally in the design optimization of the filter.