National Engineering College

Improving the thermal energy conversion efficiency of solar stills is still a key challenge to accelerating clean and sustainable desalination technologies to combat worldwide water scarcity. In this paper, an innovative Twin Wedge Solar Still (TWSS) design is experimentally investigated with two new modifications: (i) a graphene nanoplatelet (GNP)-coated absorber plate to increase solar absorption, and (ii) a mixed nano-composite phase change material (nPCM) based on aluminium oxide (Al2O3) and graphene oxide (GO) for enhanced thermal energy storage. This coupled combination has not been used previously for TWSS applications. The new system exhibits improved performance by enhancing solar absorptivity, thermal conductivity, and storage capacity, resulting in a cumulative productivity of 6.133 L/m2/day. The developed modified system shows 139.4 %, 153.2 %, and 230.7 % greater productivity, energy efficiency, and exergy efficiency compared to the existing TWSS. The evaporative heat transfer rate becomes almost double (235.89 W/m2K compared to 99.15 W/m2K), and the economic cost of distilled water decreases to $0.012/L from $0.024/L for the traditional system. The outcomes verify that the integration of the proposed GNP-coated absorber and GO-Al2O3-based nPCM provides a new and economical path to enhance solar desalination performance, making it a promising strategy for sustainable freshwater production.

The need for freshwater increases rapidly. Solar desalination is one of the economic methods to cater the freshwater requirement. Solar still is the device that can do solar desalination, but the problem with it is its low productivity. In this study, a cylindrical rotating drum is incorporated into an in-built condenser solar still (ICSSRD) and compared with a conventional solar still (CSS). A 5E analyses were carried out to compare the stills' performance, viz., condensate yield, exergy efficiency, energy efficiency, exergy destruction, economic and environmental analysis. ICSSRD gave a yield of 4.280 L/m2 when compared to the CSS yield of 2.320 L/m2. Theoretical performance was compared with the experimental investigation. The energy efficiency of ICSSRD was computed to be 18.67 % more than the CSS, whereas the exergy efficiency of ICSSRD showed an improvement of 1.21 % over the CSS. The exergy destruction vested in ICSSRD's basin was less than the CSS, as the calculated magnitude was 2997.63 W/m2, and 3510.70 W/m2, respectively. The economic analysis showed that the modified still had better payback of 1.32 years over the conventional counterpart which was calculated as 2.3 years. The environmental analysis included the total CO2 emission mitigation (TCEM) and total carbon credit earned (TCCE), which also appraised the performance of ICSSRD over CSS. 1.54 and 2.87 t was the TCEM for CSS and ICSSRD. $28.16 and $52.65 is the TCCE for the corresponding stills. Simultaneous augmentation of condensation and evaporation due to the improvement techniques was the reason for performance improvement in the modified still. Experimental values of basin water, glass cover and basin temperatures of both the stills were compared with the theoretically predicted values. The error analysis showed that the difference in the basin water temperature of CSS and ICSSRD was 5.35 % and 7.44 % respectively. 6.78 % was the error difference for both stills in the case of glass cover. Error difference was calculated to be 5.25 % and 7.32 % for the basin of CSS and ICSSRD, respectively.

In resource-constrained settings such as rural clinics, the lack of access to specialized radiologists and advanced diagnostic tools often results in delayed or inaccurate medical diagnoses. Existing deep learning-based medical image classification models are generally tailored to specific imaging modalities such as X-rays, MRIs, or dermoscopic images necessitating multiple models for different diagnostic tasks. This fragmented approach is not only computationally inefficient but also difficult to scale, particularly when working with limited datasets. Additionally, the “black-box” nature of deepest learning models undermines clinical trust, as their predictions often lack interpretable justifications. To address these challenges, we propose a unified, robust, and interpretable framework based on Vision Transformers (ViT), capable of classifying multiple medical conditions across diverse imaging modalities with high accuracy and transparency. The model utilizes a multi-task learning formulation to perform simultaneous classification across four key diagnostic categories: Bone Fractures in X-rays, Brain Tumours in MRI scans, Lung Cancer in chest imaging, and Skin Lesions in clinical photographs spanning a total of 20 distinct classes. The ViT architecture, with its self-attention mechanism, mimics the behaviour of expert radiologists by selectively focusing on the most relevant patches in an image. This enables the model to learn intricate and domain-agnostic features that generalize well across different imaging types. The framework is further strengthened through data augmentation and domain-adaptive fine-tuning strategies. To enhance clinical acceptance, the system incorporates Explainable AI (XAI) using Grad-CAM techniques, providing visual justifications for its predictions and enabling clinicians to validate the AI’s decision-making process. Our model achieved a classification accuracy of 98.7% and a validation accuracy of 92%, demonstrating its reliability and potential to support timely and accurate diagnosis in under-resourced healthcare environments. This work represents a significant step toward democratizing AI-powered healthcare with improved trust, early detection, and global accessibility.

Solar still (SS) desalination is an environmentally friendly method of freshwater generation; nevertheless, there are limitations to the production effectiveness and thermal efficiency of SS. To overcome such problems, the study experimentally analyzed the effectiveness of the Twin Wedge Solar Still (TWSS), further integrated with Al2O3-graphene oxide-based nano phase change material (nano-PCM) and natural wick materials. Based on experimentation, the jute wick and Cotton wick/nano-PCMs and bamboo wicks/nano-PCMs combinations are analyzed to study the most effective combination of wick and nano-PCMs material to have better evaporation and higher retained heat properties. For comparison purposes of different models of TWSS, an energy-exergyeconomical-environmental (4E) investigation has been carried out. Experimental study reveals that the TWSS wick material made of jute and nano-PCMs results in maximum effectiveness of all combinations. The total freshwater production was measured at 6.04 L/day, and there was an increased production value of 133.2% compared to the original SS desalination device. The net average energy and exergy efficiency of 52.5% and 4.1%, respectively, increased in value by 128.2% and 215.3%, respectively, compared to the original setup's efficiency value. The highest exergy-based sustainability value of 1.042 was measured. Economic analysis shows that the optimized design produces the lowest cost of distilled water at $0.0124 per liter with a payback time of 4.6 months. Moreover, from the environmental analysis, it is clear that the lifetime mitigation potential of CO2 is 28.14 tons. The conclusion drawn from the analysis validates that the complementing role of nano-PCM and jute wick in TWSS improves the performance of TWSS in terms of thermal, economic, as well as environmental aspects.

One of the major problems which the world faces today is water scarcity, as every life needs it for their day-to-day activity. Pollution and salinity are the two major causes of most waterbodies being unfit for use. Removing pollutants or salt from water became essential for overcoming the water shortage. Desalination technology is being used to achieve this, but the major disadvantage is its lower production. Applying solar energy and nanofluids may tackle the problem and increase the production. In this work, some physical and chemical parameters, in particular isoelectric point (pH), total dissolved solids (TDS), electric conductivity (EC), chloride (Cl–), fluoride (F–), iron(II) (Fe2+) content, were studied before and after desalination. Also, the influence of various water nanofluids (WNF) on increasing the output of glass-made double slope solar stills was studied. Physical and chemical parameter analysis revealed a reduction in values after desalination. Single and double-basin glass solar stills of 0.81 m2 basin area were fabricated, and the experiments were conducted at Kovilpatti. The experiments were carried out at a depth of 1 cm with borewell saline water and nanofluids under insulated conditions. A maximum productivity of 4350 mL per 0.81 m2/day (5405 mL/m2/day) was obtained while using Al2O3 nanofluid in a double basin glass solar still, which was higher than that obtained using water, SnO2 and ZnO WNF. This confirms the efficiency of Al2O3 aqueous nanofluid, which yielded more distillate than other two nanofluids used.