The random fluctuations in the output of new energy increase the risk of power system operation. This paper proposes a power system probabilistic power flow calculation method based on LHS-CGC method, which solves the problem that the probability density function is negative when fitting data with A-type Gram Charlie Series expansion. Taking the calculation results of the Monte Carlo method as a reference, the accuracy and effectiveness of the proposed method were verified in an improved IEEE30 testing system, with higher computational speed. Then, the probability density distribution of each state variable and output variable of the system in three typical operation modes is checked according to the actual power grid model. This article analyzes the probabilistic load flow calculation results of the system under different operating modes and believes that the seasonal output changes of new energy have a more significant impact on the load flow of the new power system, and the risk of branch load flow exceeding the limit increases. Similarly, the node voltage distribution range under the new energy generation mode is larger, reaching 0.92 to 1.03 p.u.. This provides a reference for practical power optimization scheduling problems in engineering.