Transmission lines are exposed to the atmospheric environment for a long time and are susceptible to lightning, typhoons, rainstorms and other catastrophic weather and failures. Wind damage is a more hazardous meteorological disaster in Henan Province and the spatial and temporal distribution of its hazards is of great significance to wind defence. Firstly, based on the raw data of high wind disaster statistics of substations in Henan Province in recent years, a model of transmission line fault probability under strong convective high wind disaster weather conditions was established; secondly, principal component analysis was used to analyse the raw data, extract the principal components and establish a high wind disaster hazard level prediction model. Cross-validation was used to draw training samples and prediction samples for multi-classification support vector machine prediction, and the results showed that the prediction accuracy could reach over 75%. The results show that the hazard warning model developed in this paper meets the needs of systematic risk assessment and can provide a computational model basis for the study of hazard warning of high wind hazards.

wind warning; strong convective gale; multi-classification support vector machine; early warning model