Considering the frequency regulation characteristics, power factor, size and position of the restored load on the end system security when the high voltage direct current (LCC-HVDC) makes its access, combined with the operation characteristics of load input during the system restoration process, a method for load restoration optimization considering the influence of DC access on the system security is proposed. The optimization goals are designed to minimize the frequency deviation, minimize the voltage offset of the nodes and maximize the power flow entropy of the receiving grid, the fast non-dominated sorting genetic algorithm (NSGA-II) is used to solve the Pareto optimal solution set of load restoration when considering the constraints of unit output and line power flow; then the subjective and objective weights of each attribute are determined according to the triangular fuzzy number and the information entropy method, the most satisfactory solution is selected by the gray relational analysis model. Finally, the validity of the proposed method is verified by an example of the New England 10-machine 39-node system with HVDC channel.