High R2 Values Research Articles

International Air Cargo Volume Forecasting based on the Stacking Fusion Model

Accurate forecasting of international air cargo volumes is crucial for air cargo carriers, which optimizes operational efficiency and profitability and strengthens market competitiveness and risk response capabilities. This study focuses on the open-source cargo data of over 100 international airlines in India from January 2015 to March 2017. Firstly, the missing data must be dealt with. Then, histograms and probability density plots are used to depict the distribution of the cargo volume and analyze the dynamics of the cargo volume of each company at annual and quarterly levels. Subsequently, key factors affecting cargo volumes are analyzed, with a special focus on the cargo performance of the Top 5 airlines, which is visualized in box plots. Given the time-series nature of cargo data, this study adopts the data sequential partitioning method. It innovatively utilizes the Stacking ensemble learning framework, integrating regression models such as Random Forest, Decision Tree, Extreme Gradient Boosting, and K-Nearest Neighbor as the base learners. The meta-learner is constructed to integrate the outputs of each base model, and the hyperparameters of the model are optimized by grid search and cross-validation. Comparison results show that the Stacking model has a higher R² value and lower MSE than any single base learners, indicating that it is more effective in international air cargo volume prediction. The prediction model based on stacking ensemble learning provides an effective strategy for improving air cargo prediction accuracy.

Modelling and optimizing secondary metabolites production in Spirodela polyrhiza using machine learning

Spirodela. polyrhiza as one of the potential feedstocks for industrial natural dye and pharmaceutical products, efficient process to produce from it were scarce. Here, Artificial Neural Network (ANN) and Genetic Algorithm (GA) were employed for modelling and optimization respectively. Current study focus on establishing useable models and anticipating the optimal concentration of nitrate (NO3−), phosphate (PO43−) and ammonium (NH4+) to give highest relative growth rate (RGR) and produce maximum amount of anthocyanins, chlorophylls, phenolic compounds and antioxidants. In short, 12 and 11 neurons in hidden layer were found to be most suitable neuron numbers used for the modelling of growth and secondary metabolites respectively. All the models exhibited a high degree of robustness and performance with a high r2 value (higher than 0.75) and a low mean squared error (lower than 0.05). GA also anticipated the best concentration of macronutrients to achieve the maximum growth and metabolites production with significant low errors (below 7 %). The employment of machine learning was useful for such applications.

THE ADOPTION OF AI-DRIVEN CHATBOTS INTO A RECOMMENDATION FOR E-COMMERCE SYSTEMS TO TARGETED CUSTOMER IN THE SELECTION OF PRODUCT

The research looks into the Adoption of AI-Driven chatbots into a recommendation for E-Commerce systems to targeted customer in the selection of product, particularly their function in product selection and overall customer experience. The study is based on five assumptions: improved product selection accuracy, increased user happiness, influence on customer engagement and purchase decisions, enhanced user experience and retention rates, and effect on consumer trust. The findings show that AI-powered chatbots greatly improve product selection accuracy, with a regression weight of 0.693 and a beta coefficient of 0.480, representing a 48% improvement per unit of implementation. User satisfaction has also significantly improved, as evidenced by a regression weight of 0.897 and a beta coefficient of 0.840, both with high statistical significance. Chatbots have a favourable impact on consumer engagement and purchasing decisions, as indicated by substantial regression weights and beta coefficients for each variable. AI chatbots improve user experience and retention, as evidenced by high regression weights and R² values. Finally, chatbots improve customer trust, with a regression weight of 0.447 and a beta coefficient of 0.200. Overall, the study shows that AI-powered chatbots significantly increase several aspects of e-commerce, including product selection accuracy, user pleasure, engagement, retention, and trust.

Ensemble machine learning models for compressive strength and elastic modulus of recycled brick aggregate concrete

This study delivers an in-depth analysis of predicting the compressive strength and elastic modulus of recycled brick aggregate concrete (RBAC). It assembles an extensive database of 633 compression test results and develops five ensemble learning models—random forest (RF), gradient boosting regression trees (GBRT), extreme gradient boosting (XGB), light gradient boosting machine (LGBM) and stacking (St). These models are evaluated against existing empirical formulas with respect to their effectiveness. The findings reveal that machine learning (ML) models outperform existing formulas: for compressive strength prediction, traditional models achieve a maximum determination coefficient R² of 0.38, whereas ML models attain an R² range of 0.91–0.94. For elastic modulus, the highest R² values are 0.44 for traditional models and 0.97 for ML models. Notably, the LGBM and St models excel in predicting compressive strength and elastic modulus, respectively. The study also identifies critical parameters influencing RBAC’s compressive behavior and highlights their impact trends. Remarkably, while the mass-weighted water absorption of coarse aggregates and the replacement ratio of recycled brick aggregates (RBAs) have less impact on compressive strength compared to the effective water-to-cement ratio, they become more influential for elastic modulus. Additionally, the decrease in elastic modulus due to higher RBA replacement ratios or increased mass-weighted water absorption of coarse aggregates exceeds the corresponding reduction in compressive strength. This research not only deepens the understanding of RBAC’s mechanical properties but also provides valuable predictive tools for civil engineering applications.

Multimodal adversarial informer for highway vehicle lane-changing trajectory prediction

To enhance the perception of vehicle trajectory information and lane-changing decision-making capabilities in intelligent connected vehicles during multivehicle interaction scenarios, we propose a novel method based on a Multimodal Adversarial Informer (MAI) for highway multivehicle lane-changingrajectory prediction. This method achieves spatiotemporal features of target and surrounding vehicles through graph learning of temporal features and spatial adjacency matrices. Considering the heading angle and vehicle local X-axis displacement, the vehicle trajectory samples are categorized for training and validation of the multimodal Informer. A multi-criterion discriminator is utilized to judge whether the generated trajectory fits the requirements of accuracy and rationality. After adversarial learning, the optimal vehicle lane-changing trajectory prediction is obtained using the proposed MAI. Experiments conducted with the NGSIM dataset demonstrate the comparative performance of baseline models on three different noise-added testing datasets using MAE, RMSE, and R² metrics. The MAI model consistently outperforms the others, achieving the lowest MAE and RMSE and the highest R² values across all datasets, indicating superior predictive accuracy and fit. Furthermore, the results show that the proposed MAI framework maintains a relatively low prediction error over both short-term and long-term horizons compared to baseline models.

Geoinformatics and Modeling Approaches for Estimating Above-Ground Biomass in the Moist Deciduous Forests of Uttara Kannada, Karnataka, India

The present study aims to estimate Above-Ground Biomass (AGB) in the moist deciduous forests of Mundgod Taluk, Uttara Kannada district, Karnataka, India using a combination of field sampling and remote sensing techniques. AGB was assessed using Sentinel-2A satellite imagery with a spatial resolution of 10 meters. Field data were collected using the Point-Centered Quarter (PCQ) method, it includes measuring tree girth at breast height (GBH), tree height, and canopy density. AGB was calculated using artificial form factor and specific gravity of tree species and further modelled using the Normalized Difference Vegetation Index (NDVI) derived from satellite data. The results showed a significant correlation between canopy density and biomass, with very dense forests exhibiting the highest AGB values. The field-measured AGB was 436.96 t/ha for very dense forests, 259.89 t/ha for moderately dense forests, and 161.67 t/ha for open forests. A linear regression model developed between AGB and NDVI showed a high R² value of 0.91 for open forests, 0.87 for moderately dense forests, and 0.85 for very dense forests. The total area-weighted biomass for the study area was estimated at 8.87 million tons, with 4.79 million tons in very dense forests, 3.44 million tons in moderately dense forests, and 0.76 million tons in open forests. The regression model validation indicated low Root Mean Square Error (RMSE) values, with the moderately dense forests 8.81 t/ha and the in very dense forests 7.54 t/ha. This study highlights the effectiveness of integrating field measurements with remote sensing for rapid, reliable AGB estimation. The approach offers valuable insights into biomass distribution and carbon sequestration potential, providing a scalable method for carbon inventories at state and national levels, contributing to forest management and climate change mitigation efforts.

Colour Analysis of Sausages Stuffed with Modified Casings Added with Citrus Peel Extracts Using Hyperspectral Imaging Combined with Multivariate Analysis

Recycling citrus peel waste offers several significant contributions to sustainability, transforming what would otherwise be discarded into valuable resources. In this study, the colour of sausages stored for 16 days, with varying amounts of orange extract added to the modified casing solution, was evaluated using response surface methodology (RSM) and a hyperspectral imaging system within the spectral range of 350–1100 nm for the first time. To enhance model performance, spectral pre-treatments such as normalisation, first derivative, standard normal variate (SNV), second derivative, and multiplicative scatter correction (MSC) were applied. Both raw and pre-treated spectral data, along with colour attributes, were fitted to a partial least squares regression model. The RSM results indicated that the highest R2 value, 80.61%, was achieved for the b* (yellowness) parameter using a second-order polynomial model. The interactive effects of soy oil and orange extracts on b* were found to be significant (p < 0.05), and the square effects of soy oil on b* were significant at the 1% level. The identified key wavelengths for colour parameters can simplify the model, making it more suitable for practical industrial applications.

Harnessing Moringa oleifera root powder (MORP) for the sustainable remediation of heavy metal contaminated water.

Heavy metal environmental pollution is rapidly increasing due to the increase in industrialization and urbanization. Industrial processes, such as paint production, mining, and raw materials producing industries release effluents rich in heavy metals, like Pb2+, Cd2+, Cu2+, and Cr3+. These heavy metals are dangerous because they persist in nature, are non-biodegradable and they have high tendency to accumulate in the environment and in living organisms who are exposed to them. This work studied the removal of heavy metals (Cu, Pb, Cr, and Cd) from aqueous solution using Moringa oleifera root powder (MORP) as the adsorbent. The MORP was characterized by SEM, FTIR, BET, and XRD. Batch adsorption experiments carried out investigated the effects of adsorbate concentration, adsorbent dosage, agitation time, pH and temperature on adsorption. The optimum parameters are: contact time (90 min); pH (9); adsorbent dose (0.6); metal ion concentration (30 mg L-1) for Cr and 40 mg L-1 for the rest; and temperature (50 °C) for Cu and Pb, and 70 °C for Cr and Cd. These experimental data were analyzed with 5 isotherm models (Temkin, Flory-Huggins, Langmuir, D-R and Freundlich). The result obtained fitted best to Temkin isotherm in comparison to others. Kinetic studies revealed that the pseudo-second order kinetic model best described the adsorption (with high R2 values ranging from 0.9810-0.9976) compared to pseudo-first order and intra-particle diffusion kinetics model. Results of the thermodynamic study showed that the sorption process was endothermic for Cu and Pb, but exothermic for Cd and Cr. The adsorbent showed good adsorptive tendencies toward the ions studied, and could be applied on an industrial scale for the remediation of metal contaminated water.

Optimization of Flange Design for Engine Assembly Stand using RSM and NSGA-II Based on FEA Data

This article presents the results of a research study that focused on optimizing the flange design for engine assembly stands using the Response Surface Methodology (RSM) and Nondominated Sorting Genetic Algorithm II (NSGA-II) based on finite element analysis (FEA) simulation data. The study investigated the impact of three main dimensions (d1, d2, d3) of the flange as independent variables on the mass (m, kg), Von Mises stress (V, MPa), and displacement (D, mm) as dependent variables. The regression models developed using the RSM exhibited high R2 values of 0.9896, 0.998, and 0.9997 for m, V, and D, respectively. The multi-objective optimization results obtained through NSGA-II yielded 39 Pareto solutions, with d1 ranging from 10 to 27.43 mm, d2 ranging from 30 to 50 mm, and d3 ranging from 75 to 85 mm. These values corresponded to m values ranging from 2.93 to 7.58 kg, V values ranging from 54.8 to 342.6 MPa, and D values ranging from 0.006 to 0.270 mm. For verification purposes, Solution No.17 was selected. The results showed that the redesigned flange's mass, Von Mises stress, and displacement deviated by 0.95%, 2.28%, and 1.14%, respectively, from the optimal values obtained through the optimization process. These findings provide strong evidence for the high reliability of the optimization method employed in this study.

Comparative Analysis of Machine Learning Models for Earthquake Prediction: A Case Study of Düzce, Türkiye

This paper explores the application of machine learning models, specifically XGBoost, Stacking Regressor, and Long Short-Term Memory (LSTM), for predicting earthquake magnitudes in Düzce, Turkey. The models were trained and tested on seismic data to predict moment magnitude (Mw). The performance of each model was evaluated using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and the coefficient of determination (R²). The results indicate that the XGBoost model outperforms the other models with a higher R² value and lower error metrics, providing a more accurate prediction of seismic events.

Numerical and Optimization Studies on Reactivity Controlled Compression Ignition Engine with Hydrogen and Split Injections

Abstract Effective abatement of harmful tail-pipe emissions from fossil fuel run engines is achieved through low-temperature combustion strategies; Reactivity Controlled Compression Ignition mode of operation has been successful among such concepts. The present work deals with numerical work performed using ANSYS Forte software with n-heptane as a high-reactivity fuel and hydrogen in different proportions as a low-reactivity fuel, respectively. With total energy fixed, the amount of hydrogen is varied from 0-80% fuel injection is regulated accordingly. Pertinent engine in-cylinder parameters with patterns are extracted, emphasizing the combustion phenomena of RCCI operation with the lowest possible emissions targeted, with the combined effects of hydrogen induction, Start of Injection, and split injections. The contours of fuel vapour and emission parameters are obtained to relate the performance with emissions. It is noted that with a split injection strategy at 50/50 and 75/25 split strategy and 45-50% energy share from hydrogen, the NOx, Soot reductions and thermal efficiency penalty are in the range of about 5.5%, 24% and 7.5%, respectively, also, with 30% EGR, about 95% NOx reduction but with higher soot values. A 75/25 split and advanced injection timing of 25obTDC resulted in the RCCI mode of operation with reduced soot emissions, and the use of EGR has resulted in high levels of soot and poor fuel efficiency. Among the models of machine learning tested, Random Forest regressor emerged as the most suitable, with higher R2 values, indicating better predictive capability.

Kinetic and thermodynamic analysis of engineered alginate–chitosan hydrogel based scaffolds for drug delivery applications

The synthesis of stimulus–responsive hydrogel–based scaffolds has been extensively investigated for controlled drug delivery applications. This study focused on synthesizing alginate and chitosan–based scaffolds for potential use in releasing antibiotic antimycotic solution (AAS) drugs. The controlled release of AAS from AAS–loaded alginate–chitosan scaffolds was investigated under distinct temperatures viz. 37 °C, 25 °C, and 10 °C, labeled as scaffolds A, B, and C, respectively. Porosity assessment revealed consistent and substantial mean porosity percentage of 95.06 ± 2.01 % across all scaffolds. The lower temperature of 10 °C significantly contributed to a gradual and consistent cumulative release of 91.67 ± 1.47 %, compared to 37 °C (94.07 ± 1.89 %) and 25 °C (92.75 ± 1.51 %). The Korsmeyer–Peppas model exhibited high R2 values (0.98, 0.97 to 0.99) for scaffolds A, B, and C, indicating its suitability for describing the in–vitro release of AAS. Furthermore, the obtained release exponent values (0.60, 0.65, and 0.72 for scaffolds A, B, and C, respectively) suggest non–Fickian diffusion as the primary mechanism governing drug release. The release was also pH–dependent, with slower release at acidic (pH 3) and faster release under basic conditions (pH 8) due to varying swelling ratios. Evaluation of thermodynamic parameters viz. Ea, ΔG, ΔH and ΔS, using the rate constant of the Korsmeyer–Peppas model, revealed positive values of ΔH > 0 and ΔS > 0, indicating the prevalence of hydrophobic interactions in the release process.

Microwave-assisted synthesis of γ-alooh nanoflowers as an adsorbent for cadmium removal from domestic wastewater.

In this study, cadmium ions were effectively removed from domestic wastewaters using an adsorptive treatment strategy based on γ-AlOOH nanoflowers. A novel, rapid, and simple procedure was developed for the synthesis of the nanoflowers. Characterization studies were performed using X-ray powder diffraction patterns and scanning electron microscope images. The synthesized nanoflowers were utilized as adsorbent in the batch adsorption experiments. The influential parameters of the adsorption process were optimized, and a flame atomic absorption spectrophotometry (FAAS) system was used to determine maximum percent removal of cadmium ions. Matrix-matched calibration strategy, in which the calibration plot was developed in wastewater medium, was utilized for the accurate and precise quantification of cadmium in the effluent samples. The percentage removal efficiency values were calculated between 84 and 98% for different concentrations of cadmium ions in the wastewater samples. Equilibrium data was fitted to the four different linearization methods of the Langmuir isotherm model, as well as the Freundlich isotherm model and Elovich isotherm model. The best fitting was achieved for the Langmuir model with a high R2 value of 0.9956 and maximum adsorption capacity was calculated as 6.23mg/g.

Mechanical properties of aeolian sand cemented via microbially induced calcite precipitation (MICP)

The cementation of desert aeolian sand is a key method to control land desertification and dust storms, so an economical, green and durable process to reach the binding between sand grains needs to be searched. The method based on the microbially induced calcite precipitation (MICP) appeared in recent years as a promising process that proved its efficiency. The feasibility of the MICP technique to treat aeolian sand composed by low clay content, fine particles, low water content and characterized by weak permeability was demonstrated in the present paper. The effects of initial dry density, cementation number and curing time on the permeability and strength of MICP-treated aeolian sand were investigated using permeability tests and unconfined compressive strength (UCS) tests. The microstructure of aeolian sand was observed by scanning electron microscopy (SEM) tests and X-ray diffraction (XRD), aiming to reveal the solidification principle of MICP. The tests result indicated that when the initial dry density and the cementation number rose, the hydraulic conductivity of aeolian sand decreased while the mechanical strength given by UCS values improved. When the initial dry density was 1.65 g/cm3, the curing time was 3 h and the cementation number reached 20, the hydraulic conductivity and UCS reached 0.00151 cm/s and 1050.30 kPa, respectively. With increasing curing time, the hydraulic conductivity first decreased, followed by an increase, while the UCS exhibited an up and then a downtrend. Furthermore, the correlation between UCS values and the CaCO3 content reached a high R2 value equal to 0.912, which confirmed that the cementation occurred in sandy material and governed the soil strengthening. Indeed, the calcium carbonate crystals observed by SEM and XRD enhanced the friction between particles when they wrapped around the sand grains surface, while carbonates reduced the soil permeability when filling the pores and sticking the sand particles together. Finally, the theoretical and scientific knowledge brought by the present study should help in managing sand in desert areas.

Absolute quantitation of human serum cystatin C: candidate reference method by 15N-labeled recombinant protein isotope dilution UPLC-MS/MS.

Serum cystatin C (CysC) is a reliable and ideal endogenous marker for accurately assessing early changes in glomerular filtration rate (GFR), surpassing the limitations of creatinine-based estimated GFR. To improve the precision of GFR calculation, the development of strategies for accurately measuring serum CysC is crucial. In this study, the full-length CysC pure product and fully recombinant 15N-labeled CysC internal standard were subjected to protein cleavage. Subsequently, an LC-MS/MS method was developed for the absolute quantification ofserum CysC. The traceability of the method was assigned calibrator using the amino acid reference measurement procedure (RMP). It involved calibrating the instrument using an amino acid reference material with known amino acid concentrations for calibration and comparison purposes. The total imprecision of the method was determined to be≤8.2 %, and a lower functional limit of quantification (LLoQ) was achieved. The recoveries ranged from 97.36 to 103.26 %. The relative bias between this candidate RMP for measurement of ERM-DA471-IFCC and the target value was 1.74 %. The linearity response was observed within the concentration range of 0.21-10.13 mg/L, with a high R2 value of 0.999. The results obtained using our method was consistent with those obtained using other certified RMPs. With the establishment of this highly selective and accurate serum CysC measurement method, it is now possible to assess the correlation between immunoassay results of serum CysC and the intended target when discrepancies are suspected in the clinical setting.

Application of artificial intelligence and red-tailed hawk optimization for boosting biohydrogen production from microalgae

Enhancing biohydrogen production from microalgae is crucial in addressing environmental and energy challenges. It provides a sustainable, clean energy source while reducing greenhouse gas emissions. Moreover, it advances microalgae-based biotechnology, enabling innovative biofuel production and ecological revitalization. The main target of this study is to develop a robust ANFIS model to simulate the biohydrogen production process from microalgae within photobioreactors. The study focuses on enhancing hydrogen yield by optimizing three critical process parameters: sulfur concentration (%), run time (hours), and wet biomass concentration (g/L). Initially, an adaptive neuro-fuzzy inference system (ANFIS) model for biohydrogen production process is constructed based on empirical data. Subsequently, the red-tailed hawk algorithm (RTH) is used to determine the optimal values for the process parameters, corresponding to maximum hydrogen yield. The performance of ANFIS model in predicting hydrogen yield is assessed using root mean square error (RMSE) and coefficient-of-determination (R2) values. The obtained RMSE values for training and testing data are 2.8477 × 10−05 and 1.2807, respectively, while the corresponding R2 values are 1.0 and 0.9911 for training and testing. The introduction of fuzzy logic into the model significantly improves its predictive accuracy, as evidenced by the drop in RMSE from 10.79 with ANOVA to 0.7159 with ANFIS, representing a substantial 93.4 % decrease. The remarkable precision of the ANFIS model, indicated by its low RMSE and high R2 values, underscores the success of the modeling stage. The combination between ANFIS with the RTH technique yields impressive results, leading to a hydrogen yield enhancement of 6.87 % and 26.65 % when compared to both measured data and ANOVA.

Design and Testing of a Fruit Tree Variable Spray System Based on ExG-AABB

This paper addresses the issue of pesticide waste and low utilization rates resulting from traditional plant protection via spraying operations, which apply equal dosages to different targets or to different parts of the same target. To tackle this problem, we designed a variable fruit tree spraying system based on the ExG-AABB (excess green and axis-aligned bounding box) algorithm. We used a Kinect depth camera to capture information about the fruit tree canopy and constructed a spray flow model using pulse width modulation and variable spray control technology. Variable multi-nozzle spraying was guided by combining this canopy data. We evaluated the accuracy of each model in calculating canopy volume by comparing the coefficient of determination (R2) and root mean square error (RMSE) of the ExG-AABB with the slice convex hull method, voxel method, three-dimensional alpha-shape method, and QuickHull method. The ExG-AABB algorithm had the highest R2 value (0.9334) and the lowest RMSE value (0.0353 m3) among the five models, indicating that it most accurately reflects the true volume of the fruit tree canopy. This validates the effectiveness of the ExG-AABB algorithm in calculating canopy volume. We established a correlation model between canopy volume and spray volume, designed a canopy-adaptive layering method based on point cloud processing, and achieved precise calculation of nozzle flow. Comparative field experiments were conducted to analyze the spray coverage rate and observed flow, thereby evaluating the spraying effect of this variable spraying system. The experimental results showed that compared to conventional continuous spraying, this variable spraying system not only achieves more uniform spray coverage but also significantly reduces pesticide usage by 48.1%. Furthermore, through system optimization, the average coverage rate of the middle layer of the canopy decreased by 17.53%, effectively reducing the phenomenon of overlapping spraying from multiple nozzles and improving spraying efficiency.

PRICE ESTIMATION OF SELECTED GRAINS PRODUCTS BASED ON MACHINE LEARNING FOR AGRICULTURAL ECONOMIC DEVELOPMENT IN TÜRKİYE

This study aims to estimate the price fluctuations of essential grain products, namely bread wheat (Triticum aestivum), durum wheat (Triticum durum), barley (Hordeum vulgare), and corn (Zea mays), in Türkiye using machine learning (ML) algorithms. Using data from January 2, 2020, to January 10, 2023, the study employs algorithms such as random forest (RF), neural network (NN), support vector machine (SVM), and linear regression (LR). Independent variables include oil prices, currency exchange rates, and grain production volumes. The random forest (RF) algorithm provided the best results with the highest R² values, while NN and LR showed relatively lower performance. The study highlights the significant impact of production and consumption volumes on grain prices and underscores the importance of ML algorithms in predicting these prices amidst changing conditions. Investments in agricultural technologies should be increased to improve data collection and analysis processes, as this is crucial for preventing price fluctuations in the agricultural sector. Keywords: Agricultural products; grains; durum wheat; bread wheat; corn; barley; machine learning algorithms; price estimation

Optimization of bio-oil production parameters from the pyrolysis of elephant grass (Pennisetum purpureum) using response surface methodology

Abstract The need to increase bio-oil yield from biomass and enhance its fuel properties has driven research into optimizing the pyrolysis process. This study investigated the influence of three key process parameters—temperature, heating rate, and nitrogen flow rate—on the pyrolysis of elephant grass (Pennisetum purpureum) in a fixed-bed reactor. Response surface methodology was used to study the impact of the aforementioned variables on bio-oil yield to improve its production efficiency. Proximate analysis of the biomass revealed 79.24 wt% volatile matter, 14.22 wt% fixed carbon, and 5.86% ash, with ultimate analysis showing 45.44% carbon, 5.59% hydrogen, and 40.95% oxygen. The high volatile matter content and favourable carbon and hydrogen percentages indicate that elephant grass is a viable energy source due to its potential for high bio-oil yield and energy content. The resulting bio-oil exhibited a higher heating value of 20.9 MJ/kg, indicating its suitability for various heating applications. A second-order regression model was developed for bio-oil yield, with optimal conditions identified as a temperature of 550°C, a heating rate of 17°C/min, and a nitrogen flow rate of 6 ml/min. The study achieved an optimal bio-oil yield of 59.03 wt%, and the model’s high R² value of 0.8683 from analysis of variance analysis confirmed its predictive accuracy. This research highlights elephant grass as a sustainable feedstock for bio-oil production, offering valuable insights into optimizing pyrolysis conditions to enhance bio-oil yield, thus advancing biofuel technology.

RNN-Based Monthly Inflow Prediction for Dez Dam in Iran Considering the Effect of Wavelet Pre-Processing and Uncertainty Analysis

In recent years, deep learning (DL) methods, such as recurrent neural networks (RNN). have been used for streamflow prediction. In this study, the monthly inflow into the Dez Dam reservoir from 1955 to 2018 in southwestern Iran was simulated using various types of RNNs, including long short-term memory (LSTM), bidirectional long short-term memory (Bi-LSTM), gated recurrent unit (GRU), and stacked long short-term memory (Stacked LSTM). It was observed that considering flow discharge, temperature, and precipitation as inputs to the models yields the best results. Additionally, wavelet transform was employed to enhance the accuracy of the RNNs. Among the RNNs, the GRU model exhibited the best performance in simulating monthly streamflow without using wavelet transform, with RMSE, MAE, NSE, and R2 values of 0.061 m3/s, 0.038 m3/s, 0.556, and 0.642, respectively. Moreover, in the case of using wavelet transform, the Bi-LSTM model with db5 mother wavelet and decomposition level 5 was able to simulate the monthly streamflow with high accuracy, yielding RMSE, MAE, NSE, and R2 values of 0.014 m3/s, 0.008 m3/s, 0.9983, and 0.9981, respectively. Uncertainty analysis was conducted for the two mentioned superior models. To quantify the uncertainty, the concept of the 95 percent prediction uncertainty (95PPU) and the p-factor and r-factor criteria were utilized. For the GRU, the p-factor and r-factor values were 82% and 1.28, respectively. For the Bi-LSTM model, the p-factor and r-factor values were 94% and 1.06, respectively. The obtained p-factor and r-factor values for both models are within the acceptable and reliable range.