Abstract:

With the continuous advancement of intelligent technologies, data mining plays a crucial role in anomaly detection for thermal energy storage systems, serving as a key technological means to enhance energy management and system security. This article first analyzes potential anomaly patterns within large datasets through in-depth mining, thereby improving the intelligent adaptive capabilities and early fault warning systems of thermal energy storage systems. It introduces several anomaly detection methods based on supervised learning, unsupervised learning, and deep learning techniques. Furthermore, it proposes specific applications of anomaly detection in industrial thermal energy storage systems, explores practices for anomaly detection and optimization using big data, and discusses practical applications of cloud computing-based monitoring and early warning systems for thermal energy storage anomalies. The aim is to promote the comprehensive development of intelligent, precise, and adaptive thermal energy storage systems, thereby enhancing the reliability and security of energy systems.

Key words: data mining, thermal storage, anomaly detection, supervised learning, unsupervised learning