Application Research Of Structural Damage Detection Based On Vibration Testing

Damage detection and maintenance of structures are of great significance for the safe use and economy of structures.In recent decades,more and more domestic and foreign scholars have begun to study and research them.With the rapid development of society,small ordinary structures can no longer meet the needs of people,while a large number of complex large structures appear,almost every city has its own representative buildings,and these large structures have been put into use to meet the needs of people.However,because of the complex structure of these large buildings,it is not easy to find the damage.Once the structure accidents occur,it will result in casualties and huge economic losses.Therefor it is self-evident that the damage identification of the structural is very important.Aiming at the input and output of the mathematical model of structural damage detection-the objective optimization function with constraints,this paper focuses on the following aspects:1)In this paper,particle swarm algorithm and artificial fish swarm algorithm are mixed,the advantages of each algorithm are taken to make up for the shortcomings of each algorithm,and a better Particle artificial Fish Swarm algorithm is obtained.2)Taking a 2.5m long steel beam as the experimental structure to do the dynamic experiment,including the complete working condition,the single point damage condition,the two point damage condition,the measurement method of the single point excitation with the multi-point vibration is used to carry out the measurement,and the steel beam modal parameter can be obtained.The finite element software is used to model,simulate the experimental steel beam,extract the frequency and mode of the steel beam under the complete working conditions.Comparing with the experimental data,observing the size of the error,the fish swarm algorithm is used to modify the model.3)Particle Swarm Optimization(PSO)and Particle Fish Swarm Optimization(PSO)are used to identify the damage of steel beam.Both particle algorithm and particle fish swarm algorithm can accurately find the location and the size of the damage while identifying the single point damage.Particle algorithm and particle fish swarm algorithm can also both determine the damage with the two point damage identification,but particle fish swarm algorithm has higher accuracy and faster convergence speed.In three-point damage identification,the particle algorithm is not accurate in judging the existence of the three damage,not only can the damage of three points can be found at every iteration,but also the size of the damage is not accurate.Particle fish swarm algorithm can accurately determine the location of the damage with less estimating error of the damage size,which proves the effectiveness of the improved algorithm.