Design of Comprehensive Energy System Considering Anomaly Detection of Multi-Energy Data

Considering miscellaneous data and difficult management and control of all kinds of energy equipment, a comprehensive energy management and control system is designed and a multi-energy data anomaly detection method is proposed. Data architecture of the comprehensive energy system is designed first, and then the major modules are designed, including energy monitoring, power supply and distribution monitoring, energy analysis and energy diagnosis of the system. In view of the system classification architecture, all kinds of user layer system models are designed and illustrated in detail. Finally, the spectral clustering algorithm is improved by replacing the original spectral clustering with small batch average clustering. The data detection results are obtained based on multi-energy data. Through experimental verification, the results show that the proposed method can effectively detect the multi-energy abnormal data.