The Hubei power grid faces problems such as insufficient risk assessment of power grid equipment operation and delayed disaster response due to frequent meteorological disasters such as rainfall, lightning, severe convection, strong winds, and high temperatures. To address these issues, this paper constructs a spatiotemporal simulation and risk warning system for meteorological disasters based on a unified power grid map. The methods include: (1) collecting and preprocessing multi-source meteorological data to obtain real-time and historical data from the National Meteorological Center and the Hubei Provincial Meteorological Observatory; (2) using a combination of machine learning (ML) and statistical analysis algorithms to model the spatiotemporal evolution characteristics of meteorological types; (3) establishing a correlation model between meteorological types and power grid operating status (including load, voltage, and current) to calculate the probability of line faults under different meteorological conditions; and (4) introducing a visualization module to overlay meteorological risks, affected equipment, and disaster warning information onto the unified power grid map to achieve risk zoning and dynamic early warning. The results show that the system has a 92.8% accuracy rate in predicting line faults in a historical disaster sample of the past ten years, and the average early warning time has been improved to 3.14 hours, which can effectively support the power grid’s disaster prevention, mitigation and emergency dispatch decision-making.
Fan et al. (Thu,) studied this question.