The present paper addresses the subject of structural DI (damage identification). ACO (Ant Colony Optimization) is a kind of new intelligent optimization algorithm which is based on simulation of ant group behavior. This algorithm is characterized with strong robustness, distributed computing mechanism and higher efficiency. In this paper, firstly, the feasibility and principle to make DI by FE (Finite Element) method is introduced. Then, the structure of ACO algorithm used in TSP (Traveling Salesman Problem) structure is also illustrated and the flow chart of the program is given. Furthermore, SACO (substructure ACO) method is proposed to improve efficiency and applied in a three-bay three-story frame. In conclusion, on the one hand, efficiency about ACO algorithm applied into structure DI is proved to be high during the searching process, on the other hand, the new method of SACO is applied during the whole calculation, and the conclusion indicates that it could improve the calculating efficiency and stability instead of too much complicated analysis for enter the local optimization search scope beforehand.