Research on User Abnormal Load Identification Method Based on Edge Computing

This paper proposes an edge computing-based method for user abnormal load identification to address the issues of insufficient real-time performance and limited computing resources in abnormal load detection within power systems. The study first explores in detail the application of edge computing in power systems and analyzes the limitations of traditional load identification methods in terms of real-time processing and accuracy. Based on these analyses, an integrated system architecture combining data collection, preprocessing, and abnormal load identification models based on machine learning and deep learning is designed and implemented. Strategies for efficient model deployment and resource optimization on the edge are also proposed. Experimental results show that the proposed method not only improves detection accuracy but also significantly reduces response delays, providing strong support for practical engineering applications.