Application of big data to lightning flashover warning in power grid

Abstract

With the rapid development of smart grid, the power system has entered the era of the big data. At present, some programs have been made in the application of the big data to certain aspects of the power system, but to lightning flashover warning is rarely mentioned. The study of lightning flashover warning nowadays is based on the establishment of lightning risk assessment procedures to carry out early warning. This paper innovatively applies the big data to the application of lightning flashover warming. The main steps of this paper are as follows: firstly, we need to collect a large number of the lightning history data, and build the K-Dimension tree based on them; then through the real-time monitoring of atmospheric electric field instrument, the new lightning data are obtained; afterwards, we use the K-Nearest Neighbor search algorithm to search for the nearest k-points to the new lightning data; after that, the warning results are output according to the K-Nearest Neighbor theory. At last, the method of this paper is illustrated in detail by 100 historical data, and the method is proved to be correct at the same time. This method, which combines the historical data to achieve the early warning, has the actual research value in the background of the big data.