Design of Multi Source Heterogeneous Fault Fusion System for Power Grid Based on 3D Visualization Platform

This research proposes an intelligent power grid fault analysis system based on multi-source heterogeneous data fusion through the development of a new-generation emergency command system and integrated utilization of multidisciplinary resources. By consolidating real-time monitoring data including total accident signals, dispatching automation systems, distribution automation systems, and electricity consumption information collection systems, a comprehensive fault analysis model spanning transmission networks, distribution networks, and end-user sides has been established. For transmission-level faults, the system employs an integrated analysis methodology combining multi-level time-series analysis with topological correlation to achieve precise fault localization based on total accident signals. At the distribution network level, differentiated analysis strategies have been developed for full feeder outages, branch circuit outages, and single transformer outages, incorporating grid topology relationships and load characteristic verification to form a closed-loop fault analysis process. Furthermore, a power grid digital twin platform has been constructed utilizing the Cesium 3D geographic visualization engine, enabling multidimensional rendering and spatial analysis of faulty equipment. Through technological innovations including temporal feature extraction, signal fusion, and stratified sampling validation, the system has achieved over 92 % fault analysis accuracy while reducing average processing time to under two minutes, thereby providing intelligent decision support for rapid fault response and power supply restoration in power grid operations.