Showing papers on "Smart grid published in 2025"
TL;DR: This study provides a comprehensive overview of smart grid cybersecurity, examining threats, cyberattacks, and historical incidents, and proposes a classification system to facilitate organized solutions, while highlighting future research trends and open issues in the field.
Abstract: Despite the fact that countless IoT applications are arising frequently in various fields, such as green cities, net-zero decarbonization, healthcare systems, and smart vehicles, the smart grid is considered the most critical cyber–physical IoT application. With emerging technologies supporting the much-anticipated smart energy systems, particularly the smart grid, these smart systems will continue to profoundly transform our way of life and the environment. Energy systems have improved over the past ten years in terms of intelligence, efficiency, decentralization, and ICT usage. On the other hand, cyber threats and attacks against these systems have greatly expanded as a result of the enormous spread of sensors and smart IoT devices inside the energy sector as well as traditional power grids. In order to detect and mitigate these vulnerabilities while increasing the security of energy systems and power grids, a thorough investigation and in-depth research are highly required. This study offers a comprehensive overview of state-of-the-art smart grid cybersecurity research. In this work, we primarily concentrate on examining the numerous threats and cyberattacks that have recently invaded the developing smart energy systems in general and smart grids in particular. This study begins by introducing smart grid architecture, it key components, and its security issues. Then, we present the spectrum of cyberattacks against energy systems while highlighting the most significant research studies that have been documented in the literature. The categorization of smart grid cyberattacks, while taking into account key information security characteristics, can help make it possible to provide organized and effective solutions for the present and potential attacks in smart grid applications. This cyberattack classification is covered thoroughly in this paper. This study also discusses the historical incidents against energy systems, which depicts how harsh and disastrous these attacks can go if not detected and mitigated. Finally, we provide a summary of the latest emerging future research trend and open research issues.
11 citations
TL;DR: This review examines solar PV integration with smart-grids, highlighting challenges, standards, and grid codes. It emphasizes the need for updated regulatory frameworks to ensure system stability and efficient renewable energy integration, promoting a sustainable and low-carbon energy future.
Abstract: Promoting a sustainable and low-carbon energy future through the integration of renewable energy is essential, yet it presents significant challenges due to the intermittent nature of resources such as solar and wind. This paper examines the technological and economic dimensions of AC, DC, and smart grids, concentrating on the optimization of costs, efficiency, stability, and scalability. Smart grids, enhanced by AI, IoT, and blockchain technologies, play a vital role in energy management optimization, predictive maintenance, and secure energy transactions. Furthermore, the incorporation of renewable energy sources, especially photovoltaics, presents challenges including intermittency, voltage fluctuations, and grid congestion. This paper emphasizes the necessity for updated grid codes and policies that guarantee system stability and the effective functioning of renewable energy systems. The implementation of these regulatory frameworks is crucial for facilitating the efficient integration of renewable energy into the grid, ensuring a reliable and secure power supply while advancing sustainability efforts.
5 citations
4 citations
2 citations
2 citations
TL;DR: This study proposes the Deep Contrastive Variational Network (DCVN) for unsupervised detection of False Data Injection Attacks in dynamic smart grids, leveraging a novel loss function that optimizes latent space contrast and anomaly detection through noise contrastive estimation.
Abstract: False Data Injection Attacks (FDIAs) pose a critical threat to the reliability of power systems, especially under dynamic operational conditions like line outages that cause data distribution changes and concept drift. Traditional supervised methods depend on labeled datasets, which are costly and impractical for real-time application, and often fail to adapt to new attack vectors and operational changes without extensive retraining. To address these challenges, we design the Deep Contrastive Variational Network (DCVN), an unsupervised learning framework engineered to detect FDIA without requiring labeled data or assumptions about network topology. The DCVN framework starts with a Deep Belief Network (DBN) for robust feature extraction from raw power system data, capturing underlying patterns in an unsupervised manner. Recognizing the limitations of DBNs in managing FDIA complexities under varying conditions, we enhance our model by integrating a modified Variational Autoencoder (VAE) with Noise Contrastive Estimation (NCE). This integration creates a novel loss function that optimizes the latent space of the VAE to enhance the contrast between normal operations and anomalies. The NCE component specifically sharpens the model's sensitivity to anomalies by maximizing the distinguishability of data representations. This design enables the DCVN to dynamically learn robust features, not just by minimizing reconstruction error but by enhancing anomaly detection through unsupervised learning. We provide a theoretical foundation for our method, ensuring performance guarantees in the dynamic environments of power grids. Empirical results demonstrate that the DCVN model significantly surpasses traditional approaches, offering robust detection capabilities.
TL;DR: This study presents a real-time MPC-based energy management system for a medium-sized hybrid microgrid, achieving up to 43% cost reduction and 35% grid dependency decrease while ensuring reliable critical load supply under dynamic pricing and fault conditions.
Abstract: As hybrid microgrids become increasingly widespread in real-world applications, the need for intelligent energy management strategies that ensure reliability, economic efficiency, and robustness to uncertainties is growing. This study presents a real-time capable model predictive control (MPC)-based energy management for a medium-sized hybrid microgrid at the Karabuk University Demir Çelik campus. The system comprises 100 kW photovoltaic (PV) panels, a 500 Ah battery energy storage system (BESS), a 440 kW diesel generator, and a 75 MVA utility connection. The proposed MPC approach is evaluated under ten realistic operating scenarios, incorporating dynamic pricing and fault conditions. Simulation results show up to 43% reduction in operational costs and 35% decrease in grid dependency, while keeping unserved critical loads below 3%. Compared to conventional rule-based methods, the proposed strategy offers improved scalability, adaptability, and resilience, highlighting its practical potential for deployment in smart energy systems.
TL;DR: This research integrates hyperlocal weather predictions with demand response in smart grids using machine learning, enhancing power generation forecasts with a hybrid model (SARIMAX, Prophet, Holt-Winters) achieving 0.96 R-squared and real-time user-specific forecasting insights.
Abstract: Optimizing power generation is crucial in the evolving smart grid and urban ecosystem landscape. Efficient energy use is vital in today's tech-driven world, with smart grids leading the way in managing energy consumption and distribution. Given the link between climate and energy use, integrating weather data is now essential. This research presents a method combining demand response in smart grids with hyperlocal weather predictions using machine learning to enhance power generation forecasts. Five models—SARIMAX, Prophet, XGBoost, LSTM, and Holt–Winters are trained on the smart grid and weather data, including precipitation, humidity, temperature, and cloud cover. Extensive feature engineering, such as lag terms and cyclic encoding, improved accuracy. SARIMAX outperformed others with an MAE of 3.54 and an R-squared of 0.97, while the hybrid model, combining SARIMAX, Prophet, and Holt–Winters, achieved an R-squared of 0.96. The hybrid model's dynamic recalibration improved accuracy with new data. A real-time dashboard is also developed, offering dynamic, user-specific forecasting insights. This research enhances smart grid management by integrating machine learning and demand response for more efficient energy use.
15 Dec 2025
TL;DR: This study proposes a hybrid DNN-residual tree XGBoost model for smart grid load forecasting, achieving a high R² value of 0.9985, outperforming benchmark models, and demonstrating potential for accurate short-term load forecasting in smart grids with high renewable penetration.
Abstract: Precise active power demand prediction is important for the reliable power scheduling and operation of modern smart grids. This study proposes a hybrid Deep Neural Network (DNN) with a residual tree-based gradient boosting ensemble to predict the nonlinear time-varying electricity demand profile. The hybrid framework combines the powerful feature extraction capability of deep learning with the high prediction accuracy and interpretability of boosted decision trees. Actual power demand data from the AEMO New South Wales 2024 dataset was used for the training, validation, and testing of the model. The proposed hybrid approach was compared with the benchmark models, such as DNN, ANN, LSTM-GRU, and random forest. The results show the superior performance of the hybrid model. The hybrid DNN–residual tree XGBoost achieved a high coefficient of determination R2 value of 0.9985, which demonstrates that the proposed predictive model can accurately capture the complex consumption behaviors. The results highlight the potential of incorporating deep-learning models into ensemble tree methods for short-term load forecasting in smart grids. The proposed method provides a robust, scalable, and data-driven decision-making system for intelligent energy management, demand-side planning, and operational reliability, especially in smart grids with considerable penetration of distributed and renewable resources.
23 May 2025
TL;DR: This study presents a blockchain-based smart contract system (BSCS) that optimizes energy routing in smart grids by minimizing transmission losses, ensuring secure and decentralized energy transfers, and reducing gas consumption by 12% with improved latency and throughput.
Abstract: The integration of renewable resources and prosumers into smart grids poses challenges related to scalability, transparency, and transmission efficiency. Centralized routing frequently depends on expensive technology and experiences significant losses. This study presents a blockchain-based smart contract system (BSCS) that reduces transmission losses while guaranteeing secure and decentralized energy transfers. The smart grid is represented as a weighted directed graph, with edges denoting actual power losses. Dijkstra’s shortest path algorithm generates optimal paths from the generator to the consumer with minimal loss. These optimal pathways are permanently documented and regulated by permissioned blockchain smart contracts, ensuring tamper-proof and verifiable energy settlement. Validation is performed using an enhanced IEEE 58-bus test system, which is based on the standard IEEE 57-bus network, by incorporating an additional synthetic consumer node (Bus 58) linked to Bus 12 to simulate a flexible prosumer load of 1.5 MW + 0.5 Mvar. Additionally, the synthetic consumer node employs Ganache, Truffle, and Solidity for its implementation. This modification facilitates the assessment of dynamic energy routing and decentralized transaction settlement in extended topology scenarios. The proposed BSCS demonstrates substantial enhancements compared to baseline blockchain systems. Active power losses in transmission lines are diminished, gas consumption declines by approximately 12%, latency is enhanced by as much as 21%, and throughput increases by more than 30%. The rapid deployment and execution of smart contracts within sub-second intervals validate the system's appropriateness for real-time grid operations. The proposed technique combines graph-theoretic optimization with blockchain governance to provide a safe, scalable, and hardware-independent framework for decentralized energy markets.
TL;DR: This paper proposes a dynamic event-triggered control mechanism for multiple electric vehicles in smart grids, reducing data transmission while ensuring frequency performance, and derives sufficient conditions based on Lyapunov stability theory.
Abstract: ABSTRACT This paper investigates the dynamic event-triggered control for the frequency service of EVs in smart grids (SGs). In this SG, multiple EVs are introduced, where their behavior modelling depends on the state of charge control. A novel dynamic event-triggered mechanism is designed to reduce the data transmission in SGs. By developing a reasonable dynamic threshold in advance, this communication protocol is able to reduce the transmitted data while ensuring the frequency performance of the SGs. Based on Lyapunov stability theory, some sufficient conditions are derived. Finally, two simulation examples are provided to verify the effectiveness of the proposed method.