The rapid evolution of edge computing and IIoT ecosystems demands lightweight machine learning models that deliver accurate predictions under resource constraints. Traditional classifiers, such as deep neural networks and decision trees, often struggle to balance accuracy, interpretability, and computational efficiency in such environments. This work introduces SparseBonsai, an enhanced variant of the Bonsai Tree algorithm that includes projection techniques, dynamic sparsity parameters, and adaptive regularization. Bonsai Tree models work with fixed parameters, whereas SparseBonsai dynamically adjusts sparsity and regularization during training. It improves adaptability and generalization. SparseBonsai achieves 86.91% classification accuracy with a model size of 0.048 MB, and inference time is less than 35% as of neural networks with a competitive accuracy. Model’s robustness and efficiency is evaluated using precision, recall, F1-score, and ROC-AUC. These results show that the SparseBonsai can be a practical solution for a real-time and resource-efficient fault detection system as an IIoT edge computing platform. The novelty of SparseBonsai is its dynamic adjustment of sparsity and regularization in training in comparison to the conventional Bonsai Tree algorithm. SparseBonsai reduces inference time by 35% and memory footprint by 38% compared with existing lightweight classifiers.

This study investigates the topical issues surrounding the modernisation of transport and technological machines (TTM) through the introduction of electric drives. This process represents an important step in improving the efficiency and sustainability of agricultural and transport engineering. In the context of the gradual obsolescence of the technical fleet and the growing need to reduce operating costs, modernising existing machines is becoming not only economically feasible but also necessary to maintain enterprise competitiveness. Attention is given to trends in TTM electrification, which contribute to reducing environmental damage and improving production performance. Methods and Results. This paper develops and applies a three-level optimisation framework for retrofitting TTM with electric drives. Power sizing determines an economic NHopt…NHK window that minimises present (reduced) costs with an admissible (≤5%) deviation for productivity enforcement. The configuration is selected around the chosen power plant (new/used components, transmission interfacing, cooling, and DC link under 360/600 V). Operational optimisation of Vopt (field speed) and Bopt (working width) is subject to agrotechnical and slip constraints, with power balancing between tools (NB) and traction (NT). Representative case studies yield NHopt windows of 85–125 kW for heavy tillage with KA≈11 kN/m and L=600–1000 m, and 55–85 kW for cereal seeding. Typical operational solutions are Vopt≈1.7–2.1 m·s⁻¹ and Bopt≈3.0–4.2 m, while maintaining the ≤5% cost-efficiency deviation: scope and Implications. The framework formalises modernisation decisions for electrified TTMs, consolidating morphology/set-theoretic analysis with multi-level economic–energetic criteria. It outputs site-specific power, configuration, and operating parameters that align with commercial e-tractor classes while preserving compatibility with legacy chassis and implements.

The production of heavy oil reserves has become increasingly important in recent years, as conventional oil sources become scarcer. However, extracting heavy oil is hindered by its high viscosity. One technique that has proven to be effective for heavy oil recovery is Steam-Assisted Gravity Drainage (SAGD). SAGD is a thermal recovery method that facilitates oil. In this article, we examine the main advantages of SAGD over other production methods, including higher RF, higher heating efficiency, higher production rate, lower steam-oil ratio, higher energy return on investment, better reservoir pressure maintenance, and lower environmental impact. Quantitatively, our calculations indicate that SAGD achieved a recovery factor of 39% versus 31% for CSS, with a steam–oil ratio of 2.56 versus 5.40 and an energy return on investment of 0.39 versus 0.20, respectively. Overall, these results demonstrate that SAGD outperforms CSS in recovery efficiency, energy utilisation, and environmental performance.

Researchers have developed a powerful new method to detect Alzheimer's disease (AD) by combining brain MRI scans with advanced artificial intelligence techniques. The study used sophisticated computer algorithms to analyse structural changes in the brain, particularly in regions such as the entorhinal cortex, parahippocampal area, and inferior temporal regions, which are known to be affected early in AD progression. By combining logistic regression and a support vector machine (SVM) in an ensemble learning (EL) approach, the researchers achieved remarkably high accuracy rates of 99% for distinguishing between healthy individuals and those with AD, 96% for detecting mild cognitive impairment (MCI), and 85% for identifying progression from MCI to AD. When the model was evaluated for multiclass classification distinguishing healthy controls, individuals with MCI, and patients with AD, it achieved an overall accuracy of 93%, demonstrating strong generalisation across all diagnostic categories. This reflects the efficacy of integrating structural MRI features with EL-based machine learning (ML) techniques, yielding a robust and interpretable diagnostic framework. Such an approach holds significant clinical promise, as it supports early and reliable detection of AD, potentially facilitating timely intervention and improved patient management.

The study investigates the thermal dissolution of Barzas coal using mechanochemical activation, ozone treatment, and ultrasonic processing to enhance the yield and quality of liquid products. The experimental work comprised preparing coal samples by high-energy ball milling, followed by ozone oxidation and ultrasonic pretreatment, and thermal dissolution in tetralin at 375-425 °c. The results showed that the integrated activation approach significantly increased organic matter conversion, reaching up to 88.3% at 400 °C with ultrasonic treatment, compared with 64.9% without activation. The yield of liquid fractions was 57.9% of the dry coal mass, and the light fractions exhibited a high calorific value of up to 41.2 MJ/kg, with aromatic hydrocarbon content exceeding 65%. Lower heating rates and controlled temperature conditions were essential to maximise conversion and minimise heavy residue formation. The findings highlight the effectiveness of combining mechanical, oxidative, and ultrasonic activation to improve coal dissolution and produce liquid hydrocarbons suitable for fuel components and chemical feedstock applications.

This study investigates how turbulence intensity and temperature gradients affect the macroscopic kinetics of hydrogen-air combustion at supercritical pressures from 10 to 20 MPa. A series of high-precision experiments was conducted in a custom-designed high-pressure combustion reactor with a controllable thermal field and adjustable turbulent-flow generation. The induction period for spontaneous ignition was measured under various turbulence regimes and temperature distributions. Increasing turbulence intensity from 0% to 20% reduced the induction period from 0.71 s to 0.39 s, and imposing a linear temperature gradient of 300 K over 15 cm further decreased it to 0.26 s. The maximum radical concentrations of H and OH were 3.5×10–⁴ and 3.9×10–⁴ mol/mol at 20 MPa and 1000 K, indicating enhanced chain-reaction activity in locally overheated regions. Numerical simulations using a modified global kinetic model deviated by less than 8% from the experimental data. The findings support the development of improved macrokinetic models suitable for engineering applications involving hydrogen combustion at high pressures with complex thermal and flow fields.

Construction waste management and inventory are lacking in the Jordanian construction sector. With the dramatic population growth, the construction sector plays a significant role in developing the Jordanian economy. This article aims to develop a model to quantify construction and demolition (C&D) waste based on either project hyper-categorization or waste type, and to measure the degree of awareness and readiness of local contractors to benefit from C&D waste. Therefore, it classified projects further into construction and infrastructure projects. The construction project is also classified as new construction, renovation, or rehabilitation. These classifications help improve the management of construction waste. Moreover, a survey was conducted among construction stakeholders to increase awareness and assess their perceptions and readiness to adopt C&D waste management activities in their current or future projects. Finally, a strengths, weaknesses, opportunities, and threats (SWOT) analysis was conducted to heighten major stakeholders' awareness about the challenges of C&D and to develop an effective C&D waste management plan. The results of this research can help both decision-makers and stakeholders in the construction sector properly classify their waste, reduce and manage generated C&D waste, and lower overall costs. Finally, the SWOT analysis identifies hidden opportunities to promote and develop a future C&D waste management at all levels. Meanwhile, revenue from recycling C&D waste is expected to be about US$4.2/m2 of construction.

Photovoltaic (PV) systems have gained prominence as critical tools for climate change mitigation and carbon emission reduction. These systems utilize static converters and proportional-integral controllers to regulate active and reactive power, while energy storage batteries enhance power quality by stabilizing electrical parameters (current/voltage). This study investigates a hybrid photovoltaic/thermal (PVT) system that co-generates electricity and heat using dual heat-transfer fluids with integrated thermal storage. Outdoor validation demonstrated that the hybrid cooling collector significantly reduced panel operating temperatures, achieving a maximum electrical efficiency of 15.71% in February—representing a 22% increase over conventional PV panels. Under balanced air-water flow conditions, the system attained 69.25% heat recovery efficiency and 84.40% total thermal efficiency. Deployed across four sites, PVT systems met 60% of residential heating demands and, when integrated with absorption chillers, 100% of cooling requirements. With installation costs 30–40% lower than PV-only equivalents, these systems offer a cost-effective solution for high-irradiance regions. Key factors for solar forecasting model development are identified, and findings provide actionable guidance for designing integrated energy management systems. The study positions PVT technology as a sustainable alternative for solar-rich regions, augmented by AI-driven optimization for adaptive energy networks. Future research priorities include cross-regional validation, advanced materials development, and policy frameworks for scaled deployment, collectively advancing the global transition toward renewable energy resilience.

Concrete-filled steel tubular (CFST) technology offers numerous structural benefits, including outstanding strength, fire resistance, and a considerable capacity for energy absorption. It has become crucial to investigate the ductility and stiffness of this type of structural system to gain an understanding of its capability to withstand external loads and prevent unexpected failure. This study provides insights into optimizing the design of CFST frames by offering a deeper understanding of how frame geometry affects their behavior. A finite element (FE) model was developed using specimen data from a previous study to analyze the behavior of the composite frame. The results of the FEA model were verified with the experimental data. An extensive analysis of the stress concentration in the CFST column, especially in the connecting portions, was conducted. Parametric studies were conducted to evaluate the impact of the axial load level, column length (L), and steel tube thickness (t) on the lateral stiffness and ductility behavior of CFST frames. For all frame models, no buckling occurred, and the failure mode was noticed after the formation of plastic hinges at the beam-column and base plate-column connections. The findings demonstrated that the lateral ductility of the CFST column improved with increasing L/D and D/t up to L/D equal to 15.53, at which point the development of lateral ductility was only slightly impacted by further increasing D/t. For all L/D ratios, the increase in skin thickness of the column tubes improved the stiffness of the frame; however, this improvement generally decreased noticeably as the L/D ratios increased. Increasing the length of the column with a small steel tube thickness (less than 2.67mm) insignificantly impacted the frame's ductility. On the other hand, increasing the column's length negatively impacted the frame's hardness in the case of a small steel tube thickness, i.e., less than 2.67mm.

The escalating threat of climate change demands urgent action to address rising levels of atmospheric CO2. This article reviews reactive CO2 removal and utilization techniques, including chemical, electrochemical, mineralization, carbonation, photo-catalysis, biological, and membrane separation methods. It examines their classification, mechanisms, technological advancements, integration with renewable energy, and related economic and policy frameworks, providing insights for researchers, policymakers, and industry experts. The review aims to offer valuable insights to researchers, policymakers, and industry specialists in the field of climate change, providing a detailed summary of the current state of CDR and CDU technologies. It evaluates the mechanisms and processes involved in CO2 capture, highlighting recent technological innovations and their integration with renewable energy sources to ensure sustainability. The economic and policy dimensions of these technologies are also examined, emphasizing the need for supportive regulatory frameworks and financial incentives to foster widespread adoption. Challenges related to scaling up CO2 capture and utilization are thoroughly discussed, including issues of cost, scalability, and technological barriers. The review highlights the immense energy requirements of large-scale CO2 operations and the importance of finding sustainable energy solutions to avoid counterproductive emissions. It also addresses the role of public awareness and community participation in advancing these technologies and achieving tangible climate benefits. The study concludes by highlighting recent progress in reducing costs, enhancing effectiveness, and increasing investment in carbon dioxide removal (CDR) and carbon dioxide utilization (CDU) technologies. It also identifies areas needing further research, such as cost reduction, efficient solutions, and environmental impacts. This review provides a thorough assessment of both opportunities and challenges, advancing CDR and CDU technologies toward a sustainable future.