A Comprehensive Review of the Current Status of Smart Grid Technologies for Renewable Energies Integration and Future Trends: The Role of Machine Learning and Energy Storage Systems

Energies, Journal Year: 2024, Volume and Issue: 17(16), P. 4128 - 4128

Published: Aug. 19, 2024

The integration of renewable energy sources (RES) into smart grids has been considered crucial for advancing towards a sustainable and resilient infrastructure. Their is vital achieving sustainability among all clean sources, including wind, solar, hydropower. This review paper provides thoughtful analysis the current status grid, focusing on integrating various RES, such as wind grid. highlights significant role RES in reducing greenhouse gas emissions traditional fossil fuel reliability, thereby contributing to environmental empowering security. Moreover, key advancements grid technologies, Advanced Metering Infrastructure (AMI), Distributed Control Systems (DCS), Supervisory Data Acquisition (SCADA) systems, are explored clarify related topics usage technologies enhances efficiency, resilience introduced. also investigates application Machine Learning (ML) techniques management optimization within with techniques. findings emphasize transformative impact advanced alongside need continued innovation supportive policy frameworks achieve future.

Language: Английский

---

Revolutionizing solar energy resources: The central role of generative AI in elevating system sustainability and efficiency

Applied Energy, Journal Year: 2025, Volume and Issue: 382, P. 125296 - 125296

Published: Jan. 13, 2025

Language: Английский

---

Evaluating and comparing machine learning approaches for effective decision making in renewable microgrid systems

Results in Engineering, Journal Year: 2024, Volume and Issue: 21, P. 101888 - 101888

Published: Feb. 10, 2024

This study addresses the critical issue of energy management in micro-grid (MG) systems incorporating renewable sources and hydrogen storage. The research introduces an innovative approach by conducting a comparative analysis two machine learning methods, namely k-Nearest Neighbors (k-NN) Random Forest (RF), to optimize decision-making. investigation reveals consistent superiority Forest, particularly precision F1-scores, across key components such as fuel cell relay, battery super-capacitor grid system relay. results demonstrate that RF method consistently achieves high macro average factors (90%, 86%, 84%, 82%) impressive F1-scores 87%, 88%, 85%), surpassing performance k-NN, which yields notably lower (30%, 15%, 14%, 28%) (41%, 23%, 26%, 34%). superior positions robust for decision-making, specifically realm storage sources. novelty this work lies establishing reliable tool capable handling intricacies thereby enhancing sustainable processes. Additionally, resilience imbalanced data adds its effectiveness diverse operational scenarios. sheds light on potential contribute significantly advancement solutions systems.

Language: Английский

---

Contributions of artificial intelligence and digitization in achieving clean and affordable energy

Intelligent Systems with Applications, Journal Year: 2024, Volume and Issue: 22, P. 200389 - 200389

Published: May 19, 2024

Concerned by the continuous decline in quality of life, poverty, environmental degradation, and escalated war conflicts, United Nations 2015 instituted 17 Sustainable Development Goals (SDGs) 169 targets. Access to clean, modern, affordable energy, also known as SDG 7, is one goals. Universal access electricity metrics for measuring a good life it fundamentally affects education, healthcare, food security, job creation, other socioeconomic indices. To achieve this goal targets, there has been increased traction research, development, actionable plans, policies, activities governments, scientific community, environmentalists, development experts, stakeholders achieving goal. This review presents various avenues which AI digitization can provide impetus 7. The global trends attaining clean electricity, cooking fuel, renewable energy efficiency, international public financial flows between 2005 2021 are reviewed while contribution towards meeting 7 highlighted. study concludes that deployment into sector will catalyze attainment 2030, provided ethical issues, regulatory concerns, manpower shortage, shortcomings effectively handled. recommends adequate infrastructural upgrades, modernization data collection, storage, analysis capabilities, improved awareness professional collaborative innovation, promotion legal issues ways advancing universal 2030. Going forward, more collaborations academic research institutions producers help produce experts professionals propel innovative digital technologies sector.

Language: Английский

---

Enhanced Fault Detection and Localization Strategy for High-Speed Protection in Medium-Voltage DC Distribution Networks Using Extended Kalman Filtering Algorithm

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 30329 - 30344

Published: Jan. 1, 2024

Conventional strategies are not effective in addressing the complex protection challenges medium-voltage DC distribution networks (MVDCDN). The main challenge MVDCDN is high-rising fault current, requiring a robust and fast strategy. This paper proposes use of an Extended Kalman filter (EKF) to detect various types faults using only current signal MVDCDN. In first stage, signals from positive negative poles corresponding bus obtained. EKF then applied measured DC-current generate two detection indices. index cumulative residuals (CR), calculated iterative differencing process with updated estimated state noisy measurement. second modified version total harmonic distortion, known as distortion factor (DCDF). classification/zone identification (FCZI) unit activated if changes CR DCDF detected within observation window relay. FCZI calculates filter-based predicted energy (EKFBPE) for faulty line section at both ends. polarity EKFBPE used classification localization decisions. proposed strategy requires low-band wireless communication capability smart grid. Extensive simulations MATLAB Ⓡ Simulink 2022b conducted on ±2.5 kV three feeders, considering scenarios. results demonstrate that scheme achieves 99.9% accuracy, under radial, looped, meshed topology highly resilient different time operation 1 msec. scalability method its effectiveness handling higher voltage levels associated uncertainty will investigate future research.

Language: Английский

---

Failure analysis in predictive maintenance: Belt drive diagnostics with expert systems and Taguchi method for unconventional vibration features

Heliyon, Journal Year: 2024, Volume and Issue: 10(13), P. e34202 - e34202

Published: July 1, 2024

Predictive maintenance to avoid fatigue and failure enhances the reliability of mechanics, herewith, this paper explores vibrational time-domain data in advancing fault diagnosis predictive maintenance. This study leveraged a belt-drive system with properties: operating rotational speeds 500–2000 RPM, belt pretensions at 70 150 N, three operational cases healthy, faulty unbalanced, which leads 12 studied cases. In analysis, two one-axis piezoelectric accelerometers were utilized capture vibration signals near driver pulley. Five advanced statistics calculated during signal processing, namely Variance, Mean Absolute Deviation (MAD), Zero Crossing Rate (ZCR), Autocorrelation Coefficient, signal's Energy. The Taguchi method was used test five selected features on basis Signal-to-Noise (S/N) ratio. For classifications, an expert based artificial intelligence where Random Forest (RF) model trained untraditional parameters for optimizing accuracy. resulted 0.990 0.999, accuracy AUC, demonstrate RF model's high dependability. Evidently, methodology highlights potential when progressed into systems, advances strategies systems.

Language: Английский

---

Machine Learning Models for Solar Power Generation Forecasting in Microgrid Application Implications for Smart Cities

Sustainability, Journal Year: 2024, Volume and Issue: 16(14), P. 6087 - 6087

Published: July 17, 2024

In the context of escalating concerns about environmental sustainability in smart cities, solar power and other renewable energy sources have emerged as pivotal players global effort to curtail greenhouse gas emissions combat climate change. The precise prediction generation holds a critical role seamless integration effective management systems within microgrids. This research delves into comparative analysis two machine learning models, specifically Light Gradient Boosting Machine (LGBM) K Nearest Neighbors (KNN), with objective forecasting microgrid applications. study meticulously evaluates these models’ accuracy, reliability, training times, memory usage, providing detailed experimental insights optimizing utilization driving forward. comparison between LGBM KNN models reveals significant performance differences. model demonstrates superior accuracy an R-squared 0.84 compared KNN’s 0.77, along lower Root Mean Squared Error (RMSE: 5.77 vs. 6.93) Absolute (MAE: 3.93 4.34). However, requires longer times (120 s 90 s) higher usage (500 MB 300 MB). Despite computational differences, exhibits stability across diverse time frames seasons, showing robustness handling outliers. These findings underscore its suitability for applications, offering enhanced strategies crucial advancing sustainability. provides essential sustainable practices lays foundation cleaner future, emphasizing importance accurate planning operation.

Language: Английский

---

Improving power output wind turbine in micro-grids assisted virtual wind speed prediction

Sustainable Operations and Computers, Journal Year: 2024, Volume and Issue: 5, P. 119 - 130

Published: Jan. 1, 2024

Wind energy is an alternative form of easily obtainable in the landscape. However, main challenge to extract electrical power from varying wind speeds. can be a significant production resource for electronics technologies, converters, and generators. Due their dependence on speed, output turbines experiences severe fluctuations with change ripples increase turbine. Therefore, engineers' critical research prediction will smooth these extraction fluctuations. Several speed methods have been used reduce changes turbines. One fast storage system that charged discharged seconds. Applying overcome slowness source primary approach considered this article. Also, turbine nominal 50 kW ultra-capacitor are determined, sources made MATLAB/SIMULINK softwareIn study, control signal adjusting pitch angle derived both actual predicted data. The data undergoes multiplication by 0.8, while multiplied 0.2. This serves two purposes: firstly, it helps prevent overshooting at initial stages, ensuring smoother transition. Secondly, aids maintaining consistent during subsequent moments. By combining weighted manner, achieves balanced response, effectively managing dynamics. Finally, results show utilizing improve Micro-grids (MGs) source's storage.

Language: Английский

---

Energy Intelligence: A Systematic Review of Artificial Intelligence for Energy Management

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(23), P. 11112 - 11112

Published: Nov. 28, 2024

Artificial intelligence (AI) and machine learning (ML) can assist in the effective development of power system by improving reliability resilience. The rapid advancement AI ML is fundamentally transforming energy management systems (EMSs) across diverse industries, including areas such as prediction, fault detection, electricity markets, buildings, electric vehicles (EVs). Consequently, to form a complete resource for cognitive techniques, this review paper integrates findings from more than 200 scientific papers (45 reviews 155 research studies) addressing utilization EMSs its influence on sector. additionally investigates essential features smart grids, big data, their integration with EMS, emphasizing capacity improve efficiency reliability. Despite these advances, there are still additional challenges that remain, concerns regarding privacy integrating different systems, issues related scalability. finishes analyzing problems providing future perspectives ongoing use EMS.

Language: Английский

---

Ultra High-Speed Fault Diagnosis Scheme for DC Distribution Systems Based on Discrete Median Filter and Mathematical Morphology

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 45796 - 45810

Published: Jan. 1, 2024

Language: Английский

---