A Two-Stage Bridge Structure Damage Identification Research Based On Artificial Fish-Swarm Algorithm

As an important part of transportation, bridge structures inevitably suffer structural damages caused by various external forces during its service period. To guarantee the normal use of the bridge, it is essential to do long-term health monitoring and regular security tests on the bridge structures,pay close attentions to the development of structural damages. As a core problem in the field of structural health monitoring and detecting,damage identification technologies is becoming more and more important, it has become a new subject for engineers in the engineering field to find suitable damage identification methods.Based on the researches of the existing damage identification methods at home and abroad,this paper introduces a biota intelligent algorithm- artificial fish-swarm algorithm into bridge structural damage identification researches. The main contents of this paper are as follows:(1) An improved modal flexibility difference curvature method is put forward to do the analysis with modal flexibility difference curvature method and flexibility matrix diagonal matrix rate method on a simply supported beam bridge and a continuous beam bridge. The results show that the improved modal flexibility difference curvature method is better than the other two methods and can accurately identify bridge structural damage location.(2) The artificial fish-swarm algorithm is combined with the ANSYS finite element program written by ANSYS program development language-APDL language,which not only saves large-scale time for compiling finite element structural analysis programs, but also improves the applicability and efficiency of damage identification for large complex bridge structures.(3) This paper introduces the basic artificial fish-swarm algorithm to do the researches of damage identification. Some numerical simulations are done on a simple beam bridge. The first three-order modal frequencies and modal shapes of the structure are extracted and calculated. The results show that the basic artificial fish-swarm algorithm can do structural damage identification, but not effective for it directly, so they need to be improved.(4) This paper proposes a two-stage recognition method based on the basic artificial fish-swarm algorithm. It combines the basic artificial fish-swarm algorithm with the improved modal flexibility difference curvature method. Firstly, use the improved modal flexibility difference curvature method to determine the existence of damage and locate the damage; secondly, use the artificial fish-swarm algorithm to quantify the damage. Taking a steel truss bridge as an example, it carries out modal analysis,extracts the first three orders modal frequencies and modal shapes to do iterative calculation of damage identification by the artificial fish-swarm algorithm.The results show that the two-stage damage identification method can exclude the interference of damage interference units and can accurately identify the damage. Its damage identification accuracy and recognition results are better than the basic artificial fish algorithm.