Current Issue : April-June Volume : 2026 Issue Number : 2 Articles : 5 Articles
This paper proposes a hybrid forecasting framework that combines Long Short-Term Memory (LSTM) networks with Shapley Additive Explanations (SHAPs) to quickly and accurately predict solar radiation. Historical meteorological data from the Central Weather Administration (CWA) in Taiwan, spanning 2018–2023, are processed to construct multivariate input features, including temperature, humidity, pressure, wind conditions, global radiation, and temporal encodings. The LSTM network is employed to capture nonlinear dependencies and temporal dynamics in the multivariate meteorological data. SHAPguided feature selection reduces the number of input variables, thereby lowering computational cost and accelerating convergence without sacrificing accuracy. A case study in the Penghu region—characterized by abundant solar irradiance and active photovoltaic deployment— was conducted to evaluate the model under three scenarios. Results demonstrated that if the number of features decreases from fifteen to five, the number of model parameters is reduced from 53,569 to 51,521 and the computation time is reduced from 6 ms to 4 ms. The MSE and MAE remain within the range of 0.07~0.11 and 0.13~0.18, with almost no change. The LSTM–SHAP framework not only achieves high forecasting precision but also provides transparent explanations of key meteorological drivers, with the temperature, humidity, and temporal variables identified as the most influential factors. Overall, this research contributes a scalable and interpretable methodology for solar radiation prediction, offering practical implications for photovoltaic power dispatch, grid stability, and renewable energy planning....
In this paper, we use a statistical approach to study the distribution of days of geomagnetic activity caused by the fluctuation of the Sun’s neutral plate as a function of solar phases and season during solar cycle 24. We also examine the daily response of the magnetospheric convective electric field (MCEF) to the geomagnetic disturbance caused by these days of fluctuating activity. A comparison of the different responses of the MCEF to the three main classes of geomagnetic activity disturbance is also made. A study of the occurrences of fluctuating days shows that: 1) the descending phase is the most active, with an annual occurrence of 50%, followed by the maximum phase (28%), the ascending phase (19%) and finally the ascending phase (15%); 2) spring is the most active season, with an occurrence of 25.55%, followed by autumn (25.25%), summer (24.75%) and winter (24.45%). Irrespective of the phase and time of year, one day in four the Earth’s magnetosphere is under the impact of the fluctuating solar winds responsible for the fluctuating geomagnetic activity. From the minimum phase of the solar cycle to the waning phase, the daily mean values of the MCEF are 0.08448182 mV/m, 0.1134496 mV/m, 0.11846218 mV/m and 0.1178042 mV/m, respectively. The average daily intensities of the MCEF are 0.116947784 mV/m in spring, 0.10854571 mV/m in summer, 0.12374118 mV/m in autumn and 0.10678156 mV/m in winter. Irrespective of solar phase and season, the average daily intensity of the MCEF on fluctuating days is 0.10854945 mV/m. A comparison of the results of this study with previous work on cycle 24 shows that of the three classes of disturbed geomagnetic activity, fluctuating geomagnetic activity is the one that disturbs the magnetospheric convection electric field the least....
This theoretical study compares two renewable energy configurations—Unit A (photovoltaic-only, ground-mounted panels) and Unit B (hybrid solarwind with elevated photovoltaic panels)—across varying surface areas (1, 5, 10, and 20 hectares) in four locations: Huarte/Uharte, Navarra (42.83˚N), Almería (36.84˚N), Sinaí (31˚N), and Ecuador (0˚ latitude). Unit B elevates solar panels 3 - 4 meters on posts to allow agricultural or animal activity beneath, integrating mini vertical-axis wind turbines (VAWTs)wind energy for day-and-night production. We project global photovoltaic capacity using a small-scale 1 m2 prototype built by teenagers, we used Grok 3 X’s (xAI) to project outcomes, estimating energy output in GWh/day and potential increases in food production due to preserved agricultural land. Results show Unit B outperforms Unit A by up to 50% in high-wind regions (Sinaí, Almería) and seasons (winter, spring) in high-wind regions (e.g., Sinaí winter) and low-irradiance seasons, with full significant land use benefits. Global adoption of Unit B could enhance energy production by ~600 GWh/day. Adoption of Unit B globally could enhance energy production by up to 30% in different regions worldwide while freeing ~10 millions hectares for agriculture, supporting food security....
This paper presents the implementation and outcomes of the AgroTech Project under the Komuniti@UniMADANI initiative, spearheaded by Universiti Tun Hussein Onn Malaysia (UTHM). The project investigates the feasibility and practicality of integrating sustainable resources with Internet of Things (IoT) technology to cultivate reddish pandan coconut trees within a primary school compound in Kluang, Johor, Malaysia. The school’s landscape, particularly its field, suffers from minimal vegetation and elevated ambient temperatures. Addressing this, the initiative planted 100 coconut trees and deployed IoT-enabled systems for fertigation, solar-powered irrigation, and rainwater harvesting. The system utilizes sensors, solar panels, and automated controls to monitor climate conditions and nutrient delivery, exemplifying a smart agriculture model aligned with the United Nations Sustainable Development Goals (SDGs), notably Climate Action and Responsible Consumption and Production. The study demonstrates the viability of using solar energy to power water pumps, IoT-based Blynk systems for nutrient regulation, and fertilized rainwater as a sustainable irrigation source. Beyond technical implementation, the paper evaluates the project's social, economic, and environmental impacts, highlighting its potential to enhance community resilience and ecological stewardship. It also addresses operational challenges and proposes the model as a scalable and replicable framework for green innovation in educational and community settings. This research contributes to the limited body of literature on IR4.0-enabled greening strategies in schools and underscores the transformative potential of integrating smart technologies with sustainable agriculture....
Optical performance of perovskite-based solar cells can be enhanced by utilizing fully textured interfaces. However, solution processing of perovskite films on textured surfaces is a nonstraightforward and challenging process, particularly if optically most efficient micrometer-sized textures are used. In this work, we present fully textured solution-processed perovskite solar cells on periodic inverted micropyramids. The textures have a period of 4 μm with varying pyramid depths and are fabricated by wetchemical etching of silicon with subsequent replication on glass substrates using nanoimprint lithography. Inverted pyramids are shown to enable low reflectance similar to random micropyramids on silicon. Additionally, they are able to confine perovskite precursor solution within its structure during spin coating, resulting in a conformal, fully textured perovskite film. We demonstrate that the resulting fully textured single-junction perovskite solar cells feature a reduced reflection loss of up to 1.2 mA/cm2 in short-circuit current density. Moreover, we observe that the amount of lead iodide in the perovskite precursor solution crucially impacts growth and nonradiative recombination losses of the fully textured perovskite solar cells on inverted micropyramids. Finally, we prove the versatility of our approach by also demonstrating conformal coating with slot-die coating, which is a scalable process considered for industrial application....
Loading....