To enhance the efficiency and stability of modern smart grids, accurate short-term electricity demand forecasting is essential. The objective of this study is to present the use of Long Short-Term Memory (LSTM) networks to predict electrical load based on high-resolution, real-world data from the Belgian Elia grid, sampled at 15-minute intervals. The methodology includes data preprocessing, temporal feature extraction, sequence generation, and model optimization. Exploratory data analysis highlights important consumption patterns and seasonal variations. The LSTM model effectively captures both short-term fluctuations and long-term dependencies, achieving an RMSE of 119.41 MW, a MAPE of 1.30%, and an R² score of 0.992 on the test set. Compared to alternative forecasting approaches, including more complex hybrid architectures, the LSTM model demonstrates superior accuracy and generalization capability. For instance, compared with ARIMA-LSTM models that reported a MAPE of 2.83% and CNN-LSTM models with 2.72%, the proposed model achieves markedly better performance. These findings support the integration of LSTM-based forecasting systems into smart grid operations for real-time energy management.

This study investigates the potential of renewable energy integration within the context of Parmigiano Reggiano cheese production. Using real data from dairy plants and farms, advanced approaches for proper energy management are illustrated, including performance indicators for energy benchmarking, the disaggregation of loads for energy efficiency opportunities identification, and the identification of energy demand patterns energy use optimization. The study proposes guidelines for quantifying the potential from solar and biomass energy sources, with a focus on energy self-sufficiency and resource valorization. This is illustrated with regard to a medium-sized dairy farm with a specific energy consumption of 64 kWh/t and 10 kWh/t of milk, respectively. The analysis shows that photovoltaic self-production rate decreases with plant size but can be enhanced to 88% through the integration of energy storage systems. Biogas production can entirely supply the farm’s electricity needs, while further valorization of digestate can provide an additional 79% which could be used to meet the energy demands of a corresponding dairy plant (55 kWh/t of electricity and 100 kWh/t of heat). This study demonstrates that state-of-art technologies can substantially cover the energy requirements of dairy operations, and that more advanced systems can support the achievement of full energy self-sufficiency.

This research is devoted to the electromechanical properties of axle mount devices with piezoelectric elements and their performance under various loading conditions. It performs a complex computational simulation of the interaction between mechanical energy and electrical generation. In the study, the research team examined the technical aspects and scale-up issues of integrating piezoelectric energy harvesting networks into conventional roads. The function of piezoelectric material in car wheels, which senses patient head movement and stability via pressure, is covered in the study. Important considerations include the material’s size, voltage, and Deformation. The extracted materials’ power frequency ranges from 62 Hz to 80 Hz, with 80 Hz used for mechanical energy extraction. The COMSOL 6.3 Multiphysics program was used as a simulation program. The research studies the association of piezoelectric materials, as well as power car tires, concentrating on their mechanics, electronic components, and thermal properties. The study shows that the pulse electric value is proportional to material thickness, voltage, and strain. Being a function of strain and electric power, the electromotive force, mechanical power, and electric power also change with increasing frequency. The temperature dynamics of piezoelectric mechanisms rely heavily on the resistance of the material, which results in a temperature rise that, in turn, produces an input voltage and electron movement. The 1e9 ohm resistance is the best choice, as it provides increased current flow and an electrical potential of 0.7 volts.

In a global context where buildings account for approximately 30% of final energy demand and 26% of energy-related greenhouse gas emissions, improving the building envelope is a key strategy for achieving sustainability goals. This study investigates the energy, environmental and visual performance of a dynamic Second Skin Fac¸ade (SSF) system applied as a passive retrofit solution for a typical office building located in Trondheim, Norway. The SSF integrates adaptive technologies and is composed of 3D-printed panels in Acrylonitrile Styrene Acrylate (ASA): solid panels for opaque walls and perforated panels for windows. A simulation-based methodology using TRNSYSsoftware was implemented to compare the performance of the retrofitted building against a reference case. Additionally, a gate-to-gate Life Cycle Assessment was performed to assess the environmental impact of the 3D-printed components. Results highlight a reduction in primary energy demand by up to 25.5% and an annual decrease of approximately 1.4 tCO_2eq, particularly when the dynamic shading control is based on vertical solar radiation. Although the Global Warming Potential of ASA panels is higher than that of conventional materials, the local production and Norway’s low-carbon electricity grid contribute to a favorable environmental profile. The f indings underline the potential of 3D printing for adaptive envelope solutions.

While solar still technology offers a sustainable solution to freshwater scarcity, its practical application is often limited by low productivity. This study aims to enhance the water production of a double-slope solar still through the simultaneous implementation of a glass cover cooling mechanism and a flat-plate solar collector. Three configurations were experimentally compared: a conventional solar still (SSC), a solar still with cover cooling (SST1), and a solar still integrating both cover cooling and a solar collector (SST2). Experimental results show that SST2 achieved lower glass cover temperatures than the SSC and higher water temperatures than the SST1, thereby accelerating both evaporation and condensation rates. Quantitatively, the SST2 configuration yielded a freshwater productivity of 2092 g/m², a significant increase of 147% compared to the SSC. Furthermore, its energy efficiency reached 44.53%, in contrast to 27.38% for SSC and 27.27% for SST1. Economically, SST2 demonstrated the lowest freshwater production cost at $0.082/(L·m²). These findings rigorously prove that the simultaneous use of cover cooling and a solar collector is a highly effective strategy for increasing the productivity and improving the economic viability of solar stills.

Flat plate solar collectors are widely employed in applications operating at low to moderate temperatures, including domestic water heating and various industrial uses. Their thermal performance is strongly influenced by the absorber tube, through which solar energy is transmitted to the circulating fluid. Conventional designs are often limited by low convective heat transfer, which has motivated studies on geometric enhancements to improve overall efficiency. The present work examines the thermo-hydraulic characteristics of a flat plate solar air collector fitted with twisted tape inserts having various twist ratios ($\delta$ = 3, 4, 5, 6), and compares the results with a plain tube collector. Air serves as the working fluid, and simulations were carried out over a Reynolds number range of 200–2000. A three-dimensional CFD approach was employed to study critical performance characteristics, including outlet temperature, Nusselt number, friction factor, pumping power, and thermal efficiency. The results show that twisted tape collector (TTC) provide considerably greater heat transfer compared to the plain collector (PC). At Re = 1000, the Nusselt number enhancement reached 35.19%, 44.55%, 50.15%, and 54.96% for twist ratios $\delta$ = 6, 5, 4, and 3, respectively. Although this improvement is associated with increased pressure drops, the findings confirm that twisted tape inserts substantially enhance the heat transfer effectiveness of solar collectors by promoting turbulence and better fluid mixing.

The main challenge in the development of the oil and gas industry is the reduction in hydrocarbon production using traditional methods due to the increasing share of hard-to-reach oil and gas reserves in the production structure and the insufficiency and practical unfeasibility of innovative resource-efficient methods and techniques for enhancing oil productivity from heterogeneous reservoirs. The objective of this study is to develop a new hydromechanical perforation technique for enhancing oil well productivity from heterogeneous reservoirs of oil and gas fields using an innovative method called "tunnel perforation". The application of the new approach to stimulating heterogeneous reservoirs allows for a significant increase in oil production without the use of explosives. Within a single tripping operation, three technological processes are conducted simultaneously: perforation (primary and secondary), destruction of the cement sheath behind the casing, and acid treatment of the near-wellbore zone. After the completion of the acid treatment, using the initial set of downhole tools, the products of chemical reactions are extracted to the surface without associated material components. The use of tunnel perforation technology allows for effective oil and gas production for subsurface users within a single tripping operation, compared to existing technologies such as cumulative perforation. The system was implemented and tested in three oil fields in different countries, namely, Group of reservoirs A, Group of reservoirs Ach, and Group of reservoirs B. The field test results showed an increase in the reservoir's productivity by 319% for group A, 120.2% for group Ach, and 114.8% for group B.

In many places, including Iraq, wind energy is a cheap, sustainable resource that is also environmentally benign. Despite its substantial wind potential, Iraq continues to experience an energy deficit due to the underutilization of renewable resources. To close this gap, this study will use a multi-criteria evaluation (MCE) technique in a Geographic Information System (GIS) context to determine the best places for wind energy development in Iraq. The assessment considers several geographic elements that affect wind farm placement, such as wind speed, land slope, distance from water bodies, and proximity to power lines and key roadways. A final suitability map that highlighted locations with differing degrees of acceptability for wind energy harvesting was created by integrating these parameters. According to the results, about 31% of the research region is highly favorable to wind farms, 30% is somewhat reasonable, and 39% is unsuitable. The southern and portions of central Iraq were the most promising regions for wind energy development. These results provide a sound scientific foundation for strategic planning and investment in sustainable energy infrastructure by energy planners and decision-makers. The study helps Iraq meet its sustainable development objectives, lessen its dependency on fossil fuels, and lessen its environmental effects.

The unpredictable nature of energy markets makes precise electricity price forecasting (EPF) necessary to improve bidding strategies and lower risk. For instance, this study introduces a hybrid deep learning model CNN-GRU-VAE, that learns sequences using Gated Recurrent Units (GRU), finds features using Convolutional Neural Networks (CNN), and becomes more general using a Variational Autoencoder (VAE). In tests that looked ahead one day, the CNN-GRU-VAE performed better than the CNN, ANN, GRU, and CNN-GRU models. The model’s Root Mean Squared Error (RMSE) is 0.8733, Mean Squared Error (MSE) is 0.7627, and Mean Absolute Error (MAE) is 0.6373. These findings demonstrate improved accuracy and stability across diverse market conditions. The integration of convolutional, recurrent, and generative components within a unified framework provides superior predictions compared to traditional methods, demonstrating robustness and practical applicability for day-ahead electricity price forecasting in competitive energy markets.

This study aims to optimize biogas production at the laboratory scale using a batch-mode bioreactor and Response Surface Methodology (RSM). The main objective is to assess the effects of key parameters substrate composition (household waste and cow manure), pH, fermentation temperature, and agitation speed on biogas yield.series of experiments were designed using a central composite RSM to evaluate the influence of substrate composition and temperature. The experimental data were analyzed through ANOVA to assess model significance and accuracy.The results show that the developed quadratic models are statistically significant, with a determination coefficient (R²) of 0.90 for cumulative biogas production. These findings confirm the adequacy of the models and the effectiveness of RSM in identifying optimal operating conditions for enhanced biogas yield.

The increasing energy demand in buildings and global warming issues have become a critical challenge that needs to be addressed. An energy management system (EMS) aims to manage energy efficiency and reduce environmental impact. This paper proposes the implementation of an EMS in a campus building by integrating artificial intelligence (AI) and Internet of Things (IoT) technology for prediction and energy optimization. A long short-term memory (LSTM) network is employed to predict the room environment conditions, including the concentration of carbon dioxide (CO$_2$), room temperature and humidity, outdoor temperature and humidity, and room occupancy. A genetic algorithm (GA)-based optimization method is employed to minimize energy consumption while maintaining user comfort by adjusting the room set-point temperature. The IoT-based monitoring system is used to monitor environmental parameters and power consumption in the room. The experimental results show that the proposed LSTM-based prediction achieves a low root mean square error of 50.69 ppm, 0.77°C, 1.08°C, 3.50%, 6.27% for the CO$_2$, room and outdoor temperature, room and outdoor humidity, respectively; and a high Accuracy of 0.93 for room occupancy. Additionally, machine learning techniques are proposed for occupancy modeling with high Accuracy of 0.98 and F1 score of 0.97. Furthermore, the algorithm is tested on an embedded device with a fast execution time (below two minutes), making it suitable for real-time implementation of the application.

Oil is one of the primary sources of energy worldwide; however, its production brings environmental challenges associated with the use of conventional extraction methods. Biosurfactants have emerged as a sustainable alternative for improving efficiency in Enhanced Oil Recovery (EOR). The objective of this research is to analyze the role of biosurfactants in EOR through a bibliometric analysis and a systematic review, identifying trends, key microorganisms, and their impact on recovery efficiency. The research methodology consisted of three phases: (i) selection of data for analysis, (ii) application of scientific metrics, and (iii) systematic review using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, focusing on contributions from the last six years. The bibliometric analysis compiled data from 1988 to 2025, with 745 academic publications indexed in the Scopus and WoS databases from 48 countries, with the main contributions coming from China, India, and Iran, attributing their dominance to state investments in research and development for energy and biotechnology innovation. The systematic review found that the most studied biosurfactants are from Pseudomonas and Bacillus, with rhamnolipids and surfactins being the most prevalent. They act through multiple mechanisms and show potential in applications involving in situ and ex situ production, with additional oil recovery rates exceeding 10% in laboratory studies.