Background: The utilisation of solar radiation for generating thermal energy has garnered a significant amount of attention, especially with the rising demand for sustainable power and heating sources. Nanofluids appeared as promising options for enhancing the efficiency of solar-thermal systems owing to their superior heat transmission properties. In this aspect, the present research deals with Ree-Eyring hybrid nanofluid ( SWCNT − MWCNT / H 2 O ) flow in the presence of thermal radiation effect and shape factor to examine the thermal behaviour in solar panels. Additionally, the innovation lies in the field of artificial neural networks and fluid flow analysis as an integrated approach. Methodology: The considered physical interpretation is stated in the form of nonlinear partial differential equations. To facilitate the computation process, the governing equations turned into nondimensional ordinary differential equations with the help of invariant transformations. The converted equations are treated scientifically using the NDSolve methodology, which automatically adjusts the step size. In further, an artificial neural network-based multi-layer perceptron deploying the Levenberg–Marquardt model is employed to forecast the measurements of physical quantities. Core Findings: The markable outcomes from this advanced work imply that the stretching surface has an important role in analysing the thickness of the boundary layer. The improving Weissenberg number enhances the boundary layer thickness, leading to a magnification in the momentum field of the non-Newtonian fluid. In addition, thermal radiation is captured as a second significant influence on temperature distribution. It follows that enhancing the Radiation parameter upsurges the thermal layer. The variation of the volume fraction of nanoparticles resulted in higher temperatures due to the Hamilton–Crosser model, which involves the cylindrical shape of the nanoparticles. Validation: The validation of the current work is explored by comparing it with the existing literature in certain instances. The incorporation of multi-layer perceptron for hybrid nanofluid delivers the performance capacity of high prediction with the error 4.83E−06, 1.11E−10 and 2.4E−08. Applications: The existing optimisation approach provides a new perspective on solar-powered offshore, solar vehicles, solar-powered charging stations, solar power greenhouse and solar water pump implementation.

A power control method for an HVDC system integrating a DFIG-based offshore wind farm supplying a weak onshore grid is the prime focus of this work. The real power generated by the wind farm is regulated by the LCC-based rectifier station and transmitted to the VSC-based inverter station through the HVDC line. The work proposes a control mechanism to integrate these two stations with better power quality management. The inverter station facilitates the effective transmission of generated power to the onshore grid under nominal onshore grid voltages. Nevertheless, the abnormalities in onshore grid voltage lead to converter failures due to huge harmonic grid currents and power pulsations beyond permissible limits; uninterrupted power transfer without compromising power quality is the target. The proposed method utilises a dynamic current control technique at the inverter station to inject compensating current. This enables independent regulation of active and reactive power, mitigating power oscillations and minimising grid current THD induced by onshore grid voltage deviations. The simulations in PSCAD/EMTDC and hardware-in-loop (HIL) experiments using OPAL-RT demonstrate that the proposed control strategy significantly reduces power oscillations while maintaining a grid current THD of 1.47% up to 50% unbalance in onshore grid voltage.

Seawater distillation often relies on fossil fuels, emitting greenhouse gases, but solar energy offers a sustainable, cost-free alternative. This study enhances solar still productivity by incorporating a condensation chamber and thermoelectric cooling using Peltier modules and fans. Two models were tested in Khemis Miliana, Algeria: a conventional solar still and a modified version with a cooling system. Experimental results from September to November showed the cooling system increased productivity by 74% compared to the uncooled system and 14% compared to the conventional still. The optimal cooling system operation was determined to be 3 minutes every 30 minutes, influenced by solar radiation. The modified system achieved a water production cost of $0.09237 per litre, with a payback time of 2.5 years. Daily output data helped evaluate CO2 emissions, mitigation potential and carbon credits over the system's lifespan. Future work includes integrating a flat plate collector and optimising the cooling system's operation relative to solar radiation for higher efficiency.

The estimation of global horizontal irradiance (GHI) is crucial for assessing solar energy potential, especially for investment purposes in specific regions. This study employs two feature selection techniques such as recursive feature elimination (RFE) and least absolute shrinkage and selection operator (LASSO) to identify key variables from two datasets, which are then used to train four machine learning (ML) models such as Decision Tree (DT), Random Forest (RF), Extreme Gradient Boost (XGB), and Extra Trees (ET) regressors. The performance of these models is evaluated using three statistical metrics such as mean absolute error (MAE), root mean squared error (RMSE), and R-squared (R 2). The results show that the ET regressor, when combined with LASSO, achieves the best predictive performance, with an MAE of 1.36 W/m² and an RMSE of 2.46 W/m2. The study further employs Shapley Additive Explanations (SHAP) to interpret the model, revealing that parameters like Diffuse Horizontal Irradiance, Solar Zenith Angle, and Direct Normal Irradiance significantly impact GHI prediction accuracy. The combination of feature selection, advanced ML models, and SHAP analysis offers a comprehensive and transparent framework for solar energy resource assessment, addressing the need for accuracy and interpretability in GHI estimation.

The post-harvest loss of agricultural commodities is a matter of universal concern. So, establishing cold storage near the production site is the need of the hour. Thus, the present work proposes a solar-biomass-powered absorption cold storage scheme for potatoes, considering the climatic conditions of developing country like India. The system considers that the primary energy requirement of the generator comes from an array of Evacuated Tube Collectors. The additional energy requirement is met using a biomass pellet burner for the continuous operation in off-grid mode. The capacity of the plant is determined to be about 70 kW. The effect of various performance parameters on the energetic and exergetic performance of the plant has been examined. An Artificial Neural Network (ANN) model is developed to forecast the system’s performance metrics, such as cooling capacity, Coefficient of Performance (COP), and refrigerant mass flow rate. The ANN model shows a coefficient of determination (R) of 0.99, establishing that the ANN model accurately predicts the output parameters. A genetic algorithm allows a population of many people to evolve to a condition that optimises ‘fitness’. Here, optimisation study employing GA indicates that the COP of the plant is 0.73, at the maximised operating condition.

Combined infrared and hot air drying (IRHAD) are widely used due to their affordability, simplicity, and efficient heat dispersion. This work aims to decrease the moisture content and the colour difference (ΔE), influencing the quality of turmeric drying. The drying optimisation conditions were determined by the WASPAS technique based on RSM, with an entropy weight that had the highest Q value of 0.8999 and a coefficient (R 2) of 0.9396, suggesting a robust correlation. The results were a temperature of 70 °C, an air velocity of 1.5 m/s, a drying time of 137 minutes, and an infrared power of 1500 watts. This combination leads to a moisture content of 3.14% and ΔE of 20.43. Meanwhile, the MAPE values are at an excellent level of 3.640 for moisture content and 1.422 for ΔE. These conditions can be used to improve product quality and long-term shelf life.

This work examines the attributes of a magnetohydrodynamic natural convective numerical analysis in a hexagonal cavity due to applications in thermal engineering. The bottom and inclined walls are treated as cold exteriors, the middle circular surface has a heat source, and other walls are insulated. The entire cavity contains an Ag-MgO-water hybrid nanofluid. The governing equations are simulated by the finite element method. Exploiting streamlines, isotherms and line graphs, the results are physically explained for Rayleigh number ( 10 3 ≤ Ra ≤ 10 6 ), nanoparticle volume fraction ( 0 ≤ ϕ ≤ 0.06 ), and Hartmann number ( 0 ≤ Ha ≤ 100 ). Using response surface methodology (RSM), a best-fitted correlation is built to examine the rate of sensitivity. Increasing the Ra value and adding hybrid nanoparticles results in thermal actuation. The use of Ag-MgO-H2O raises the water's heat transmission rate (Nuav) to 11%. Also, when the magnetic field has a direct influence, it reduces the Nuav by more than 25%. The Ra and ϕ have positive sensitivity on Nuav, but Ha is inversely sensitive. The numerical and statistical analysis in such an enclosure filled in hybrid nanofluid is conducted for the very first time. Lastly, the findings might guide the design of a successful mechanical device.

This paper deals with a detailed investigation into the boundary layer flow of a kerosene-based ferro-nanofluid (Fe 3 O 4) thin film over an unsteady expandable sheet, taking into account the effects of Arrhenius activation energy, aligned magnetic fields, and nonlinear thermal radiation. Also, the role of chemical reactions, non-uniform heat source/sink and convective boundary conditions are thoroughly explored. The conservation of momentum, energy, and mass diffusion equations seem to be complex and nonlinear, making it difficult to solve with some conventional analytical methods. Arrhenius activation energy provides deeper insight into chemical kinetics at the liquid–solid interface, which is relevant in catalytic processes where surface reactions greatly depend on the system’s performance. The study employs numerical bvp4c solver based on the three-stage Lobatto-IIIA method, given the limitations of conventional analytical methods for this problem. It is observed that as the volume fraction of ferroparticles increases, velocity and temperature gradients are enhanced, indicating improved heat and mass transfer characteristics. The increase in thermal radiation leads to a reduction in the temperature gradient. Further, the temperature gradient rises with an increment in the non-uniform heat source/sink parameter and Prandtl number. The rate of mass transport was enhanced by the rise in the Schmidt number.