There is few effective solving methods for the high-dimensional multi-objective optimization problem with interval parameters in real life. Therefore a new method is proposed using to concept of set. Firstly , three performance indicators including convergence, distribution and diversity of a solution set are used to transform the original optimization problem into a tri-objective one. Then, the set-based Pareto dominant relationships and a measure of uncertainty are used to distinguish the solutions of the converted optimization problem. Additionally, the adaptive crossover and mutation rates are suggested to improve the global and local search ability of the population. Finally, four benchmark function optimization problems are used to test the performance of the proposed method and comparison method. The test results show that, the uncertainty, diversity, and distributivity of a Pareto solution set obtained by the proposed method is superior to that by the contrast method, in addition to convergence.