Torrefaction has emerged as a promising thermochemical pretreatment to enhance the fuel properties of agricultural biomass, yet predicting product characteristics remains challenging due to feedstock variability. This study developed and compared four machine learning (ML) models, support vector regression (SVR), artificial neural network (ANN), gradient boosting machine (GBM), and random forest (RF), to predict the higher heating value (HHV), mass yield (MY), and energy yield (EY) of torrefied biomass. A comprehensive dataset comprising 117 experimental samples from 11 biomass types torrefied at 200–280°C for 20–60 min was utilized. Input variables included the contents of proximate composition (ash, fixed carbon, and volatile matter), ultimate composition (C, H, N, O, S), and process conditions (temperature and residence time). SHapley Additive exPlanations (SHAP) analysis revealed that torrefaction temperature was the most influential parameter, followed by volatile matter and ash content. Results demonstrated that GBM consistently outperformed the other models, achieving an acceptably high coefficient of determination, with minimal prediction errors across different input configurations. Notably, simple proximate‐analysis data provided performance comparable to that of more detailed and instrument‐intensive ultimate analysis (UL), highlighting the practical value of GBM for operators handling diverse biomass feedstocks with limited access to specialized analytical instruments.
This study presents a 28‐case computational fluid dynamics (CFD) parametric investigation of Al 2 O 3 /water and Al 2 O 3 –TiO 2 /water hybrid nanofluid cooling for an LG Chem 15 Ah lithium‐ion pouch cell subjected to a prescribed steady thermal load of 277.5 W. The imposed load is used as a severe thermal‐stress condition for coolant comparison and is not interpreted as the electrochemical heat generated by the cell at a specific C‐rate. Four Reynolds numbers (Re = 2300–10,000) and three particle volume fractions (1%, 2%, and 4%) were examined in ANSYS Fluent. The Re = 2300 cases were solved as laminar, whereas Re = 4434, 7000, and 10,000 employed the realizable k–ε model with enhanced wall treatment. Nanofluid properties were estimated using a single‐phase effective‐property approach with Maxwell thermal conductivity and Brinkman viscosity correlations. The battery‐side thermal model was compared with FLIR thermography at 32 W under natural convection; this comparison supports the hotspot location and cell‐level temperature pattern but is not presented as forced‐convection nanofluid validation. Reynolds number remained the dominant control parameter. At Re = 2300, the 4% hybrid nanofluid reduced the maximum hotspot temperature by 8.1°C relative to water and increased the bulk‐temperature‐based heat‐transfer coefficient by ~13.1%. For the turbulent cases, all nanofluids produced PEC values below unity, while at Re = 2300 the PEC values were effectively unity (0.997–1.003). Above Re = 7000, the predicted temperature advantage of nanofluid addition decreased below 0.9°C. The results indicate that hybrid nanofluids primarily provide incremental hotspot reduction at low‐to‐moderate Reynolds numbers, whereas water remains the stronger thermal‐hydraulic baseline over the turbulent range.
Enhancing the electrochemical performance and structural durability of electrode materials is a critical frontier in energy storage research. Boron carbide (B 4 C), characterized by its exceptional chemical stability and mechanical robustness, emerges as a promising candidate for next‐generation supercapacitor electrodes. In this study, B 4 C was synthesized in two distinct morphologies equiaxed and whisker via a sol–gel method. Phase and morphological analyses via X‐ray diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM) verified the successful synthesis of crystalline B 4 C with different morphologies. Complementary electron spin resonance (ESR) spectroscopy analysis revealed a stronger paramagnetic defect‐related signal for the equiaxed particles than for the whisker‐type particles. The influence of these morphological and atomic‐scale variations on electrochemical behavior was systematically investigated in a three‐electrode configuration using 1 and 6 M potassium hydroxide (KOH) electrolytes. Cyclic voltammetry (CV) analysis revealed clear redox peaks, indicating a Faradaic contribution to the overall charge‐storage response. The 6 M KOH electrolyte produced an enhanced charge‐storage response, as evidenced by larger CV areas and longer discharge times and supported by the lower solution and charge‐transfer resistances obtained from electrochemical impedance spectroscopy (EIS). The whisker‐type B 4 C outperformed the equiaxed morphology, delivering a maximum specific capacitance of 348.3 F g −1 at 5 mV s −1 in 6 M KOH, surpassing the 285.56 F g −1 attained by its equiaxed counterpart. Furthermore, a symmetric supercapacitor device fabricated with whisker‐type B 4 C in 6 M KOH achieved an energy density of 0.376 Wh kg −1 and a power density of 407.64 W kg −1 . The device demonstrated good long‐term stability, retaining 97.7% of its initial capacitance after 10,000 cycles at 4 A g −1 , highlighting the potential of whisker‐type B 4 C for stable aqueous supercapacitor operation.
The International Maritime Organization (IMO) has strengthened regulations on ship emissions, increasing interest in eco‐friendly marine propulsion systems. Among these, the polymer electrolyte membrane fuel cell (PEMFC) is a promising alternative for small ships. This study aims to develop a system‐level high‐fidelity integrated fuel cell–battery hybrid small ship model and evaluate the performance of different power management strategies, focusing on energy efficiency and thermal management. A comprehensive model was developed using Matlab/Simulink and Matlab/Simscape, integrating a PEMFC stack, high‐voltage battery, power conversion units, motor, and thermal management system (TMS). Two power management systems (PMSs), rule‐based and fuzzy‐based, were implemented and analyzed under realistic operational profiles, including trim angle effects. Simulation results showed that the fuzzy‐based PMS reduced hydrogen consumption by 8.6%, improved overall energy efficiency by 17.8%, decreased overshoot by 21.6%, and shortened settling time by 121 s compared to the rule‐based PMS. TMS energy consumption was also reduced due to precise control of coolant pumps and valves. Qualitatively, the fuzzy‐based PMS stabilized battery state of charge (SOC) and mitigated fuel cell overload, demonstrating improved operational stability during dynamic ship operation.
This review critically synthesizes research on carbon‐based materials, including both nanomaterials and bulk forms, for enhancing solar still (SS) desalination. It is structured into four key application areas: carbon‐based absorber coatings and plates; carbon‐enhanced thermal energy storage in phase change materials (PCMs) and sensible media; carbon‐based nanofluids (NFs); and carbon‐based interfacial solar evaporators and wick structures. Significant performance improvements are documented, such as using a graphite absorber with glass‐cover cooling to achieve ~9.19 L/m 2 /day (107.7% enhancement), integrating 20 wt% graphite into paraffin PCM to achieve 8.52 L/m 2 /day with a daily energy efficiency of 65.13%, and employing carbon black NF in hybrid systems yielding 9.28 L/m 2 /day. Interfacial designs, like floating carbon nanotube (CNT) evaporators, further boost evaporation rates up to 4.64 kg/m 2 ·h. The review consolidates these advances, highlighting carbon’s role in improving productivity, efficiency, and cost‐effectiveness for sustainable solar‐powered desalination.
