Abstract
The aim of the study was to explore and analyse the potential of applying artificial intelligence for optimising energy distribution processes and identifying vulnerabilities in energy networks. The work focused on the study of methods, algorithms, and approaches that enabled increased efficiency in managing energy systems, reduced energy losses, improved network resilience to external threats, and ensured more accurate forecasting of supply and demand. Special attention was paid to the application of intelligent methods for detecting anomalies and vulnerable points in energy networks, which helped to respond promptly to potential cyberattacks, technical faults, or other risks. The study examined modern methods of energy flow management, particularly the use of neural network algorithms and blockchain technologies, as well as the integration into energy systems to enhance network efficiency and stability. The application of machine learning algorithms, such as convolutional and recurrent neural networks, significantly improved load forecasting accuracy and adaptability to changing network conditions. Load forecasting methods, including neural networks, decision trees, and reinforcement learning, contributed to reducing energy consumption and preventing overloads. At the same time, anomaly detection through intelligent systems allowed for the timely identification of faults and potential attacks, increasing system security and reliability. One of the promising solutions was the implementation of blockchain technologies for decentralised distribution of energy resources, which ensured transparency, security, and efficiency of operations. Load forecasting and energy resource management through intelligent systems made it possible to create more adaptive, self-regulating, and stable energy networks