INTELLIGENT ENERGY CONSUMPTION MANAGEMENT IN EDGE COMPUTING NETWORKS BASED ON GAME THEORY

With the development of edge computing and the Internet of Things (IoT) technologies, the relevance of optimizing energy consumption in distributed networks is increasing. Traditional energy management methods are often inefficient due to the lack of adaptation to dynamic load conditions and limited coordination capabilities between nodes. This paper presents a cooperative dynamic energy management model in edge computing networks based on game theory principles. The introduction identifies the main problems of modern edge computing systems, including high energy consumption, uneven load distribution, and limited adaptation capabilities to changing operating conditions. A review of current research in the field of energy-efficient management in edge computing is provided. The theoretical foundations section describes the key concepts underlying the proposed model. The principles of cooperative and non-cooperative games, Nash equilibrium, and energy optimization algorithms used for balancing node loads in the network are considered. The methodology section formulates a mathematical model that describes the behavior of edge computing network nodes as a dynamic game with incomplete information. The utility function of nodes is defined, taking into account energy consumption, quality of service (QoS), and data transmission costs. A mechanism for finding Nash equilibrium for optimal resource allocation among nodes is described. The research results demonstrate the effectiveness of the proposed model in reducing energy consumption and improving edge network performance. The proposed approach allows nodes to adaptively change their strategies depending on changes in load, minimizing overall energy costs without losing productivity. The use of machine learning in decision-making further enhances system efficiency. Practical applications of the model cover areas such as smart cities, IoT infrastructures, and autonomous energy management systems. The use of distributed decision-making mechanisms ensures stable and efficient operation of computing nodes even in resource-constrained environments. The conclusions summarize the key research findings and highlight prospects for further model development, including the integration of deep learning algorithms, expansion of cooperation mechanisms between nodes, and implementation of energy consumption forecasting technologies.