• Energy Research

AbstractRecently, smart grids introduce significant challenges to power system protection due to the high integration with distributed energy resources (DERs) and communication systems. To effectively manage the impact of DERs on power networks, researchers are actively formulating adaptive protection strategies, requiring robust communication schemes. However, concerns remain over the occurrence of communication connection failures and the potential risks presented by cyber‐attacks. This work addresses these challenges by investigating the impact of cyber‐attacks on different adaptive overcurrent relays (OCRs) approaches. Here, modern adaptive OCR coordination approaches using different group settings has integrated in evaluating high voltage/medium/low voltage (HV/MV/LV) network model with real network parameters at the MV/LV level. Additionally, a voltage‐based relay is developed and employed to enhance protection system performance under various cyber threats, aiming to reduce tripping time and to minimize energy that is not supplied. The results show that voltage‐based scheme outperform the traditional adaptive OCRs in terms of response time and mis coordination events under cyber‐attacks. In the proposed MV/LV real network scenario characterized by an 89% availability of a 4 MW photovoltaic system, even a brief interruption caused by cyber‐attacks can result in significant cost consequences.

In recent years, the integration of Distributed Energy Resources (DERs) and communication networks has presented significant challenges to power system control and protection, primarily as a result of the emergence of smart grids and cyber threats. As the use of grid-connected solar Photovoltaic (PV) systems continues to increase with the use of intelligent PV inverters, the susceptibility of these systems to cyber attacks and their potential impact on grid stability emerges as a critical concern based on the inverter control models. This study explores the cyber-threat consequences of selectively targeting the components of PV systems, with a special focus on the inverter and Overcurrent Protection Relay (OCR). This research also evaluates the interconnectedness between these two components under different cyber-attack scenarios. A three-phase radial Electromagnetic Transients Program (EMTP) is employed for grid modeling and transient analysis under different cyber attacks. The findings of our analysis highlight the complex relationship between vulnerabilities in inverters and relays, emphasizing the consequential consequences of affecting one of the components on the other. In addition, this work aims to evaluate the impact of cyber attacks on the overall performance and stability of grid-connected PV systems. For example, in the attack on the PV inverters, the OCR failed to identify and eliminate the fault during a pulse signal attack with a short duration of 0.1 s. This resulted in considerable harmonic distortion and substantial power losses as a result of the protection system’s failure to recognize and respond to the irregular attack signal. Our study provides significant contributions to the understanding of cybersecurity in grid-connected solar PV systems. It highlights the importance of implementing improved protective measures and resilience techniques in response to the changing energy environment towards smart grids.

The advancement of sustainable energy sources necessitates the development of robust forecasting tools for efficient energy management. A prominent player in this domain, solar power, heavily relies on accurate energy yield predictions to optimize production, minimize costs, and maintain grid stability. This paper explores an innovative application of tiny machine learning to provide real-time, low-cost forecasting of solar energy yield on resource-constrained edge internet of things devices, such as micro-controllers, for improved residential and industrial energy management. To further contribute to the domain, we conduct a comprehensive evaluation of four prominent machine learning models, namely unidirectional long short-term memory, bidirectional gated recurrent unit, bidirectional long short-term memory, and simple bidirectional recurrent neural network, for predicting solar farm energy yield. Our analysis delves into the impacts of tuning the machine learning model hyperparameters on the performance of these models, offering insights to improve prediction accuracy and stability. Additionally, we elaborate on the challenges and opportunities presented by the implementation of machine learning on low-cost energy management control systems, highlighting the benefits of reduced operational expenses and enhanced grid stability. The results derived from this study offer significant implications for energy management strategies at both household and industrial scales, contributing to a more sustainable future powered by accurate and efficient solar energy forecasting.