High R-square Research Articles

Predictive analysis of fuel oil consumption in vessels: interpretable modelling with emphasis on load conditions

ABSTRACT Fuel oil consumption (FOC) in vessels is influenced by various factors, with vessel load conditions being a critical determinant. This research aimed to develop an interpretable regression-based machine learning model, specifically an XGBoost Regressor, to predict FOC, and leveraged the analysis with Explainable Artificial Intelligence (XAI) techniques to enhance transparency and understanding of the factors affecting FOC in maritime operations. Hyperparameter tuning optimized model parameters, achieving a high R-squared (R2) value of 0.99. XAI methods clarified how operational (e.g., speed, load, draft) and environmental factors (e.g., wind, wave, current, sea state) contribute to FOC increases. As a result, operational factors, notably average draft, exerted a substantial influence, with an average draft of 18.425 m resulting in a significant increase in predicted FOC to 1,729.40 kg/h during the laden condition, compared to the base value of 1,496.82 kg/h. Additionally, environmental factors, particularly Relative Wind Angle, significantly impacted FOC prediction, with a Mean Absolute SHAP Value of 11.97 kg/h, notably higher than in ballast (2.59 kg/h) and empty (2.85 kg/h) conditions. These findings emphasize the importance of tailored fuel efficiency strategies, as operational and environmental factors may vary in their impact across different loading conditions. ‘This article is a revised and expanded version of a paper entitled “Predictive Analysis of Fuel Oil Consumption in Vessels: Interpretable Modeling with Emphasis on Load Conditions” presented at The 11th International Conference on Logistics and Maritime Systems (LOGMS 2023) on 6 September 2023 at the Busan Port International Exhibition & Convention Center (BPEX)’

Effective thermal conductivity of composites using homogenization

Composite materials are extensively used in engineering applications due to their customizable properties and high performance. Determining the equivalent homogenized properties of composites, such as thermal conductivity, is crucial for their effective use. Various theoretical, analytical, and experimental methods have been developed to assess these properties. This study investigates the effective thermal conductivity of composites using a deterministically based procedure for thermal analysis. This procedure accounts for the combined influences of the inclusions’ volume fractions, shapes, orientations, and locations within the matrix to determine the effective thermal conductivity of composites. The specific composite analyzed consists of a cubical PLA matrix with a single spherical or elliptical void inclusion with perfect interfaces. For that purpose, an analytical approach was developed, and MATLAB® code was created to calculate the effective thermal conductivity tensor. To benchmark the analytical results, comparisons were made against numerical finite element modeling (FEM) results conducted using ANSYS®; in addition, to previously reported analytical models from the literature. Corroboration was also obtained by comparing the results against experimental data from the literature. The accuracy of the proposed homogenization scheme was demonstrated by achieving a low mean absolute percentage error (MAPE) compared to FEM (2.88%) and to the experimental results (2.72% for void inclusion and 6.99% for filled inclusion). Additionally, a high R-squared (R2) value of 0.986 was achieved compared to FEM, and values of 0.97 and 0.998 were achieved compared to the experimental results for void and filled inclusions, respectively.

ENTREPRENEURIAL ORIENTATION AND SME SUCCESS: UNVEILING THE MEDIATING IMPACT OF AMBIDEXTERITY

This research seeks to assess how ambidexterity in exploration and exploitation mediates the relationship between entrepreneurial orientation and the performance of small and medium-sized enterprises (SMEs). Specifically, this study investigates the indirect influence of entrepreneurial orientation on business performance through the mediating roles of ambidexterity exploration and exploitation. A quantitative approach using structural equation modeling (SEM) is employed to examine the relationships among these variables. The research sample comprises 200 SMEs operating primarily in East Java, Indonesia. The findings demonstrate a significant mediating effect (p<0.05) of ambidexterity exploration and exploitation, which enhances the influence of entrepreneurial orientation on business performance. This indirect effect proves to be stronger than the direct impact of entrepreneurial orientation itself. The structural model shows a high R-square value of 68.6%, indicating a strong predictive capability. Additionally, the goodness-of-fit index reflects high model suitability (0.543 > 0.36), and the predictive power of the PLS model is considered medium. The study’s results underscore the importance of fostering both ambidexterity in exploration and exploitation alongside entrepreneurial orientation to optimize SME performance.

NIMG-26. GLIOBLASTOMA TISSUE INVASION MAPPING WITH NOVEL HYPOXIA-TARGETED BLOOD OXYGENATION-LEVEL DEPENDENT MRI

Abstract BACKGROUND Glioblastoma is a devastating incurable malignant primary brain tumor exhibiting an erratic growth pattern and brain tissue invasion exceeding the typically depicted contrast-enhancing lesion on MRI. This behaviour is strongly associated with the presence of aberrant vasculature. Advanced clinical neuroimaging techniques, however, have not provided convincing imaging markers to determine glioblastoma tissue infiltration and monitor treatment effects. We investigated a new imaging technique based on blood oxygenation-level dependent (BOLD) MRI contrast with precise hypoxic respiratory targeting in order to better determine tumor tissue invasion not restricted to the typically depicted focal glioblastoma lesion on contrast-enhanced MRI. METHODS A computer-controlled gas blender was used to induce transient and standardized hypoxic stimulus in isocapnic conditions during echo-planar-imaging acquisition in patients with brain tumors.8,9 Data were preprocessed using SPM and in-house written MATLAB scripts. The induced BOLD signal changes were used to calculate %BOLD signal change during hypoxic stimulus with respect to the baseline, contrast to noise (CNR) ratio and goodness-of-fit (Rsquared) for each voxel in the whole-brain and automatically segmented region of interest (contrast-enhancement, necrosis, edema) and overlayed using color-maps onto anatomical high resolution T1. RESULTS Hypoxic targeting during gas modulation could be performed with high reproducibility, caused no discomfort in patients with brain tumors and was able to induce strong signal change during hypoxic stimulus with high contrast-noise ratio and Rsquared. In particular, when compared with other tumors, BOLD hypoxia glioblastoma tissue invasion maps show very distinct hypoxic signal responses in some cases extending beyond the focal -contrast-enhancing- glioblastoma lesion. CONCLUSION The ability to map at voxel level BOLD hypoxia signal patterns may determine a novel therapeutic imaging marker based on structural vascular alterations in the tumour tissue infiltration zone before these become appreciable with standard neuroimaging. This may support a refinement and tailoring of tumor resection and improve follow-up after resection and during treatment i.e. chemoradiotherapy.

Data augmentation using SMOTE technique: Application for prediction of burst pressure of hydrocarbons pipeline using supervised machine learning models

Accurate burst pressure prediction is critical for ensuring oil and gas pipeline safety, guiding maintenance decisions, and lowering costs and risks. Traditional methods have limitations, including high experimental costs, conservative empirical models, and computationally expensive numerical algorithms. Machine learning (ML) models have supplanted traditional methods in recent years. However, small and imbalanced datasets are the big challenge to build a ML model that can generate more accurate results. Moreover, the lack of generalization in ML models trained on a dataset of pipelines with specific material grids prevents them from producing superior results on other pipeline types. First, FEA was used to make a dataset. Then, a new way to improve machine learning (ML) model generalization for burst pressure prediction is suggested: combine publicly available datasets of different pipeline specifications. In this combined dataset, some pipelines have a higher number of data samples, and some have fewer, which causes a class imbalance issue. The Synthetic Minority Oversampling Technique (SMOTE) technique was applied to address the issue of class imbalance. The performance of various ML models, Extra Trees (ET), Extreme Gradient Boosting (XGBR), Random Forest (RF), Light Gradient Boosting Machine (LGBM), and Decision Tree (DT), was evaluated to validate the model's prediction and generalization on pipelines of various material grids. Results show that all the selected ML models produced high R-squared, i.e., >0.95, on balanced data compared to the imbalance dataset. These results show that SMOTE-based augmentation is a beneficial way to fix dataset imbalance and make ML models better at predicting burst pressure in oil and gas pipelines.

Long-term AI prediction of ammonium levels in rivers using transformer and ensemble models

This study provides a cutting-edge machine learning approach to forecast ammonium (NH4+) levels in River Lee London. Ammonium concentrations were predicted over several time intervals using a complete dataset that includes temperature, turbidity, chlorophyll, dissolved oxygen, conductivity, and pH. Our technique captures the intricate connections between environmental conditions and ammonium concentrations using developed algorithms, including Temporal Fusion Transformer (TFT), Random Forest (RF) and Extreme Gradient Boosting (XGBoost) levels versus the important factors, considerably improving prediction accuracy. The novel aspect of this study is the utilisation of the TFT model for multi-horizon forecasting, which offers high accuracy and interpretability in hydrological predictions by combining convolutional components with an attention mechanism. The study also demonstrates the effectiveness of the TFT model in capturing short-term fluctuations while retaining accuracy over long time periods, which is a major difficulty in environmental modelling. The models used, have exceptional forecasting skills, predicting 150, 200, 365, 730, and 1095 days based on daily average and 12, 24 and 30 months based on monthly average. This dual-scale model combines flexibility and resilience, making it an effective tool for forecasting both short- and long-term environmental changes. The RF model excelled in long-term forecasts, sustaining high R-squared (R²) (0.97) values and low root mean square error (RMSE) (0.18), and the second best one was the XGBoost with optimiser with R2 of (0.92) and RMSE of (0.25) with forecasting 1095 days. The results also found that whilst the TFT captured the fluctuations in the short-term, it struggled with the longer-term predictions due to data granularity. The XGBoost model did remarkably well in monthly forecasts up to 12 months, maintaining low RSME. The findings also highlight the necessity of proactive water management techniques to reduce the risk of potential ecological effects, including hypoxia and oxygen depletion. The findings support resource managers in addressing prospective ammonium toxicity concerns such as oxygen depletion and ecological stress.

Performance improvement of THz MIMO antenna with graphene and prediction bandwidth through machine learning analysis for 6G application

This article provides the findings of a study that integrated simulation, an RLC equivalent circuit, and machine learning (ML) techniques to improve wireless indoor communications clusters with future 6 G applications. The antenna being presented is constructed on a polyimide substrate. It exhibits an isolation of 27 dB and has a bandwidth of 4.331 THz, ranging from 0.631 THz to 4.962 THz. Along with its small size (95.52 × 227.24) µm2, it boasts an impressive maximum gain of 13.3 dB and an efficiency rating of 95 %. The ECC value drops below 0.0002 when the DG goes over 9.99. An advanced design system (ADS) creates a model like the proposed MIMO antenna to compare the return loss caused by CST (Computer Simulation Technology). Subsequently, following extensive data sampling with CST MWS (Microwave Studio) simulation, we employed supervised regression ML techniques. Gaussian process regression demonstrates exceptional accuracy, reaching almost 99 %, as evidenced by the high R-square and var scores. Additionally, it achieves the lowest error, less than one, while predicting bandwidth. The proposed antenna demonstrates strong potential as a formidable contender for 6 G THz band applications, as evidenced by the outcomes of the CST simulations and the prognostications derived from the machine learning techniques.

Predicting smartphone prices using machine learning algorithms

This study investigates the prediction of smartphone prices using machine learning algorithms, motivated by the increasing demand for accurate pricing information in the highly competitive and rapidly evolving mobile phone market. With the proliferation of smartphones and the constant introduction of new models with varying specifications, it is crucial for consumers, retailers, and manufacturers to have a reliable method for estimating prices based on technical features. Accurate price prediction can assist consumers in making informed purchasing decisions, retailers in setting competitive prices, and manufacturers in optimizing production costs and pricing strategies. Utilizing Decision Tree Regression, Support Vector Regression (SVR), Random Forest Regression and Convolution Neural Network (CNN), the research investigates the efficacy of these methods in estimating prices accurately. Principal Component Analysis (PCA) is applied to reduce dimensionality and enhance model performance. Through rigorous experimentation and hyperparameter tuning, Random Forest Regression emerges as the most effective method, achieving the lowest Mean Squared Error (MSE) and the highest R-squared (R) score. The findings highlight the potential of ensemble learning techniques in addressing complex prediction tasks.

PENERAPAN REGRESI LINEAR UNTUK PREDIKSI SUHU BADAN SAPI MENGGUNAKAN DATA SMART TERNAK DARI PT TELKOM INDONESIA: STUDI KASUS DI DESA PENGENGAT

Abstract West Nusa Tenggara (NTB) is one of the regions in Indonesia with significant potential in the field of livestock farming. In the context of the agricultural industry, livestock is considered a valuable asset, and monitoring their health is crucial for improving livestock productivity. This research aims to determine the relationship between the body temperature of cattle (Y) and environmental factors, namely air humidity (), environmental temperature (), and barometric pressure (), using smart livestock data provided by PT Telkom. The study was conducted in Pengengat Village, NTB, involving correlation and regression analysis. The analysis results indicate a significant positive correlation between the body temperature of cattle and environmental temperature () and a negative correlation with humidity (). Additionally, there is a strong positive correlation between the body temperature of cattle and barometric pressure (). The multiple linear regression equation obtained is as follows: = –26.816842418191094 + -0,0014707 + 0.39772325 + 0.04927779. Model testing shows excellent prediction quality with low Mean Squared Error (MSE) values (0.3849 for training and 0.4623 for testing). Furthermore, the high R-squared (R^2) values (0.7283 for training and 0.7110 for testing) indicate that the linear regression model can predict the body temperature of cattle based on the analyzed environmental factors. Keywords: Cattle Body Temperature, Smart Livestock Data, Prediction, Machine Learning, Linear Regression

Impact of digital marketing and E-Payments on Saudi online startup consumers

This study explores the impact of digital marketing and payment systems on consumer behavior in Saudi Arabia's online startups. A sample of 291 participants was selected from Hail’s region, in Saudi Arabia. The research employs a mixed-method approach, using SPSS version 2025 for descriptive statistical analysis and Structural Equation Modeling with Partial Least Squares (SEM-PLS) to examine the relationships between the key constructs and their effects on online purchase outcomes (OP). The study tests nine hypotheses, revealing that Perceived Innovation (PI) has a weak positive effect on OP, but this relationship is not statistically significant. Similarly, Perceived Security (PS) shows a minimal impact on OP, suggesting that these factors may not be as influential in the context of digital payment adoption as initially expected. The model's high R-square values for intention to purchase online (IPO) (0.654) and OP (0.594) indicate that it successfully explains a substantial portion of the variance in these outcomes. These findings offer valuable insights into the factors driving digital payment adoption and online purchasing behavior in Saudi Arabia. The study underscores the need for businesses to consider regional and cultural contexts when developing digital marketing strategies for effectively promoting digital commerce in diverse markets.

Integrating fuzzy modelling and war strategy optimization for identifying optimal operating factors of direct ethanol fuel cell

-Operating factors, including temperature, pressure, alcohol concentration, and reactant flow rate, significantly impact how much energy is produced by direct alcohol fuel cells. The cell's performance may go down or up depending on how these factors vary. For instance, greater alcohol temperatures and concentrations can speed up reactions and boost fuel cell efficiency, whereas lower alcohol temperatures and concentrations can have the opposite effect. Consequently, optimizing the operating circumstances is essential to getting the most energy possible out of direct alcohol fuel cells. Therefore, the primary goal is to develop a solid fuzzy model to simulate direct ethanol fuel cells (DMFC). Three process variables—ethanol flow rate, ethanol concentration, and temperature—are considered to increase the power density of DEFC. First, a fuzzy model of the DEFC was developed using experimental data. The best operating conditions to increase power density are then determined using war strategy optimization (WSO). Thanks to fuzzy, the RMSE decreased from 0.529 using RSM to 0.0292 using fuzzy (decreased by 94.5 %), compared to 0.529 using RSM. By around 11.7 %, the squared-R for prediction increases from 0.88 (using RSM) to 0.9831 (using fuzzy). The fuzzy model's low RMSE and high R-square values show the successful modelling phase. Compared to measured data and RSM, integrating fuzzy and WSO increased DEFC's power density by 5 % and 7.26 %, respectively. The ideal ethanol flow rates, concentrations, and temperatures are 9.4 ml/min, 0.25 M, and 74C, respectively.

Accurate determination of alcohol-based diesels using optimal chemical factors

Sustainable environmental policies and energy crises have led to a trend of blending different alcohols into diesel to partly replace the decreasing fossil fuels. To improve the rapidity and accuracy of determining alcohols exist in methanol and ethanol diesel, optimal chemical factors (OCF) feature selection schemes were presented based on different near infrared (NIR) characteristic absorption bands generated by different chemical structure information utilizing support vector machine (SVM). Through comparative analysis with SVM based on entire spectra, Monte Carlo uninformative variable elimination (MC-UVE) spectra and competitive adaptive reweighted sampling (CARS) spectra, the proposed OCF-SVM not only achieved 100 % accuracy, precision, recall and F1-score in classification, but also exhibited the best performance in prediction analysis with the smallest sum of squares due to error (SSE), root mean squared error (RMSE), mean absolute percentage error (MAPE) and the highest R-square. The overall outcomes indicate that the OCF method based on molecular chemical structures can select more pertinent and interpretable spectral features, thereby making the classification and prediction of alcohol-based diesels more exact and credible. Further, the developed OCF strategy together with SVM could supplement existing methods for analysis of alcohol-based diesels and is expected to be recommended for detecting the chemical composition of other fuels.

DeepInvesting: Stock market predictions with a sequence-oriented BiLSTM stacked model – A dataset case study of AMZN

Intelligent forecasters are now being considered in the stock market, providing essential insights and strategic guidance to investors and traders by presenting analytical tools and predictive models, thus enabling informed decision-making and mitigating financial risks in this dynamic market. The importance of intelligent analyzers in stock trading routines is considered in this work, where DeepInvesting, a multimodal deep learning model tailored for stock price prediction, is introduced. Employing a Sequence-Oriented, Long-Term Dependent (SoLTD) architecture featuring Bidirectional Long Short-Term Memory (BiLSTM) networks, DeepInvesting is applied to essential features of the Amazon Corp. (AMZN) market dataset, gathered from Yahoo Finance, including Closing, Opening, High, Low, Volume, and Adj Close prices. Exceptional performance in forecasting Closing, Opening, High, Low, and Adj Close prices is demonstrated, with minimal Mean Absolute Percentage Error (MAPE) and Root Mean Squared Percentage Error (RMSPE) scores, coupled with high R-squared (R2) values, manifesting a robust fit to the data, as well as computational complexity, and Rates Per Second (RPS) metrics in comparison to other models of KNN, LSTM, RNN, CNN, and ANN. Finally, challenges in the accurate prediction of trading volumes are identified, highlighting an area for future enhancement.

Modeling of specific fuel consumption for compression ignition engines fueled with polymer-based fuel: a response surface methodology approach

Background. Diesel engines play a crucial role in ensuring human comfort and well-being across residential, commercial, transportation, and emergency response sectors because of their reliability and versatility. However, identifying alternative fuels remains a significant challenge. Objective. This study aims to develop a comprehensive mathematical model using Response Surface Methodology (RSM) to optimize the performance of Compression Ignition (CI) engines utilizing different types of plastic pyrolysis oil. Methods. Through systematic data collection and analysis, this study examines the importance of design parameters, specifically injection pressure, compression ratio, engine load, and type of plastic pyrolysis oil, which are important for specific fuel consumption. A prediction model was developed to identify the complex correlations between these factors and the fuel use. Results. The developed model serves as an effective tool for optimizing the CI engine performance under diverse operational conditions. Experimental validation involved testing diesel engines with conventional diesel fuel and various plastic pyrolysis oils, followed by optimization using RSM to achieve optimal engine performance. The engine load was identified as the most significant parameter affecting the specific fuel consumption, followed by the fuel type, injection pressure, and compression ratio. The high R-squared (99.35%) and adjusted R-squared (98.02%) values indicate that the proposed model effectively fits the experimental data. Conclusion. The RSM-based model effectively optimizes CI engine performance under varied operational conditions. It significantly reduces the time and effort required to optimize engine design variables, thus enhancing engine performance and sustainability.

Enhanced Rubber Yield Prediction in High-Density Plantation Areas Using a GIS and Machine Learning-Based Forest Classification and Regression Model

Rubber is a perennial plant grown for natural rubber production, which is used in various global products. Ensuring the sustainability of rubber cultivation is crucial for smallholder farmers and economic development. Accurately predicting rubber yields is necessary to maintain price stability. Remote sensing technology is a valuable tool for collecting spatial data on a large scale. However, for smaller plots of land owned by smallholder farmers, it is necessary to process productivity estimates from high-resolution satellite data that are accurate and reliable. This study examines the impact of spatial factors on rubber yield and evaluates the technical suitability of using grouping analysis with the forest classification and regression (FCR) method. We developed a high-density variable using spatial data from rubber plots in close proximity to each other. Our approach incorporates eight environmental variables (proximity to streamlines, proximity to main river, soil drainage, slope, aspect, NDWI, NDVI, and precipitation) using an FCR model and GIS. We obtained a dataset of 1951 rubber yield locations, which we split into a training set (60%) for model development and a validation set (40%) for assessment using area under the curve (AUC) analysis. The results of the alternative FCR models indicate that Model 1 performs the best. It achieved the lowest root mean square error (RMSE) value of 19.15 kg/ha, the highest R-squared (R2) value (FCR) of 0.787, and also the highest R2 (OLS) value of 0.642. The AUC scores for Model 1, Model 2, and Model 3 were 0.792, 0.764, and 0.732, respectively. Overall, Model 4 exhibited the highest performance according to the AUC scores, while Model 3 performed the poorest with the lowest AUC score. Based on these findings, it can be concluded that Model 1 is the most effective in predicting FCR compared to the other alternative models.

Numerical simulation of wormhole propagation in fractured carbonate rocks during acidizing using a simplified Stokes–Brinkman model

Most numerical simulations for modeling acid reactive fluid transport and wormhole propagation during matrix acidizing, waterflooding, and CO2 sequestration in carbonate formations are computationally expensive, limiting real-time reservoir management and deep learning training datasets generation for inverse modeling research. Therefore, there is a need for less computationally expensive acid-reactive fluid flow models with adequate accuracy. This study developed and validated a simplified acid reactive-transport model by integrating a simplified Stokes–Brinkman model (as opposed to Darcy’s law), an averaged continuum model, and a pseudo-fracture model. Using FEniCS, the model effectively simulates acid-reactive fluid transport and wormhole propagation in carbonate rocks, achieving a high R-square value of about 0.97 based on a quantitative comparison of the breakthrough volume with other models. The simplified model can also simulate wormhole propagation for the reciprocal of the Damköhler number (1/Da) ranging from 0.001 to 1 with adequate accuracy. Sensitivity studies on the natural fracture parameters such as orientation, length, width, and density showed that higher fracture density, wider fracture aperture, longer fracture length, and orientation aligned with the direction of acid injection contribute to lower pore volume to breakthrough ratio but may not increase long-term acid stimulation efficiency. Also, the presence or absence of fractures in the matrix does not alter the dissolving patterns and optimum injection rate. This simple acid reactive-transport model can generate large training datasets for developing surrogate models in deep learning research. Finally, the FEniCS code in this paper is shared so future researchers can reproduce the results or extend the research work.

ANALISIS SPASIAL BERBASIS PEMETAAN MENGGUNAKAN GEOGRAPHICALLY WEIGHTED REGRESSION (GWR) PADA SEBARAN JUMLAH WISMA DI KABUPATEN LOMBOK BARAT

This research aims to analyze the effect of population density, number of food stalls, and population on the number of homesteads using the Geographically Weighted Regression (GWR) model. GWR modeling is carried out based on spatial data using secondary data sourced from the Central Statistics Agency (BPS). First, multiple linear regression analysis was carried out to evaluate the relationship between variables. The results of multiple linear regression show that only the variable number of food stalls has a significant effect on the number of homesteads. Next, spatial aspects were tested, including spatial autocorrelation tests and spatial heterogeneity tests. Testing spatial aspects shows the existence of spatial autocorrelation and spatial heterogeneity in the data. Therefore, we continued with GWR modeling to obtain a more in-depth picture of the distribution of the number of homesteads in West Lombok Regency. The results of modeling with GWR show that the variables used have a spatially varying influence throughout the West Lombok Regency area. Mapping GWR results produces better modeling of the factors that influence the number of homesteads in each village. A comparison between the multiple linear regression model and the GWR shows that the GWR model is able to explain variations in the number of homesteads with a higher R-square and a smaller AIC value

Comparative studies of machine learning models for predicting higher heating values of biomass

This study addresses the challenge of efficiently determining the higher heating value (HHV) of biomass, a crucial parameter in large-scale biomass-based energy systems. The conventional method of measuring HHV using an oxygen bomb calorimeter is time-consuming, expensive, and less accessible to researchers, particularly in developing nations. To overcome these limitations, we employed four machine learning (ML) models, namely Random Forest (RF), Decision Tree (DT), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost). These models were developed by using proximate and ultimate analysis parameters as input features. Up to 200 datasets were compiled from literature and used for the ML models. Our results demonstrate the effectiveness of all ML models in accurately predicting the HHV of biomass materials. Notably, the XGBoost model exhibited superior performance with the highest R-squared (R2) values for both training (0.9683) and test datasets (0.7309), along with the lowest root mean squared error (RSME) of 0.3558. Key influential input features identified for HHV prediction include carbon (C), volatile matter (Vm), ash, and hydrogen (H). Consequently, this research provides a reliable alternative for predicting HHV without the need for costly and time-intensive experimental measurements, facilitating broader accessibility in biomass energy research.

IoT-based monitoring system and air quality prediction using machine learning for a healthy environment in Cameroon.

This paper is aimed at developing an air quality monitoring system using machine learning (ML), Internet of Things (IoT), and other elements to predict the level of particulate matter and gases in the air based on the air quality index (AQI). It is an air quality assessor and therefore a means of achieving the Sustainable Development Goals (SDGs), in particular, SDG 3.9 (substantial reduction of the health impacts of hazardous substances) and SDG 11.6 (reduction of negative impacts on cities and populations). AQI quantifies and informs the public about air pollutants and their adverse effects on public health. The proposed air quality monitoring device is low-cost and operates in real-time. It consists of a hardware unit that detects various pollutants to assess air quality as well as other airborne particles such as carbon dioxide (CO2), methane (CH4), volatile organic compounds (VOCs), nitrogen dioxide (NO2), carbon monoxide (CO), and particulate matter with an aerodynamic diameter of 2.5 microns or less (PM2.5). To predict air quality, the device was deployed from November 1, 2022, to February 4, 2023, in certain bauxite-rich areas of Adamawa and certain volcanic sites in western Cameroon. Therefore, machine learning algorithm models, namely, multiple linear regression (MLR), support vector regression (SVR), random forest regression (RFR), XGBoost (XGB), and K-nearest neighbors (KNN) were applied to analyze the collected concentrations and predict the future state of air quality. The performance of these models was evaluated using mean absolute error (MAE), coefficient of determination (R-square), and root mean square error (RMSE). The obtained data in this study show that these pollutants are present in selected localities albeit to different extents. Moreover, the AQI values obtained range from 10 to 530, with a mean of 132.380 ± 63.705, corresponding to moderate air quality state but may induce an adverse effect on sensitive members of the population. This study revealed that XGB regression performed better in air quality forecasting with the highest R-squared (test score of 0.9991 and train score of 0.9999) and lowest RMSE (test score of 1.5748 and train score of 0. 0073) and MAE (test score of 0.0872 and train score of 0.0020), while the KNN model had the worst prediction (lowest R-squared and highest RMSE and MAE). This embryonic work is a prototype for projects in Cameroon as measurements are underway for a national spread over a longer period of time.

Thermal degradation assessment, impact strength, and hardness of combination epoxy and polystyrene powder composite

The present study investigates the reactions of polystyrene and epoxy/polystyrene powder composites, exploring different weight percentages (8%, 12%, and 16%) of polystyrene during decomposition under highly elevated temperatures within a nitrogen atmosphere. The findings reveal that polystyrene content significantly influences the thermal behavior of pure epoxy, with a higher polystyrene content correlating with the increased thermal stability of the composites. This study observed a three-stage mass loss process by examining the reaction kinetics of thermal decomposition using the Coats-Redfern approach. Initially, weight reduction occurs due to moisture and volatile elimination, followed by a smooth mass loss phase attributed to weaker polymer chain linkages and, ultimately, char formation. Chemical kinetic parameters, such as activation energy and frequency factor, were successfully determined using the Coats-Redfern approach, with a first-order reaction observed across all data points, demonstrating a high R-square mean value of 99.8%. Differential scanning calorimetry (DSC) and thermal gravitational analysis (TGA) analyses revealed an increase in glass temperature from 86.95 to 92.85 °C and an elevation in activation energy from 75.38 to 92.85 kJ/mol with increasing polystyrene content (0–16 wt%). Furthermore, the study investigated the influence of thermodynamic parameters. In terms of mechanical properties, increasing polystyrene content reduced impact strength (from 8.5 to 4.23 Kj/m2). In contrast, hardness increased from 77.5 for pure epoxy to 80.4 at 12 wt% polystyrene content in the composite. These findings underscore the importance of thermal stability results in developing a comprehensive understanding of thermal decomposition processes, informing future research endeavors and industrial applications.