Considering the importance, the frequency adjustment characteristic and the position of the load cut off on the overall effect of the low-frequency load shedding scheme, combined with the load shedding requirements of the cycle to be optimized and the action characteristics of devices in the system, an optimization method of low-frequency load shedding scheme considering the influence of power flow transferring on the power system is proposed. The optimization goals are designed to minimize power loss, minimize the unit power regulation of load cut off and maximize the flow transfer entropy, considering all kinds of the constraints, the Pareto optimal solution set is solved by using the fast and elitist Non-dominated Sorting Genetic Algorithm (NSGA-II). Then multiple attribute decision-making should be made for it through the gray relational analysis model to determine the most scientific and reasonable low-frequency load shedding scheme. Finally, the rationality and validity of the proposed method are demonstrated by an example of the New England 10-machine 39-node system. This work is supported by National Natural Science Foundation of China (No.51277076).