The invention relates to a stochastic dynamic unit commitment method for an electric power system based on a Newton method for solving the quantile, and belongs to the technical field of the operationof the electric power system. The method comprises the steps of building a chance-constrained stochastic dynamic unit commitment model for determined parameters of the electric power system; solvingthe quantile of stochastic variables obeying mixed Gaussian distribution by using the Newton method, and converting the chance constraints into deterministic linear constraints so as to convert the original problem into a mixed integer linear constraint optimization problem which is easy to solve; and finally solving a unit commitment optimization model, and obtaining a unit start-stop strategy and an active power plant. The chance constraints containing the risk level and stochastic variables are converted into the deterministic linear constraints by using the Newton method, so that the modelsolving efficiency is improved, the conservativeness of the traditional robust unit commitment is eliminated, and a more reasonable unit start-stop and active power plan is provided for decision makers. The stochastic dynamic unit commitment method can be applied to the unit commitment of the electric power system including large-scale renewable energy grid connection.