In recent years, extreme weather and climate events such as lightning have been intensifying, causing certain adverse effects on the operation of the power grid. However, there are still shortcomings in the mechanism research, prediction, and early warning application of high-risk events such as lightning disasters. It is urgent to rely on electric power meteorological artificial intelligence technology to achieve the recognition of microscale characteristics of the external environment of the power grid and the analysis of the mechanism of equipment and facilities causing disasters, overcome the last mile problem from basic research to business applications, and effectively improve the protection and early warning capabilities of power grid lightning disasters. Therefore, this article conducts research on the identification and extraction techniques of key parameters related to lightning meteorological characteristics and lightning disasters in power grids. Firstly, the meteorological characteristics of lightning were elaborated, including the formation process of lightning and the formation process of ground flashes. Then, the key parameters of lightning disasters in the power grid, namely lightning current and ground flash density, were provided. Subsequently, the probability distribution of lightning current amplitude, as well as lightning distribution of the global region, in the surrounding areas of China, and in the jurisdiction of State Grid Corporation of China, were presented. The research results can lay the foundation for subsequent lightning warning.