Search Results (222)

The use of power smoothing for renewable energy resources is attracting increasing attention. One widely used resource that could benefit from this technique is the grid-tied photovoltaic (PV) system. Solar energy production typically follows a Gaussian bell curve, with peaks at midday. This paper confirms this pattern by using the bell curve as a reference; however, climate variations can significantly alter this pattern. Therefore, this study aimed to smooth the power supplied to the grid by a PV system. The proposed controller manages the charge and discharge processes of the energy storage system (ESS) to ensure a smooth Gaussian bell curve output. It adjusts the parameters of this curve to closely match the generated energy, absorbing or supplying fluctuations to maintain the desired profile. This system also aims to provide accurate predictions of the power that should be supplied to the grid by the PV system, based on the capabilities of the ESS and the overall system performance. Although experimental results were not included in this analysis, the system was implemented in SIMULINK using real-world data. The controller utilizes a hybrid ESS comprising a vanadium redox battery (VRB) and supercapacitors (SCs). The design and operation of the controller, including curve tuning and ESS charge–discharge management, are detailed. The simulation results demonstrate excellent performance and are thoroughly discussed. Full article

The increasing penetration of photovoltaic solar energy has intensified the need for accurate production forecasting to ensure efficient grid operation. This study presents a critical comparison of traditional statistical methods and machine learning approaches for forecasting solar irradiance using the benchmark Folsom PLC dataset. Two primary research questions are addressed: whether machine learning models outperform traditional techniques, and whether time series modelling improves prediction accuracy. The analysis includes an evaluation of a range of models, including statistical regressions (OLS, LASSO, ridge), regression trees, neural networks, LSTM, and random forests, which are applied to physical modelling and time series approaches. The results reveal that although machine learning methods can outperform statistical models, particularly with the inclusion of exogenous weather features, they are not universally superior across all forecasting horizons. Furthermore, pure time series approach models yield lower performance. However, a hybrid approach in which physical models are integrated with machine learning demonstrates significantly improved accuracy. These findings highlight the value of hybrid models for photovoltaic forecasting and suggest strategic directions for operational implementation. Full article

In recent years, recovering precious and base metals such as silver and copper from end-of-life products has become a fundamental factor in the sustainable development of many countries. This not only supports environmental goals but is also a profitable economic activity. Therefore, in this study, we investigate the recovery of silver and copper from an end-of-life photovoltaic panel powder using an alternative leaching system containing sulfuric acid and ferric sulfate instead of nitric acid-based leaching systems, which are susceptible to producing hazardous gases such as NO_x. To obtain this goal, a series of experiments were designed with the Central Composite Design (CCD) approach using Response Surface Methodology (RSM) to evaluate the effect of reagent concentrations on the leaching rate. The leaching results showed that high recovery rates of silver (>85%) and copper (>96%) were achieved at room temperature using a solution containing only 0.2 M sulfuric acid and 0.15 M ferric sulfate. Analysis of variance was applied to the leaching data for silver and copper recovery, resulting in two statistical models that predict the leaching efficiency based on reagent concentrations. Results indicate that the models are statistically significant due to their high R² (0.9988 and 0.9911 for Ag and Cu, respectively) and the low p-value of 0.0043 and 0.0003 for Ag and Cu, respectively. The models were optimized to maximize the dissolution of silver and copper using Design Expert software. Full article

Hydrogen is a potential key component of a carbon-neutral energy carrier and an input to marine industrial processes. This study examines the consequences of coupled hydrogen release and marine environmental factors during floating photovoltaic hydrogen production (FPHP) system failures. A validated three-dimensional numerical model of FPHP comprehensively characterizes hydrogen leakage dynamics under varied rupture diameters (25, 50, 100 mm), transient release duration, dispersion patterns, and wind intensity effects (0–20 m/s sea-level velocities) on hydrogen–air vapor clouds. FLACS-generated data establish the concentration–dispersion distance relationship, with numerical validation confirming predictive accuracy for hydrogen storage tank failures. The results indicate that the wind velocity and rupture size significantly influence the explosion risk; 100 mm ruptures elevate the explosion risk, producing vapor clouds that are 40–65% larger than 25 mm and 50 mm cases. Meanwhile, increased wind velocities (>10 m/s) accelerate hydrogen dilution, reducing the high-concentration cloud volume by 70–84%. Hydrogen jet orientation governs the spatial overpressure distribution in unconfined spaces, leading to considerable shockwave consequence variability. Photovoltaic modules and inverters of FPHP demonstrate maximum vulnerability to overpressure effects; these key findings can be used in the design of offshore platform safety. This study reveals fundamental accident characteristics for FPHP reliability assessment and provides critical insights for safety reinforcement strategies in maritime hydrogen applications. Full article

Photovoltaic panels are pivotal in transforming solar irradiance into electricity, making them a key technology in renewable energy. Despite their potential, the distribution of photovoltaic systems in Romania remains sparse, requiring advanced data analytics for effective management, particularly in addressing the intermittent nature of photovoltaic energy. This study investigates the predictive capabilities of Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) architectures for forecasting hourly photovoltaic energy production in Romania. The results indicate that CNN models significantly outperform LSTM models, with 77% of CNNs achieving an R² of 0.9 or higher compared to only 13% for LSTMs. The best-performing CNN model reached an R² of 0.9913 with a mean absolute error (MAE) of 9.74, while the top LSTM model achieved an R² of 0.9880 and an MAE of 12.57. The rapid convergence of the CNN model to stable error rates illustrates its superior generalization capabilities. Moreover, the model’s ability to accurately predict photovoltaic production over a two-day timeframe, which is not included in the testing dataset, confirms its robustness. This research highlights the critical role of accurate energy forecasting in optimizing the integration of photovoltaic energy into Romania’s power grid, thereby supporting sustainable energy management strategies in line with the European Union’s climate goals. Through this methodology, we aim to enhance the operational safety and efficiency of photovoltaic systems, facilitating their large-scale adoption and ultimately contributing to the fight against climate change. Full article

This study investigates the performance prediction of poly-crystalline photovoltaic (PV) systems in Jordan using experimental data, analytical models, and machine learning approaches. Two 5 kWp grid-connected PV systems at Applied Science Private University in Amman were analyzed: one south-oriented and another east–west (EW)-oriented. Both systems are fixed at an 11° tilt angle. Linear regression, Least Absolute Shrinkage and Selection Operator (LASSO), ElasticNet, and artificial neural networks (ANNs) were employed for performance prediction. Among these, linear regression outperformed the others due to its accuracy, interpretability, and computational efficiency, making it an effective baseline model. LASSO and ElasticNet were also explored for their regularization benefits in managing feature relevance and correlation. ANNs were utilized to capture complex nonlinear relationships, but their performance was limited, likely because of the small sample size and lack of temporal dynamics. Regularization and architecture choices are discussed in this paper. For the EW system, linear regression predicted an annual yield of 1510.45 kWh/kWp with a 2.1% error, compared to 1433.9 kWh/kWp analytically (3.12% error). The south-oriented system achieved 1658.15 kWh/kWp with a 1.5% error, outperforming its analytical estimate of 1772.9 kWh/kWp (7.89% error). Productivity gains for the south-facing system reached 23.64% (analytical), 10.43% (experimental), and 9.77% (predicted). These findings support the technical and economic assessment of poly-crystalline PV deployment in Jordan and regions with similar climatic conditions. Full article

This article presents a modeling and analysis approach for a hybrid photovoltaic wind turbine (PV-WT) hydrogen production system. This study uses the TRNSYS simulation platform to evaluate the system under coastal climate conditions in Perth, Australia. The system encapsulates an advanced alkaline electrolyzer (ELE) and an alkaline fuel cell (AFC). A comprehensive 4E (energy, exergy, economic, and environmental) assessment is conducted. The analysis is based on hourly dynamic simulations over a full year. Key performance metrics include hydrogen production, energy and exergy efficiencies, carbon emission reduction, levelized cost of energy (LCOE), and levelized cost of hydrogen (LCOH). The TRNSYS model is validated against the existing literature data. The results show that the system performance is highly sensitive to ambient conditions. A sensitivity analysis reveals an energy efficiency of 7.3% and an exergy efficiency of 5.2%. The system has an entropy generation of 6.22 kW/K and a sustainability index of 1.055. The hybrid PV-WT system generates 1898.426 MWh of renewable electricity annually. This quantity corresponds to 252.7 metric tons of hydrogen production per year. The validated model shows a stable LCOE of 0.102 USD/kWh, an LCOH of 4.94 USD/kg, an energy payback time (EPBT) of 5.61 years, and cut CO₂ emissions of 55,777.13 tons. This research provides a thorough analysis for developing green hydrogen systems using hybrid renewables. This study also offers a robust prediction model, enabling further enhancements in hybrid renewable hydrogen production. Full article

Photovoltaic (PV) energy production in Western countries increases yearly. Its production can be carried out in a highly distributed manner, not being necessary to use large concentrations of solar panels. As a result of this situation, electricity production through PV has spread to homes and open-field plans. Production varies substantially depending on the panels’ location and weather conditions. However, the integration of PV systems presents a challenge for both grid planning and operation. Furthermore, the predictability of rooftop-installed PV systems can play an essential role in home energy management systems (HEMS) for optimising local self-consumption and integrating small PV systems in the low-voltage grid. In this article, we show a novel methodology used to predict the electrical energy production of a 48 kWp PV system located at the Campus Feuchtwangen, part of Hochschule Ansbach. This methodology involves hybrid time series techniques that include state space models supported by artificial intelligence tools to produce predictions. The results show an accuracy of around 3% on nRMSE for the prediction, depending on the different system orientations. Full article

This research introduces a novel hybrid machine learning framework for automated quality prediction and classification of silicon solar modules in production lines. Unlike conventional approaches that rely solely on encapsulation loss rate (ELR) for performance evaluation—a method limited to assessing encapsulation-related power loss—our framework integrates unsupervised clustering and supervised classification to achieve a comprehensive analysis. By leveraging six critical performance parameters (open circuit voltage (V_OC), short circuit current (I_SC), maximum output power (P_max), voltage at maximum power point (VPM), current at maximum power point (IPM), and fill factor (FF)), we first employ k-means clustering to dynamically categorize modules into three performance classes: excellent performance (ELR: 0–0.77%), good performance (0.77–8.39%), and poor performance (>8.39%). This multidimensional clustering approach overcomes the narrow focus of traditional ELR-based methods by incorporating photoelectric conversion efficiency and electrical characteristics. Subsequently, five machine learning classifiers—decision trees (DT), random forest (RF), k-nearest neighbors (KNN), naive Bayes classifier (NBC), and support vector machines (SVMs)—are trained to classify modules, achieving 98.90% accuracy with RF demonstrating superior robustness. Pearson correlation analysis further identifies V_OC, P_max, and VPM as the most influential quality determinants, exhibiting strong negative correlations with ELR (−0.953, −0.993, −0.959). The proposed framework not only automates module quality assessment but also enhances production line efficiency by enabling real-time anomaly detection and yield optimization. This work represents a significant advancement in solar module evaluation, bridging the gap between data-driven automation and holistic performance analysis in photovoltaic manufacturing. Full article

The selection of the peak power of a photovoltaic system to meet the energy demand of a building is a key task in the energy transformation. This article presents an algorithm for assessing the correctness of the selection of a photovoltaic system with a peak power of 50 kWp for the needs of a university administration building. This is made possible due to the use of an advanced photovoltaic inverter, which is a device of the Internet of Things and the smart metering system. At the beginning of the review, the authors employed the naked eye measurement data of the time series related to the power production by the photovoltaic system and its consumption by the university building. Then, traditional statistical analyses were performed, characterizing the generated power divided into self-consumption power and that fed into the power grid. The analysis of the total consumed power was performed with the division into the power produced by the photovoltaic system and that taken from the power grid. The analyses conducted were subjected to expert validation aimed at explaining the nature of the behavior of the power generation and consumption systems. The main goal of this article is to determine the signatures of the power generated by the photovoltaic system and consumed by the administration building. As a result of unsupervised clustering, the power generation and consumption space were divided into states. These were then termed based on their nature and their usefulness in managing the power produced and consumed. Presentation of clustering results in the form of heatmaps allows for localization of specific states at specific times of the day. This leads to their better understanding and quantification. The signatures of power generated by the photovoltaic system and consumed by the university building confirmed the possibility of using an energy storage system. The presented computational algorithm is the basis for determining the correctness of the photovoltaic system selection for the current energy needs of the building. It can be the basis for further analysis related to the prediction of both the power generated by Renewable Energy Sources and the energy consumed by diverse types of buildings. Full article

This review provides a comprehensive and multidisciplinary overview of recent advancements in solar tracking systems (STSs) aimed at improving the efficiency and adaptability of photovoltaic (PV) technologies. The study systematically classifies solar trackers based on tracking axes (fixed, single-axis, and dual-axis), drive mechanisms (active, passive, semi-passive, manual, and chronological), and control strategies (open-loop, closed-loop, hybrid, and AI-based). Fixed-tilt PV systems serve as a baseline, with single-axis trackers achieving 20–35% higher energy yield, and dual-axis trackers offering energy gains ranging from 30% to 45% depending on geographic and climatic conditions. In particular, dual-axis systems outperform others in high-latitude and equatorial regions due to their ability to follow both azimuth and elevation angles throughout the year. Sensor technologies such as LDRs, UV sensors, and fiber-optic sensors are compared in terms of precision and environmental adaptability, while microcontroller platforms—including Arduino, ATmega, and PLC-based controllers—are evaluated for their scalability and application scope. Intelligent tracking systems, especially those leveraging machine learning and predictive analytics, demonstrate additional energy gains up to 7.83% under cloudy conditions compared to conventional algorithms. The review also emphasizes adaptive tracking strategies for backtracking, high-latitude conditions, and cloudy weather, alongside emerging applications in agrivoltaics, where solar tracking not only enhances energy capture but also improves shading control, crop productivity, and rainwater distribution. The findings underscore the importance of selecting appropriate tracking strategies based on site-specific factors, economic constraints, and climatic conditions, while highlighting the central role of solar tracking technologies in achieving greater solar penetration and supporting global sustainability goals, particularly SDG 7 (Affordable and Clean Energy) and SDG 13 (Climate Action). Full article

This research focuses on developing an intelligent irrigation solution for agricultural systems utilising solar photovoltaic-thermal (PVT) energy applications. This solution integrates PVT applications, prediction, modelling and forecasting as well as plants’ physiological characteristics. The primary objective is to enhance water management and irrigation efficiency through innovative digital techniques tailored to different climate zones. In the initial phase, the performance of PVT solutions was evaluated using ANSYS Fluent software R19.2, revealing that scaled PVT systems offer optimal efficiency for PV systems, thereby optimising electrical production. Subsequently, a comprehensive approach combining integral feature selection (IFS) with machine learning (ML) and deep learning (DL) models was applied for reference evapotranspiration (ETo) prediction and water needs forecasting. Through this process, 301 optimal combinations of predictors and best-performing linear models for ETo prediction were identified. Achieving R² values exceeding 0.97, alongside minimal indicators of dispersion, the results indicate the effectiveness and accuracy of the elaborated models in predicting the ETo. In addition, by employing a hybrid deep learning approach, 28 best models were developed for forecasting the next periods of ETo. Finally, an interface application was developed to house the identified models for predicting and forecasting the optimal water quantity required for specific plant or crop irrigation. This application serves as a user-friendly platform where users can input relevant predictors and obtain accurate predictions and forecasts based on the established models. Full article

The study addresses the impact of selected environmental factors on the energy production of photovoltaic systems under real outdoor conditions, with particular emphasis on the application of evolutionary computation techniques. The experiment was carried out on a dedicated test stand, where measurements were made under natural environmental conditions. Parameters such as solar irradiance, wind speed, temperature, air pollution, and obtained PV power were continuously recorded. Initial correlation analysis using Pearson and Spearman coefficients confirmed associations between environmental factors and power output, especially solar irradiance. In order to advance the analysis beyond conventional methods, a linear regression model was developed in which the model weights were optimized using evolutionary algorithms, allowing for a more robust assessment of the contribution of each parameter. The results showed that solar irradiance accounted for 97.79% of the variance in photovoltaic power, while temperature (0.95%), air pollution (0.72%), and wind speed (0.54%) had significantly lower impacts. The implementation of evolutionary algorithms represents a novel approach in this context and has proven to be effective in quantifying environmental influence under complex real-world conditions. Furthermore, the findings highlight the indirect role of air pollution in attenuating irradiance and reducing system efficiency. These insights provide a foundation for the development of adaptive control strategies and predictive models to optimize the performance of the photovoltaic system in dynamic environmental settings. Full article

Floating photovoltaic systems (FPV) are one of the emerging technologies that are able to support the “green” energy transition. In Greece, the environmental impact assessment of such projects is still under early development. The scope of the present study was to provide insights into the potential impacts of a small-scale FPV system on the water quality of the oligotrophic Kremasta Lake, an artificial reservoir. For this reason, a hydrodynamic and water quality model was employed. The results showed that the water quality parameter variations were insignificant and limited only in the immediate area of the FPV construction and gradually disappeared toward the shoreline. Likewise, this variation was restricted to the first few meters of depth of the water column and was eliminated onwards. The water temperature slightly decreased only in the area of close proximity to the installation. Average annual dissolved oxygen, chlorophyll-a, and nutrient concentrations were predicted not to change considerably after the panels’ construction. FPV systems can provide an attractive alternative for energy production in artificial reservoirs, especially in regions of land use conflicts that are associated with land allocation for alternative energy development. Given the limited data on the long-term impact of such projects, robust monitoring programs are essential. These initiatives rely on public support, making collaboration between stakeholders and the local community crucial. Full article

Agrophotovoltaic (APV) systems represent innovative agricultural farms and solar power plants, capable of producing electricity and crops simultaneously. Since the solar radiation required to optimize harvests varies by crop type, traditional PV panels face challenges in efficiently adjusting the shading ratio of APV systems. This study evaluates the economic viability of APV systems integrated with building-integrated photovoltaic (BIPV) systems for biofuel production. Specifically, it assesses the production forecast for corn-based biofuel—demand for which is rising due to the mixed-fuel use policy of the Korean government—and the economic feasibility of production in the APV system enhanced by BIPV integration (i.e., the APV–BIPV system). To this end, LCOE (levelized cost of energy) and NPV (net present value) are employed as performance indicators. Additionally, yield data from corn and corn stover harvested in actual APV facilities are utilized to predict bioenergy production. Consequently, the study will analyze the impact of renewable energy production from the proposed APV–BIPV system on achieving the Korean government’s renewable energy production goals and will provide guidelines on the potential benefits for farmers involved in renewable energy production and energy crop harvesting. Full article