The paper is a review of chosen reports on the integration of photovoltaics and agrotechnics and also presents the concept of an agro-photovoltaic foil tunnel module. The concept of an agro-photovoltaic module assumes the possibility of integrating the modules into the intelligent agro-photovoltaic fields. The proposed solution combines the cultivation in a foil tunnel with a photovoltaic farm. The paper reviews electronic solutions to control ambient parameters particularly those influencing the rate of photosynthesis in foil tunnels. The article also presents an analysis of energy and water balance of the module in 4 different climatic zones. The climate zones include locations in Poland, Spain, Algeria, and Colombia. The module’s water requirements include drip irrigation and fogging systems. Water is assumed to be obtained from rainfall, desalination, non-conventional (treated, desalinated) resources, groundwater, and sub-artesian wells. The module’s energy needs include electricity consumption for pumping freshwater, desalination and water treatment, irrigation and fertilization, additional lighting and heating of plants, folding and unfolding of foil covers, and energy consumption related to control. The article also presents an analysis of crop shading by solar panels and its impact on the selected crop. The functionality of the presented solution was assessed for tomato cultivation in the four selected areas. A foil tunnel of the total area 240 m2 was considered. 60 m2 of the tunnel was covered by the PV panels. Results of the analysis show that the most energetically effective agro-photovoltaic cultivation of tomatoes appears to be in the Saharan region of Africa where 16 MWh annual energy surplus has been obtained. Slightly less effective seem to be cultivations in Spain (Cartagena) and Colombia (Cali) with the approximately 15.5 MWh/a. The least effective agro-photovoltaic cultivation of tomatoes proved to be in Poland where the energy surplus reached 8.5 MWh/a. However, economic return from the cultivation strongly depends on local energy and tomato prices. The system of smart tunnels proposed by the authors combines photovoltaics with controlled protection of crops against unfavorable and extreme climatic conditions. In addition, the system allows plants to be grown with the maximum possible number of days with natural growing conditions, i.e. with uncovered tunnels, with natural bio-fertilization and with natural sunlight.

Assessing groundwater quality is essential for sustainable water resource management in arid regions, where hydrochemical processes are complex and monitoring resources are often limited. This study proposes a hybrid deep learning framework that integrates convolutional neural networks (CNNs) with transformer-based self-attention mechanisms for multiclass water quality index (WQI) classification. A dataset comprising 621 groundwater samples from the Bechar region in southwestern Algeria was analyzed using 12 hydrochemical parameters. To evaluate model robustness under different monitoring conditions, 12 feature combinations representing distinct hydrochemical groups were tested using both CNN and CNN–Transformer architectures. Model performance was assessed using accuracy, precision, recall, F1-score, the area under the receiver operating characteristic (ROC) curve (AUC), and an error-overlap Venn diagram. The CNN–Transformer consistently outperformed the conventional CNN across most configurations. The highest performance was achieved using the full hydrochemical profile, with an accuracy of 92.5% and a micro-AUC of 0.99. Because the WQI target was deterministically calculated from the same hydrochemical variables used as model inputs, the proposed framework should be interpreted as a surrogate WQI classification model rather than as an independent predictor of groundwater quality. Accordingly, the SHAP importance patterns partly reflect the adopted WQI formulation, including its parameter weights, regulatory limits, and class thresholds. Importantly, several reduced models based on only four to five commonly measured parameters also achieved strong predictive performance, with accuracies of up to approximately 0.85, demonstrating the potential of the proposed framework for streamlined groundwater monitoring when comprehensive laboratory analyses are unavailable. Graphical Abstract