Assessment of machinery energy ratio in potato production by means of Artificial Neural Network

Nowadays , nonlinear time series and artificial neural networks (ANN) models are used for forecasting in the field of business, agriculture and soon. Recent studies have shown, ANN have been successfully used for forecasting of financial and agriculture data series The classical methods used for time series prediction like Box-Jenkins or ARIMA assumes that there is a linear relationship between inputs and outputs. ANN have more advantages that can approximate to model both linear and nonlinear structures in time series, they are not able to handling both structures equally well. The autoregressive integrated moving average (ARIMA) model and two ANN models namely, Radial basis function neural networks (RBFNN), and Elman recurrent neural networks (ERNN) methods were applied to Hyderabad airport traffic data. The data obtained for 15 years from 2002–2003 to 2016–2017 about domestic and international passenger of International Airport of Hyderabad, India. In this research paper, we compared the performances of ARIMA, RBFNN and ERNN were based on three measures: mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean square error (RMSE). The results showed that RBFNN obtained the smallest MAE, MAPE and RMSE in both the modeling and forecasting processes. The performances of the three models ranked in ascending order were: ARIMA, ERNN and the RBFNN model. Keywords: T ime series, forecasting, artificial neural networks, ARIMA models, radial basis function neural networks, and Elman recurrent neural networks Cite this Article R. Ramakrishna, Berhe Aregay, Tewodros Gebregergs. The Comparison in Time Series Forecasting of Air Traffic Data by Autoregressive Integrated Moving Average Model, Radial Basis Function and Elman Recurrent Neural Networks. Research & Reviews: Journal of Statistics . 2018; 7(3): 75–90p.

ANN based simulation and experimental verification of analytical four- and five-parameters models of PV modules

The search for better climate change adaptation techniques for addressing environmental and economic issues due to changing climate is of paramount interest in the current era. One of the many ways Pacific Island regions and its people get affected is by dry spells and drought events from extreme climates. A drought is simply a prolonged shortage of water supply in an area. The impact of drought varies both temporally and spatially that can be catastrophic for such regions with lack of resources and facilities to mitigate the drought impacts. Therefore, forecasting drought events using predictive models that have practical implications for understanding drought hydrology and water resources management can allow enough time to take appropriate adaption measures. This study investigates the feasibility of the Artificial Neural Network (ANN) algorithms for prediction of a drought index: Standardized Precipitation-Evapotranspiration Index (SPEI). The purpose of the study was to develop an ANN model to predict the index in two selected regions in Queensland, Australia. The first region, is named as the grassland and the second as the temperate region. The monthly gridded meteorological variables (precipitation, maximum and minimum temperature) that acted as input parameters in ANN model were obtained from Australian Water Availability Project (AWAP) for 1915–2013 period. The potential evapotranspiration (PET), calculated using thornthwaite method, was also an input variable, while SPEI was the predictand for the ANN model. The input data were divided into training (80%), validation (10%) and testing (10%) sets. To determine the optimum ANN model, the Levenberg-Marquardt and Broyden-Fletcher-Goldfarb-Shanno quasi-Newton backpropagation algorithms were used for training the ANN network and the tangent sigmoid, logarithmic sigmoid and linear activation algorithms were used for hidden transfer and output functions. The best architecture of input-hidden neuron-output neurons was 4-28-1 and 4-27-1 for grassland and temperate region, respectively. For evaluation and selection of the optimum ANN model, the statistical metrics: Coefficient of Determination (R 2 ), Willmott’s Index of Agreement (d), Nash-Sutcliffe Coefficient of Efficiency (E), Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) were employed. The R 2 , d, E, RMSE and MAE for optimum ANN models were 0.9839, 0.9909, 0.9838, 0.1338, 0.0882 and 0.9886, 0.9935, 0.9874, 0.1198, 0.0814 for grassland and temperate region, respectively. When prediction errors were analysed, a value of 0.0025 to 0.8224 was obtained for the grassland region, and a value of 0.0113 to 0.6667 was obtained for the temperate region, indicating that the ANN model exhibit a good skill in predicting the monthly SPEI. Based on the evaluation and statistical analysis of the predicted SPEI and its errors in the test period, we conclude that the ANN model can be used as a useful data-driven tool for forecasting drought events. Broadly, the ANN model can be applied for prediction of other climate related variables, and therefore can play a vital role in the development of climate change adaptation and mitigation plans in developed and developing nations, and most importantly, in the Pacific Island Nations where drought events have a detrimental impact on economic development.

The goal of this study is to compare the forecasting performance of classical artificial neural network and the hybrid model of artificial neural network and genetic algorithm. The time series data used is the monthly gold price per troy ounce in USD from year 1987 to 2016. A conventional artificial neural network trained by back propagation algorithm and the hybrid forecasting model of artificial neural network and genetic algorithms are proposed. Genetic algorithm is used to optimize the of artificial neural network neurons. Three forecasting accuracy measures which are mean absolute error, root mean squared error and mean absolute percentage error are used to compare the accuracy of artificial neural network forecasting and hybrid of artificial neural network and genetic algorithm forecasting model. Fitness of the model is compared by using coefficient of determination. The hybrid model of artificial neural network is suggested to be used as it is outperformed the classical artificial neural network in the sense of forecasting accuracy because its coefficient of determination is higher than conventional artificial neural network by 1.14%. The hybrid model of artificial neural network and genetic algorithms has better forecasting accuracy as the mean absolute error, root mean squared error and mean absolute percentage error is lower than the artificial neural network forecasting model.

Reference evapotranspiration (ET0) is used to determine crop water requirements under different climatic conditions. In this study, soft computing tools viz. artificial neural network (ANN) and k-nearest neighbors (KNN) models were evaluated for forecasting daily ET0 by comparing their performance with the Penman-Monteith model (PM) using climatic data from 1990 to 2020 of the Indian Agricultural Research Institute (IARI) farm observatory, New Delhi, India. The performance of these models was assessed using statistical performance indices viz., mean absolute error (MAE), mean squared error (MSE), correlation coefficient (r), mean absolute percentage error (MAPE), and index of agreement (d). Results revealed that the ANN model with sigmoid activation function and L-BFGS (Limited memory-Broyden-Fletcher-Goldfarb-Shanno) learning algorithm was selected as the best performing model amongst 36 ANN models. Amongst 4 KNN models developed and tested, the K4 KNN model was observed to be the best in forecasting daily ET0. Overall, the best ANN model (M11) outperformed the K4 KNN model with MAE, MSE, r, MAPE, and d values of 0.075, 0.018, 0.997, 2.76 %, and 0.974, respectively and 0.091, 0.053, 0.984, 3.16 %, and 0.969, respectively during training and testing periods. Thus, we conclude that the ANN technique performed better than the KNN technique in forecasting daily ET0. Sensitivity analysis of the best ANN model revealed that wind speed was the most influential input variable compared to other weather parameters. Thus, the ANN model to forecast daily ET0 accurately for efficient irrigation scheduling of different crops in the study region may be recommended.

The growing demand of development of appropriate and relevant wastewater treatment technology is drastically increasing in rural and urban communities in many parts of the world including South Africa. This is largely exacerbated by the escalation of water demand and decreasing potable water availability. As a result, advanced research related to the development and optimization of water treatment technologies is becoming an urgent necessity including research focusing on wastewater recycling and reclamation. Meanwhile, horizontal roughing filter (HRF) technology is one such physical water pre-treatment system that can effectively and efficiently treat wastewater and thus reduce the reliance on potable water use. Therefore, this study aimed at modelling HRF in order to investigate the option of domestic greywater reuse for delivering desired water quality for nonpotable applications. The overall aim of the study was modelling the HRF in order to improve its performance and several objectives were investigated in this study. The first one was the characterization of biological and physico-chemical strength of greywater originated from kitchen, bath and laundry sources. The second objective investigated the HRF performance/efficiency after treating various domestic greywater pollutants. The third objective investigated the controlling factors affecting the performance and optimization of the HRF during its operation. This was investigated based on design of experiments (DOE) and response surface methodology (RSM). Based on the artificial neural network (ANN), the first objective investigated the filter duration in a HRF using ANN modelling for high level of contaminants in domestic greywater. Secondly, the ANN models applicable to a HRF were investigated and used for the prediction of greywater quality variables from the output stream of the HRF based on experimental data obtained from the operation of the HRF equipment. The first step in water treatment processes requires quality analysis in order to understand the constituent of water pollutants. Therefore, the experimental analysis of biological and physicochemical contents in greywater sources was conducted in this study. The next aspect involved treatment of mixed domestic greywater using a three compartment HRF unit which was fixed at a low filtration rate of 0.3 m/h. The effect of operating parameters on the HRF performance was studied factorial design and optimization. The factorial design application in HRF defines performance based on derivation of right factor settings for the effective operation of HRF. The aspect of ANN was undertaken to investigate the applicability, effectiveness and predictive ability of ANN within a HRF equipment. The use of ANN in HRF can serve as a monitoring tool in terms of performance and also as an indicator of any quality deviation that might be occurring during the filter operation. The key findings were obtained on qualitative analysis of domestic greywater originating from a peri-urban community for the quantification of biological and physico-chemical contaminants. The significant quality difference was recorded in greywater sources and the kitchen greywater source recorded the highest load of pollutants compared to the laundry and bathing sources at p<0.05 significant level. Furthermore, the quality difference was evident in greywater sources in terms of daily households’ social conditions, activities and practices. Also, the analysis of microbes in domestic greywater recorded high values of Escherichia coli (E. coli) and total coliform contamination which poses health related risks in domestic greywater reuse. Therefore, further treatment of domestic greywater prior to reuse remained necessary. The effectiveness of HRF was evident in removing biological and physico-chemical pollution load in domestic greywater at 0.3 m/h filtration rate. An average of 90% turbidity removal was obtained with 86% removal of conductivity and 84% of total solids and more than 50-70% removal of chemical oxygen demand (COD) within the HRF system. The E. coli and total coliforms were totally removed in the three compartment HRF. Based on DOE analysis, the significant factors identified were flowrate, gravel media, filter bed height and filter length and most significant contributing factor identified was filtration rate. Furthermore, the optimization of the HRF resulted in a high efficiency of 76% for the removal of turbidity. Results on ANN modelling for the prediction of turbidity of the effluent stream from the HRF showed good learning abilities of the ANN and the optimal ANN structure obtained was 4-7-1 structure using the trainlm algorithm. The mean square error (MSE) value below 10% was obtained after training and the R correlation coefficient >0.9 was obtained in training, testing, validation and all data sets. For the prediction of COD, the optimal ANN architecture was 3-10-1 which was obtained with trainlm training algorithm. A satisfactory mean absolute percentage error (MAPE), low mean absolute error (MAE) and high R correlation coefficients close to 1 for the training and testing sets were also recoded for this ANN model for the prediction of COD. The other objective was the investigation of filter duration in HRF using ANN and a 4-8-2 optimal structure was obtained with the trainlm algorithm which outperformed other training algorithms for the prediction of filter duration along with turbidity. Also, a high R correlation coefficient and low MSE value was obtained for this optimal ANN model for the predicted filter duration. For this model, satisfactory R correlation values for training, testing, validation and all data were close to 1. Results on feedforward multi-input multi-output (MIMO) ANN showed good accuracy in predicting multioutput parameters of domestic greywater effluent from the HRF. The optimal ANN architecture obtained through a trial-and-error approach for MIMO ANN was 7-15-4. During training, different structures of ANN were investigated through varying training functions, neurons and combination of physico-chemical parameters and learning functions. For the optimal ANN model, the MSE of 0.001 was finally obtained based on the training data set. Furthermore, the R correlation values above 0.9 for training, testing, validation and all data sets were obtained. The optimal ANN model also showed good prediction and satisfactory accuracy when a new set of sample data was presented to the network. Therefore, based on the objectives and findings of this study, the pollution load in domestic greywater characteristics can contain a number of pollutants and can significantly vary with greywater sources. It is also important to note that the HRF significantly showed effectiveness in treating physical pollutants and large amounts of chemical and biological pollutants. From the findings and based on the HRF, it was also noted that the chemical pollutants can be significantly removed using a combination of physical and chemical treatment processes in order to remove more pollutants. This was observed by a high removal of physical pollutants such as turbidity, conductivity and solids while domestic greywater biodegradability ratio was lower than 0.5. Furthermore, for the DOE/RSM techniques, it was also observed that the effective filter performance of the HRF is a function of multi-design parameters such as filtration rate, filter length, gravel media and bed height and multi factor optimization was useful in this research work. Finally, the ANN showed effective characteristics and accuracy in the HRF equipment for the prediction of multi-output variables of the effluent greywater from the HRF following mixed domestic greywater pre-treatment.

Prediction of performance parameters in friction stir processing using ANN and multiple regression models

Ultrasound estimation of fetal weight with the use of computerized artificial neural network model

Renewable energy sources (RES) are being used and integrated into the electrical grid as a result of the environment’s effects and the ever-increasing demand for energy. Reliable and accurate forecasts are necessary to address environmental concerns and improve grid management due to the intermittent availability of renewable energy sources. This study focuses on improving ANN-based techniques for precise solar irradiance prediction as the prediction accuracy of an artificial neural network (ANN) is impacted by the random assignment of weights to its edges. As a result, we proposed hybrid solar irradiance forecasting models in which the cuckoo search algorithm (CSA) and adaptive moment estimation (ADAM) are used to optimize the weights assigned to the ANN’s edges. Two models were presented in this study namely: ADAM-optimized ANN model and a novel two-stage optimization technique known as CSA-ADAM optimized ANN model for accurate and reliable forecasting of solar irradiance. Both models were tested using actual weather data, and standard error metrics like mean squared error (MSE), mean absolute percentage error (MAPE), mean absolute error (MAE), and root mean square error (RMSE) to assess their accuracy. The outcomes demonstrate that ADAM-optimized ANN model produced MSE = 0.52, MAPE = 0.18%, MAE = 0.64, and RMSE = 0.72, and CSA-ADAM optimized ANN model obtained MSE = 0.25, MAPE = 0.17%, MAE = 0.43, and RMSE = 0.50. We evaluated the practicality of both models by comparing their average prediction times using the same test dataset. While the ADAM-optimized ANN model took an average of 0.1093 ± 0.0085 seconds to make predictions on the test data, the CSA-ADAM optimized ANN model took 0.1110 ± 0.0058 seconds. These findings demonstrate that using CSA to optimize the ANN weights increases the accuracy of solar irradiance predictions.

Non-destructive estimation of leaf area of durian (Durio zibethinus) – An artificial neural network approach

Fly-rock induced by blasting is an undesirable phenomenon in quarries. It can be dangerous for humans, equipment, and buildings. To minimize its undesirable hazards, we proposed a state-of-the-art technology of fly-rock prediction based on artificial neural network (ANN) models and their robust combination, called EANNs model (ensemble of ANN models); 210 fly-rock events were recorded to develop and test the ANN and EANNs models. Of thi sample, 80% of the whole dataset was assigned to develop the models, the remaining 20% was assigned to confirm the models developed. Accordingly, five ANN models were designed and developed using the training dataset (i.e., 80% of the whole original data) first; then, their predictions on the training dataset were ensembled to generate a new training dataset. Subsequently, another ANN model was developed based on the new set of training data (i.e., EANNs model). Its performance was evaluated through a variety of performance indices, such as MAE (mean absolute error), MAPE (mean absolute percentage error), RMSE (root-mean-square error), R2 (correlation coefficient), and VAF (variance accounted for). A promising result was found for the proposed EANNs model in predicting blast-induced fly-rock with a MAE = 2.777, MAPE = 0.017, RMSE = 4.346, R2 = 0.986, and VAF = 98.446%. To confirm the performance of the proposed EANNs model, another ANN model with the same structure was developed and tested on the training and testing datasets. The findings also indicated that the proposed EANNs model yielded better performance than those of the ANN model with the same structure.

Among all solutions for disrupted vortex formation in shaft spillways, an innovative one called Circular Piano Key Spillway, based upon piano key weir principles, has been experimented less. In this study, the potential of Artificial Neural Networks (ANN) in estimating the amounts of discharge coefficient of Circular Piano Key Spillway has been evaluated. In order to pursue this purpose, the results of some physical experiments were used. These experiments have been conducted in the hydraulic laboratory using different physical models of Circular Piano Key Spillway including three models with different angles of 45, 60 and 90 degrees. Data from those experiments were used in training and test steps of ANN models. Multilayer Perceptron (MLP) network with Levenberg-Marquardt backpropagation algorithm was used. The performance of artificial neural network was measured by these statistical indicators: coefficient of determination (R2), mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE) and optimum quantities of statistical indicators for test step were assessed 0.9999, 0.4988, 0.5963 and 0.9999 respectively, for Circular Piano Key Spillway with an angle of 90 degree and for training step were assessed 0.9999, 0.5479, 0.6305 and 0.9999 respectively, for Circular Piano Key Spillway with an angle of 90 degree. In other words, Circular Piano Key Spillway with an angle of 90 degrees has the optimum performance, both in training and test steps. Artificial Neural Network model can successfully estimate the amounts of discharge coefficient of Circular Piano Key Spillway.

Quality prediction and rivet/die selection for SPR joints with artificial neural network and genetic algorithm

Financial time series forecast is challenging due to nonlinear mean dynamics, volatility clustering, and long-memory effects. Traditional hybrid models such as Autoregressive Integrated Moving Average – Generalised Autoregressive Conditional Heteroscedasticity (ARIMA–GARCH) and Fractional Generalised Integrated Autoregressive Conditional Heteroscedasticity – Artificial Neural Network (FIGARCH–ANN), improve forecasting performance but remain limited by linear mean assumptions, short-memory volatility, or incomplete treatment of nonlinearities. These constraints are particularly evident in emerging markets like Nigeria, where financial returns display pronounced nonlinear and persistent volatility patterns. Thus, this study developed a hybrid model to address volatility, nonlinearity, and long memory in residuals. Daily Nigeria All Share Stock Index Data (2001-2019), exhibiting these characteristics was used to assess the forecast performance of the new hybrid Exponential Autoregressive – Fractional Integrated Generalized Autoregressive Conditional Heteroscedasticity – Artificial Neural Network (ExpAR-FIGARCH-ANN) model in comparison to the existing Exponential Autoregressive – Fractional Integrated Generalized Autoregressive Conditional Heteroscedasticity (ExpAR-FIGARCH) and Artificial Neural Network (ANN) models using error-based metrics, viz Mean Absolute Percentage Error (MAPE), Mean Absolute Error (MAE) and Mean Squared Error (MSE). The empirical findings show that the hybrid ExpAR-FIGARCH-ANN model outperformed the standalone ExpAR-FIGARCH and ANN model. It achieved the lowest error metrics (MSE = 0.0029, MAE = 0.0352, MAPE = 1.68%), confirming superior predictive performance. This enhanced performance is ascribed to the novel capability of the model to concurrently address nonlinear mean dynamics, long-memory volatility, and residual nonlinearities. It provides a more accurate forecast than existing hybrid models, thus, has potential applications beyond stock indices.

Cavitation, a physical phenomenon that detrimentally affects pump performance and reduces pump life, can cause wear on pump elements. Various engineering methods have been developed to identify the initiation and full development of the cavitation process. One such method is the determination of the net positive suction head (NPSH) through a 3% decrease in total head (Hm) at a constant flow rate. In radial pumps, commonly used in agricultural irrigation and industry, cavitation conditions result in a sudden drop in the Hm-Q curve, making it challenging to detect the 3% Hm value drop. This study differs from others in the literature by modelling NPSH, noise, and vibration levels using three machine learning models, specifically artificial neural networks (ANN), support vector machines (SVM), and decision tree regression (DTR). The best-performing model predicts NPSH, noise, and vibration levels corresponding to a 3% decrease in Hm level. The present study determined the NPSH values of a horizontal shaft centrifugal pump at different flow rates and constant operating speed, and the vibration and noise levels were measured for these NPSH values. For each of the NPSH, noise, and vibration levels, ANN, SVM and DTR models were created. The performances of these models were evaluated using criteria such as root mean squared error (RMSE), Mean Absolute Error (MAE) and mean absolute percentage error (MAPE). In addition, Taylor and error box diagrams were created. The ANN model and DTR yielded high accuracy predictions for NPSH values (R2 = 0.86 and R2 = 0.8, respectively). The ANN model provided the best prediction performance for noise and vibration levels. By entering the level of 3% drop in the Hm value of the pump as external data input to the ANN model, NPSH3, noise, and vibration levels were determined. The ANN models can be effectively employed to determine NPSH3, noise, and vibration levels, particularly in radial flow pumps, where detecting 3% reductions in manometric height value is challenging.