In order to study the probability distribution characteristics of lightning peak current and the fitting effectiveness of cumulative probability for peak current, and the demands of lightning protection engineering design and lightning risk assessment, this article utilizes the peak current data observed by the LLS (Lightning Location System) from 2007 to 2013 to analyze the distribution characteristics of probability and density for peak current. The results show that there is a big difference in the cumulative probabilities for different polar peak currents; the distribution of negative lightning is more concentrated than that of positive lightning; the distribution curve of peak current cumulative probability for negative lightning is basically the same with the total one; and the peak currents of lightning are concentrated mostly in 10 to 50 kA. According to the distribution expression suggested by IEEE for cumulative probability of peak currents, this article fits the distribution expression of cumulative probability for different polar peak currents. The statistical analysis shows that when the peak current is less than 110 kA, the relative error between fitting cumulative probability calculated by the fitting expression of IEEE and the actual one is small, but when the peak current is above 110 kA, the error increases with the peak current. According to the segment revision expression given by this article, with the peak current being above 110 kA, the error between calculated cumulative probability and the actual one is less than 2%. When use the expression suggested by IEEE to fit the cumulative probability and density of peak currents, the fitting effectiveness for negative lightning and total one is better than the positive one. The reason seems to be connected with the distribution characteristics of positive lightning.