Search

Abstract For developing countries, scarcity of climatic data is the biggest challenge, and model development with limited meteorological input is of critical importance. In this study, five data intelligent and hybrid metaheuristic machine learning algorithms, namely additive regression (AR), AR-bagging, AR-random subspace (AR-RSS), AR-M5P, and AR-REPTree, were applied to predict monthly mean daily reference evapotranspiration (ET0). For this purpose, climatic data of two meteorological stations located in the semi-arid region of Pakistan were used from the period 1987 to 2016. The climatic dataset includes maximum and minimum temperature (T max, T min), average relative humidity (RHavg), average wind speed (U x), and sunshine hours (n). Sensitivity analysis through regression methods was applied to determine effective input climatic parameters for ET0 modeling. The results of performed regression analysis on all input parameters proved that T min, RHAvg, U x, and n were identified as the most influential input parameters at the studied station. From the results, it was revealed that all the selected models predicted ET0 at both stations with greater precision. The AR-REPTree model was located furthest and the AR-M5P model was located nearest to the observed point based on the performing indices at both the selected meteorological stations. The study concluded that under the aforementioned methodological framework, the AR-M5P model can yield higher accuracy in predicting ET0 values, as compared to other selected algorithms.

Abstract Assessing irrigation water quality is one of the most critical challenges in improving water resource management strategies. The objective of this work was to predict the irrigation water quality index of the Bahr El-Baqr, Egypt, based on non-expensive approaches that requires simple parameters. To achieve this goal, three artificial intelligence (AI) models (Support vector machine, SVM; extreme gradient boosting, XGB; Random Forest, RF) and four multiple regression models (Stepwise Regression, SW; Principal Components Regression, PCR; Partial least squares regression, PLS; Ordinary least squares regression, OLS) were applied and validated for predicting six irrigation water quality criteria (soluble sodium percentage, SSP; sodium adsorption ratio, SAR; residual sodium carbonate, RSC; potential of salinity, PS; permeability index, PI; Kelly’s ratio, KR). Electrical conductivity (EC), sodium (Na+), calcium (Ca2+) and bicarbonate (HCO3−) were used as input exploratory variables for the models. The results indicated the water source is not suitable for irrigation without treatment. A good soil drainage system and salinity control measures are required to avoid salt accumulation within the soil. Based on the performance statistics of the root mean square error (RMSE) and the scatter index (SI), SW emerged as the best (0.21% and 0.03%) followed by PCR and PLS with RMSE 0.22% and 0.21% for SAR, respectively. Based on the classification of the SI, all models applied having values less than 0.1 indicate good prediction performance for all the indices except RSC. These results highlight potential of using multiple regressions and the developed machine learning methods in predicting the index of irrigation water quality, and can be rapid decision tools for modelling irrigation water quality.

Abstract Floods caused by dam failures can cause huge losses of life and property, especially in estuarine areas and valleys. In spite of all the capabilities and great improvements reached by man in the construction of dams and their structures, they will remain helpless before the powerful forces of nature, especially those related to tectonic activation, and the occurrence of earthquakes of different intensities. The region extending from the Ilisu Dam in Turkey to the Mosul Dam in Iraq was chosen as an area for this study, and the HEC-RAS application was used to simulate the collapse of the Ilisu Dam due to a major earthquake, to know the magnitude of the risks and losses that could result from this. The Ilisu Dam was built very close to a highly tectonically active fault system, particularly the East Anatolian Fault (EAF), which is one of the largest tectonically active faults in the world with a length of 500 km. This region has witnessed past and present earthquakes of high magnitude (M > 7), especially in the EAF, so the construction of the Ilisu Dam near the EAF fault system is of great concern, as it was built in a basin with very complex seismic activity and geology. Using the HEC-RAS simulation application, the study found that the flood resulting from the collapse of the Ilisu Dam would reach the edges of the Mosul Dam Lake in just 13 h. With a flow of more than 100,000 m3/s, more than 10 billion m3 of water will flow into the Mosul Dam Lake within four days of the disaster. This will lead to the collapse of the Mosul Dam and direct the flood wave of the collapse of these dams towards Baghdad through Mosul, Tikrit, and Samarra. This could pose risks to all Iraqi cities located within the Iraqi sedimentary plain (Mesopotamia), from south of the Mosul Dam up to Basra, in a scenario similar to Noah’s Flood.

Abstract A study was conducted at Punjab Agricultural University, Ludhiana, Punjab, with the aim of monitoring soil loss and infiltration rate in loamy sand soil. The study focused on the effects of applying polyvinyl alcohol (PVA) and mulch under simulated rainfall conditions. The experimental setup involved testing three levels of PVA (0.5%, 0.75%, and 1.0%), one level of wheat-straw mulch (600 g/m2), and a Control treatment (untreated soil). Each of these treatments was replicated four times. The lowest soil loss (20.9 g/m2) was recorded under the 1.0% PVA treatment, while the highest (120.1 g/m2) was seen under the 0.5% PVA treatment. The 1.0% PVA treatment showed a significant reduction in soil loss compared to the 0.5% PVA, 0.75% PVA, mulch, and Control treatments, with reductions of approximately 82.6%, 45.1%, 81.2%, and 89.6%, respectively. Regarding infiltration rates, the Control treatment exhibited the lowest rate (2.4 cm/h), while the 1.0% PVA treatment displayed the highest rate (9.6 cm/h). Additionally, the use of mulch led to a 44.7% reduction in soil loss compared to the treatment without mulch, likely due to the mitigated impact of raindrops. The infiltration rate was significantly higher (4.8 cm/h) under the mulched treatment compared to the unmulched treatment (2.4 cm/h). Overall, the application of PVA and mulch resulted in a drastic reduction in soil loss, likely attributable to the enhanced stability of soil aggregates, improved infiltration rate, and reduced runoff.

Abstract Evapotranspiration plays a pivotal role in the hydrological cycle. It is essential to develop an accurate computational model for predicting reference evapotranspiration (RET) for agricultural and hydrological applications, especially for the management of irrigation systems, allocation of water resources, assessments of utilization and demand and water use allocations in rural and urban areas. The limitation of climatic data to estimate RET restricted the use of standard Penman–Monteith method recommended by food and agriculture organization (FAO-PM56). Therefore, the current study used climatic data such as minimum, maximum and mean air temperature (T max, Tmin, T mean), mean relative humidity (RHmean), wind speed (U) and sunshine hours (N) to predict RET using gene expression programming (GEP) technique. In this study, a total of 17 different input meteorological combinations were used to develop RET models. The obtained results of each GEP model are compared with FAO-PM56 to evaluate its performance in both training and testing periods. The GEP-13 model (T max, T min, RHmean, U) showed the lowest errors (RMSE, MAE) and highest efficiencies (R 2, NSE) in semi-arid (Faisalabad and Peshawar) and humid (Skardu) conditions while GEP-11 and GEP-12 perform best in arid (Multan, Jacobabad) conditions during training period. However, GEP-11 in Multan and Jacobabad, GEP-7 in Faisalabad, GEP-1 in Peshawar, GEP-13 in Islamabad and Skardu outperformed in testing period. In testing phase, the GEP models R 2 values reach 0.99, RMSE values ranged from 0.27 to 2.65, MAE values from 0.21 to 1.85 and NSE values from 0.18 to 0.99. The study findings indicate that GEP is effective in predicting RET when there are minimal climatic data. Additionally, the mean relative humidity was identified as the most relevant factor across all climatic conditions. The findings of this study may be used to the planning and management of water resources in practical situations, as they demonstrate the impact of input variables on the RET associated with different climatic conditions.

Abstract Thermal energy is utilized as an environmentally friendly energy source for seasonal heat and cold storage on a global scale. Specifically, the aquifer thermal energy storage system is highlighted for being cost-effective in cooling and heating applications. The study assesses the sustainability of the aquifer thermal energy storage in the Halabja-Khurmal sub-basin by evaluating six critical criteria: groundwater transmissivity, groundwater temperature, groundwater discharge, groundwater chemistry, population density, and per capita GDP. A multi-criteria decision analysis judgment is applied to analyze all criteria, resulting in a consistency ratio of 0.3% in the analytical hierarchy process. Consequently, the sustainability map for Aquifer Thermal Energy Storage in the Halabja-Khurmal sub-basin for heating reveals that 26.45% of the area is strongly sustainable located in the north and southwestern part of the sub-basin, 73.53% is moderate in the east, central, southeast, and southern regions, 0.02% is weakly sustainable as a tiny area in the southwestern. On the other hand, the sustainability map for Aquifer Thermal Energy Storage in the Halabja-Khurmal sub-basin for cooling reveals that 19% of the area is strongly sustainable located in the north, and southwestern parts of the sub-basin, 78% is moderate in the northeast, east, southeast, west, central, and southern regions, 3% is weakly sustainable as spots in the west and southwestern areas.

Abstract Accurate prediction of reference evapotranspiration (ETo) is crucial for many water-related fields, including crop modelling, hydrologic simulations, irrigation scheduling and sustainable water management. This study compares the performance of different soft computing models such as artificial neural network (ANN), wavelet-coupled ANN (WANN), adaptive neuro-fuzzy inference systems (ANFIS) and multiple nonlinear regression (MNLR) for predicting ETo. The Gamma test technique was adopted to select the suitable input combination of meteorological variables. The performance of the models was quantitatively and qualitatively evaluated using several statistical criteria. The study showed that the ANN-10 model performed superior to the ANFIS-06, WANN-11 and MNLR models. The proposed ANN-10 model was more appropriate and efficient than the ANFIS-06, WANN-11 and MNLR models for predicting daily ETo. Solar radiation was found to be the most sensitive input variable. In contrast, actual vapour pressure was the least sensitive parameter based on sensitivity analysis.

Abstract Scour depth downstream of weirs is considered one of the most important hydraulic problems, which greatly influences the stability of weirs. Recently, artificial intelligence (AI) methods have become increasingly popular in modeling hydraulic variables, especially scour depth, because they can capture nonlinear relationships between input variables and their associated objectives. Despite their importance, these models have problems with hyperparameter tuning in scour depth modeling due to their structures, so algorithms must be used to tune the hyperparameters. Moreover, these algorithms are usually tuned by using the trial-and-error method to select the hyperparameters such as the number of hidden nodes, transfer function, and learning rate, and in this case, the main problem is overfitting during the training phase. To solve these problems, the high-order response surface method (HORSM), an improved version of the response surface method (RSM), is used as an alternative approach for the first time in this study to predict the scour depth. The HORSM model is based on high-order polynomial functions (from two to six) compared with the artificial neural network model (ANN). The findings indicate that the fifth order of the HORSM polynomial function yields the most precise predictions, with a higher coefficient of determination (R 2 ) of 0.912 and Willmott Index (WI) of 0.972 compared to the values obtained using ANN (R 2 = 0.886 and WI = 0.927). Moreover, the accuracy of the predictions is represented by a reduction of the mean square error by up to 44.17 and 29.01% compared to the classical RSM and ANN, respectively. The suggested model established an excellent correlation and accuracy with experimental values.

Abstract Sediment and nutrient pollution in water bodies is threatening human health and the ecosystem, due to rapid land use changes and improper agricultural practices. The impact of the nonpoint source pollution needs to be evaluated for the sustainable use of water resources. An ideal tool like the soil and water assessment tool (SWAT) can assess the impact of pollutant loads on the drainage area, which could be beneficial for developing a water quality management model. This study aims to evaluate the SWAT model’s multi-objective and multivariable calibration, validation, and uncertainty analysis at three different sites of the Yarra River drainage area in Victoria, Australia. The drainage area is split into 51 subdrainage areas in the SWAT model. The model is calibrated and validated for streamflow from 1990 to 2008 and sediment and nutrients from 1998 to 2008. The results show that most of the monthly and annual calibration and validation for streamflow, nutrients, and sediment at the three selected sites are found with Nash–Sutcliffe efficiency values greater than 0.50. Furthermore, the uncertainty analysis of the model shows satisfactory results where the p-factor value is reliable by considering 95% prediction uncertainty and the d-factor value is close to zero. The model's results indicate that the model performs well in the river's watershed, which helps construct a water quality management model. Finally, the model application in the cost-effective management of water quality might reduce pollution in water bodies due to land use and agricultural activities, which would be beneficial to water management managers.

Abstract Stepped weirs are used in a wide range of applications, designed to increase energy dissipation. In this study, laboratory experiments were conducted in a flume on six stepped weir models, with a downstream angle of θ = 26.6°. The physical models used were on a scale of 10:1, and tests of discharges up to 0.055 m3/s were carried out. Several step geometries including traditional step, sill and curve geometries were used to study flow behavior and overall energy dissipation. The laboratory investigations were augmented by modelling numerically the within step flow and energy behavior using a 2-D CFD model, incorporating the k-ε model for turbulence closure. The results showed that energy dissipation was greatest for the curved steps by about 10.5%, where it was observed that the skimming flow regime was shifted to a higher discharge range. Numerical modelling results showed good agreement with the experimental results. An inspection of the modelled streamlines highlighted the increase in vortex intensity for the curve model, reflecting the strong circulation observed. The predicted stepwise energy dissipation showed the energy dissipation increase when the step number Ns increases. For the range of step height hs, tested, our results showed that energy dissipation increased with step height. The results from this study can be used to inform engineering design for steps with θ = 26.6° and provide estimates of the expected energy dissipation and residual energy.

Abstract The river stage is certainly an important indicator of how the water level fluctuates overtime. Continuous control of the water stage can help build an early warning indicator of floods along rivers and streams. Hence, forecasting river stages up to several days in advance is very important and constitutes a challenging task. Over the past few decades, the use of machine learning paradigm to investigate complex hydrological systems has gained significant importance, and forecasting river stage is one of the promising areas of investigations. Traditional in situ measurements, which are sometime restricted by the existing of several handicaps especially in terms of regular access to any points alongside the streams and rivers, can be overpassed by the use of modeling approaches. For more accurate forecasting of river stages, we suggest a new modeling framework based on machine learning. A hybrid forecasting approach was developed by combining machine learning techniques, namely random forest regression (RFR), bootstrap aggregating (Bagging), adaptive boosting (AdaBoost), and artificial neural network (ANN), with empirical mode decomposition (EMD) to provide a robust forecasting model. The singles models were first applied using only the river stage data without preprocessing, and in the following step, the data were decomposed into several intrinsic mode functions (IMF), which were then used as new input variables. According to the obtained results, the proposed models showed improved results compared to the standard RFR without EMD for which, the error performances metrics were drastically reduced, and the correlation index was increased remarkably and great changes in models’ performances have taken place. The RFR_EMD, Bagging_EMD, and AdaBoost_EMD were less accurate than the ANN_EMD model, which had higher R≈0.974, NSE≈0.949, RMSE≈0.330 and MAE≈0.175 values. While the RFR_EMD and the Bagging_EMD were relatively equal and exhibited the same accuracies higher than the AdaBoost_EMD, the superiority of the ANN_EMD was obvious. The proposed model shows the potential for combining signal decomposition with machine learning, which can serve as a basis for new insights into river stage forecasting.

Abstract The commonly used precipitation-based drought indices typically rely on probability distribution functions that can be suitable when the data exhibit minimal discrepancies. However, in arid and semi-arid regions, the precipitation data often display significant discrepancies due to highly irregular rainfall patterns. Consequently, imposing any probability distributions on the data for drought analysis in such regions may not be effective. To address this issue, this study employs a novel drought index called the Discrepancy Precipitation Index (DPI), specifically designed for arid regions. Unlike traditional methods, the DPI does not impose a probability distribution on the precipitation data; instead, it relies on the discrepancy between the data and the mean value. Drought severity classifications (i.e., Drought-I, Drought-II, and Drought-III) are proposed based on the DPI values. The DPI is used to characterize and assess the meteorological drought years based on annual and monsoonal precipitation over nineteen districts in Western Rajasthan, India, during 1901–2019. Additionally, a novel statistic called Discrepancy Measure (DM) is employed to assess the degree of discrepancy in the precipitation climatology of the districts for annual and monsoon precipitation time series. Based on annual precipitation, Jaisalmer district exhibited the highest number of historical drought years (35), whereas three districts, i.e., Jhunjhunu, Dausa, and Bhilwara exhibited the lowest number of drought years (11). Similarly, based on monsoon precipitation, Jaisalmer and Bhilwara encountered the highest (34) and the lowest (11) number of drought years, respectively. The return period of Drought-II is lower for monsoon precipitation-based DPI as compared to that of the annual precipitation-based DPI for all the districts. The DM and DPI-based total number of droughts are found to be strongly correlated for both annual and monsoon precipitation. The DM value is highest for Jaisalmer and lowest for Bhilwara district. The findings reveal DPI as an efficient tool for assessing drought years, particularly in arid climatic conditions. Moreover, as the DM value increases for a precipitation series, the DPI becomes more effective in capturing drought events.

Abstract Exploration of groundwater is an integral part of viable resource growth for society, economy, and irrigation. However, uncontrolled utilization is mainly reported in urban and industries due to the increasing demand for water in semi-arid and arid regions of the world. In the background, groundwater demarcation for potential areas is vital in meeting necessary demand. The current study applied an integrated method comprising the analytical hierarchy process (AHP), multiple influence factors (MIF), combined with a linear regression curve and observatory well data for groundwater prospects mapping. Thematic maps such as flow direction, flow accumulation, elevation map, land use land cover, slope, soil texture, hill shade, geomorphology, normalized vegetation index, and groundwater depth map were generated utilizing remote sensing techniques. The relative weight of each parameter was estimated and then assigned to major and minor parameters. Potential zones for groundwater were classified into five classes, namely very good, good, moderate, poor, and very poor, based on AHP and MIF methods. A spatially explicit sensitivity and uncertainty analysis method to a GIS-based multi-criteria groundwater potential zone model is presented in this research. The study addressed a flaw in the way groundwater potential mapping results are typically presented in GIS-based multi-criteria decision analysis studies, where discrete class outputs are used without any assessment of their certainty with respect to variations in criteria weighting, which is one of the main contributors to output uncertainty. The study region is categorized based on inferred results as very poor, poor, marginal, and very good in potential ground quality 3.04 km2 is considered extremely poor, 3.33 km2 is considered poor, 64.42 km2 is considered very good, and 85.84 km2 is considered marginal zones, which shows reliable and potential implementation. The outcomes of AHP and MIF were validated by linear regression curve and actual water table in a study area. The study results help to formulate the potential demarcation of groundwater zones for future sustainable planning and development of groundwater sources. This study may be helpful to provide a cost-effective solution to water resources crises. The current study finding may be helpful for decision-makers and administrative professionals for sustainable management of groundwater resources for present and future demands.

Abstract Recently, the spread of white spot disease in shrimps has a major impact on the aquaculture activity worldwide affecting the economy of the countries, especially South-East Asian countries like Vietnam. This deadly disease in shrimps is caused by the White Spot Syndrome Virus (WSSV). Researchers are trying to understand the spread and control of this disease by doing field and laboratory studies considering effect of environmental conditions on shrimps affected by WSSV. Generally, they have not considered spatial factors in their study. Therefore, in the present study, we have used spatial (distances to roads and factories) as well as physio-chemical factors of water: Chemical Oxygen Demand (COD), Dissolved Oxygen (DO), Salinity, NO3, P3O4 and pH, for developing WSSV susceptibility maps of the area using Decision Tree (DT)-based Machine Learning (ML) models namely Random Tree (RT), Extra Tree (ET), and J48. Model’s performance was evaluated using standard statistical measures including Area Under the Curve (AUC). The results indicated that ET model has the highest accuracy (AUC: 0.713) in predicting disease susceptibility in comparison to other two models (RT: 0.701 and J48: 0.641). The WSSV susceptibility maps developed by the ML technique, using DT (ET) method, will help decision makers in better planning and control of spatial spread of WSSV disease in shrimps.

Abstract Perfluorooctane sulfonic acid (PFOS), a perfluoroalkyl substance, has engendered alarm over its presence in water sources due to its intrinsic toxicity. Hence, there is a pressing need to identify efficacious adsorbents capable of removing PFAS derivatives from water. To achieve this, batch adsorption studies under various circumstances were employed to tune amorphous polymer networks regarding their morphological configuration, heat durability, surface area and capacity to adsorb PFOS in water. A facile, one-pot nucleophilic substitution reaction was employed to synthesize amorphous polymer networks using triazine derivatives as building units for monomers. Notably, POP-3 exhibited a superlative adsorption capacity, with a removal efficiency of 97.8%, compared to 90.3% for POP-7. POP-7 exhibited a higher specific surface area (SBET) of 232 m2 g−1 compared to POP-3 with a surface area of 5.2 m2 g−1. Additionally, the study emphasizes the importance of electrostatic forces in PFOS adsorption, with pH being a significant element, as seen by changes in the PFOS sorption process by both polymeric networks under neutral, basic and acidic environments. The optimal pH value for the PFOS removal process using both polymers was found to be 4. Also, POP-7 exhibited a better thermal stability performance (300 °C) compared to POP-3 (190 °C). Finally, these findings indicate the ease with which amorphous polymeric frameworks may be synthesized as robust and effective adsorbents for the elimination of PFOS from waterbodies.

Abstract The present research work focused on predicting the electrical conductivity (EC) of surface water in the Upper Ganga basin using four machine learning algorithms: multilayer perceptron (MLP), co-adaptive neuro-fuzzy inference system (CANFIS), random forest (RF), and decision tree (DT). The study also utilized the gamma test for selecting appropriate input and output combinations. The results of the gamma test revealed that total hardness (TH), magnesium (Mg), and chloride (Cl) parameters were suitable input variables for EC prediction. The performance of the models was evaluated using statistical indices such as Percent Bias (PBIAS), correlation coefficient (R), Willmott’s index of agreement (WI), Index of Agreement (PI), root mean square error (RMSE) and Legate-McCabe Index (LMI). Comparing the results of the EC models using these statistical indices, it was observed that the RF model outperformed the other algorithms. During the training period, the RF algorithm has a small positive bias (PBIAS = 0.11) and achieves a high correlation with the observed values (R = 0.956). Additionally, it shows a low RMSE value (360.42), a relatively good coefficient of efficiency (CE = 0.932), PI (0.083), WI (0.908) and LMI (0.083). However, during the testing period, the algorithm’s performance shows a small negative bias (PBIAS = − 0.46) and a good correlation (R = 0.929). The RMSE value decreases significantly (26.57), indicating better accuracy, the coefficient of efficiency remains high (CE = 0.915), PI (0.033), WI (0.965) and LMI (− 0.028). Similarly, the performance of the RF algorithm during the training and testing periods in Prayagraj. During the training period, the RF algorithm shows a PBIAS of 0.50, indicating a small positive bias. It achieves an RMSE of 368.3, R of 0.909, CE of 0.872, PI of 0.015, WI of 0.921, and LMI of 0.083. During the testing period, the RF algorithm demonstrates a slight negative bias with a PBIAS of − 0.06. The RMSE reduces significantly to 24.1, indicating improved accuracy. The algorithm maintains a high correlation (R = 0.903) and a good coefficient of efficiency (CE = 0.878). The index of agreement (PI) increases to 0.035, suggesting a better fit. The WI is 0.960, indicating high accuracy compared to the mean value, while the LMI decreases slightly to − 0.038. Based on the comparative results of the machine learning algorithms, it was concluded that RF performed better than DT, CANFIS, and MLP. The study recommended using the current month’s total hardness (TH), magnesium (Mg), and chloride (Cl) parameters as input variables for multi-ahead forecasting of electrical conductivity (ECt+1, ECt+2, and ECt+3) in future studies in the Upper Ganga basin. The findings also indicated that RF and DT models had superior performance compared to MLP and CANFIS models. These models can be applied for multi-ahead forecasting of monthly electrical conductivity at both Varanasi and Prayagraj stations in the Upper Ganga basin.

Abstract By using microorganisms and the microalgae Chlorella vulgaris in conjunction with sequencing batch reactors (SBRs), the performance of a wastewater treatment facility was studied. For this purpose, the effect of pH, temperature, $${\mathrm{COD}}_{\mathrm{inlet}}$$ COD inlet , and air flowrate on COD removal rate and residual was investigated. A single-factorial optimization method is utilized to optimize the amount of COD removal, and the best result is obtained with a pH of 8, $${\mathrm{COD}}_{\mathrm{inlet}}=600\, \mathrm{mg}/\mathrm{l}$$ COD inlet = 600 mg / l , and an airflow rate of 55 l/min. Under optimal conditions, the amount of residual COD in the effluent reached 36 $$\mathrm{mg}/\mathrm{l}$$ mg / l , showing an augmentation in the efficiency of the desired system. Moreover, empirical correlations are proposed for double-factorial optimization of residual COD and COD removal. Also, a multilayer perceptron artificial neural network is proposed to model the process and predict the residual COD concentration. The useful technique of hyperparameter tuning is utilized to obtain the best result for the predictions. All the effective parameters, including the number of hidden layers, neurons, epochs, and batch size, are adjusted. Data from the experiments agreed well with the artificial neural network modeling results. For this modeling, the values of the correlation coefficient ( $${R}^{2}$$ R 2 ) and mean absolute error (MAE) were obtained as 0.98 and 2%, respectively.