Photovoltaic–thermal (PVT) applications have been widely studied in recent years, though commercialisation has become critical due to their operational characteristics and size. In this study, a portable PVT system was developed for mobilisation with assistance from an organic phase-change material (PCM). Two different PCM composites were developed using the PCM with charcoal (PCM + C) and charcoal and metal flakes (PCM + C + M). Considering the portability of the PVT system, conventional metal-container-based PCM storage units were avoided, and the shape-stabilised PCMs (SS-PCMs) were fitted directly on the back surface of the PV module. Further, a serpentine copper tube was placed on the SS-PCMs to extract heat energy for hot water applications. It was found that PV_PCM+C+M exhibited a higher cooling rate, with peak reductions of 24.82 °C and 4.19 °C compared to the PV_noPCM and PV_PCM+C, respectively. However, PV_PCM+C exhibited a higher outlet water temperature difference of 11.62 °C. Secondly, an increase of more than 0.2 litres per minute showed a declining trend in cooling in the PV module. Considering the primary concern of electrical power generation, it was concluded that PV_PCM+C+M is suitable for PVT mobilisation applications, owing to it having shown the highest thermal cooling per 190 g of PCM and a 1-Watt (TCPW) cooling effect of 2.482 °C. In comparison, PV_PCM+C achieved a TCPW cooling effect of 1.399 °C.

Transformer-based embedding models are widely used for similarity search as they are reliable and efficient for capturing semantic similarity. This study uses all-MiniLM-L6-v2, paraphrase-MiniLM-L6-v2 and all-distilroberta-v1 transformer-based embedding models to find the similarity search for Wikipedia documents. All three transformer models are ensembled for enhanced semantic search, and Principal Component Analysis (PCA) is applied to ensure smooth assembly of a different dimensionality model. To understand the strength of the proposed transformer models, 2,000 Wikipedia documents were arbitrarily selected and converted into vectors before storing them in MongoDB. The ground truth of the proposed transformer-based models was examined using 996 TREC questions. The all-MiniLM-L6-v2 and paraphrase-MiniLM-L6-v2 consume less memory than all-distilroberta-v1 model. However, the ensemble process abruptly increased the memory usage to 924.79 MB, higher than individual models. Following that, the average execution time for each query increased to 0.1031 seconds. Beneficially, the ensemble+PCA attained higher precision@10 and recall, resulting in a higher F1 score with an average of 0.5094. The error analysis method indicates that the ensemble+PCA approach significantly improved the semantic search with a higher relevant rate to the raised query. Furthermore, ensemble-based PCA methods are recommended for large dataset handling and are suitable for real-time applications.