As the global push towards renewable energy intensifies, it becomes imperative to comprehensively assess the environmental impacts and sustainability of renewable energy systems throughout their operational lifecycle. Traditional lifecycle assessment (LCA) methods, while useful, often fall short in handling the complex, dynamic data associated with renewable energy systems. This study explores the application of artificial intelligence (AI) and machine learning (ML) techniques to enhance lifecycle assessments of wind, solar, and green hydrogen energy systems, aiming to provide more accurate, efficient, and comprehensive evaluations. AI-driven LCA models leverage extensive datasets from various stages of the lifecycle of renewable energy systems, including raw material extraction, manufacturing, installation, operation, maintenance, and decommissioning. By employing ML algorithms, these models can identify patterns and relationships within the data, predict potential environmental impacts, and provide insights into sustainability performance over time. The research focuses on developing and validating ML models that incorporate diverse data inputs such as material usage, energy consumption, emissions, and waste generation. These models are trained using historical data from multiple renewable energy projects and are capable of adapting to new data inputs, ensuring continuous improvement in assessment accuracy. Key findings demonstrate that AI-enhanced LCA models significantly improve the precision and depth of environmental impact assessments. For wind energy systems, ML models help in predicting turbine lifespan and maintenance needs, thereby optimizing resource use and minimizing environmental footprints. In solar energy systems, AI techniques assist in forecasting degradation rates and energy yield, contributing to more sustainable design and operation. For green hydrogen production, ML models optimize the electrolysis process and assess the overall sustainability of hydrogen supply chains. The integration of AI in LCA facilitates real-time monitoring and dynamic adjustments, ensuring that renewable energy systems operate at peak sustainability. This approach not only enhances the environmental performance of individual systems but also supports strategic decision-making in renewable energy deployment and policy development. In conclusion, the application of AI and ML techniques in lifecycle assessment offers a transformative approach to evaluating the environmental impact and sustainability of renewable energy systems. This research underscores the critical role of advanced analytics in advancing the global transition to sustainable energy and calls for further exploration and adoption of AI-driven LCA methodologies. Keywords: Machine Learning, Renewable Energy Systems, Environmental Impact, Sustainability, AI-Enhanced Lifecycle.
The growing demand for sustainable energy solutions has spurred the development of hybrid renewable energy systems (HRES), which combine multiple renewable sources like solar and wind to enhance energy reliability and efficiency. However, optimizing the performance of HRES and managing energy storage remain significant challenges. This study explores the application of machine learning (ML) techniques to model hybrid renewable energy systems, integrating data from solar and wind sources to predict system performance and improve energy storage solutions. Machine learning algorithms are employed to analyze large datasets generated from solar panels and wind turbines, including variables such as solar irradiance, wind speed, temperature, and historical power output. By identifying patterns and correlations within these datasets, ML models can predict the performance of the hybrid system under various environmental conditions, enabling more efficient management and utilization of renewable resources. The research focuses on the development of various ML models, including regression analysis, neural networks, and ensemble methods, to enhance the predictive accuracy of HRES performance. These models are trained on extensive historical data from multiple renewable energy installations, ensuring robustness and reliability. Feature selection techniques are used to identify the most significant factors affecting system performance. Key findings demonstrate that ML-driven modelling significantly improves the accuracy of performance predictions for hybrid renewable energy systems. This improved predictive capability allows for better planning and optimization of energy storage solutions, ensuring that surplus energy generated during peak periods can be effectively stored and utilized during low production periods. The integration of ML models with energy management systems also facilitates real-time adjustments to optimize the balance between energy production, storage, and consumption. Furthermore, the study highlights the potential of ML in enhancing the scalability and adaptability of HRES. By continuously learning from new data, ML models can adapt to changing environmental conditions and evolving system configurations, ensuring sustained efficiency and reliability. The application of machine learning to hybrid renewable energy systems modelling offers a transformative approach to optimizing system performance and improving energy storage solutions. This research underscores the importance of leveraging advanced ML techniques to enhance the integration and management of renewable energy sources, supporting the transition to a more sustainable and resilient energy future. Keywords: Energy Storage Solutions, ML, System Performance, Wind Energy Sources, Hybrid Renewable Energy System.
A large set of undrained compression triaxial tests was carried out on different types of cohesionless soils, from sands to silty sands and silts. Shear wave velocity measurements were also carried out. These tests exhibit distinct state transitions ranging from flow liquefaction to strain softening or strain hardening. With the purpose of defining a framework to assess soil liquefaction, it was found that the ratio between the shear wave velocity (V-S(0)) and the peak undrained deviatoric stress (q(peak)), V-S(0)/q(peak), could be accurately used to define a boundary between liquefaction and strain hardening for sands and between strain softening and strain hardening for silty sands and silts. Since this ratio is a function of the tested material, the prediction of these boundaries can be made as a function of soil grading, namely via the coefficient of uniformity, C-U. Despite not being regarded as a strong geomechanical parameter, C-U is easily determined from a grain-size distribution test and has an empirically proven correlation with critical state parameters.
The identification of transition field factor boundary between streamer initiation dominated breakdown and streamer propagation dominated breakdown is important for understanding the breakdown mechanisms of transformer liquids and for the transformer insulation design in practice. Most previous studies focused on either very divergent field or nearly uniform field. This article reports on breakdown voltage tests with three transformer liquids in moderately uniform fields. The results show that the transition field factor boundary between streamer initiation dominated breakdown and streamer propagation dominated breakdown under negative lightning impulse is liquid-dependent. It is 9.6 for a mineral oil, 7.3 for a gas-to-liquid (GTL) oil, and 42.0 for a synthetic ester. The difference in the transition field factor boundary helps explain the observed breakdown voltage distinction among the three transformer liquids. In addition, the characteristics of negative streamers leading to breakdown were also investigated in the moderately uniform fields. Streamer mode transitions with gap distance were observed and clarified, which is from 1st mode streamer to 2nd mode streamer in both the mineral oil and GTL oil, while it is the 2nd mode streamer to fast streamer in the synthetic ester.
Bubble formation in transformers could lead to their failure, resulting in severe outages in the electricity network. Hence, understanding the parameters that impact the bubble formation process is crucial to maintain the transformer’s dependable operation. Commonly, mineral oil and non-thermally upgraded Kraft paper have been used as the insulating materials for power transformers, and the risk of bubble formation has been studied for this material combination. However, new insulating liquids such as esters and gas-to-liquid technology-based oils are recently gaining more popularity and studies on bubble formation for these alternative liquids are limited. In this paper, the impact of alternative liquids on bubble formation has been investigated by using a small-scale test tube-based system. Studies have shown that when considering the water content in the paper at the time of bubble formation, the bubble formation temperature is comparable across different liquid types.
