Beam Structure Damage Detection Based On Improved Particle Swarm Optimization

In recent years,the collapse of buildings has occurred frequently,leading to the loss of people's lives and property of the country.The reason is that the damage can not be found and repaired in time.In order to ensure the safety and durability of the structure,it is necessary to monitor the damage of the existing structure.Using the effective damage identification method to obtain the damage information,so that the damage position can be repaired and reinforced to prevent the damage spreading.Therefore,it is of great significance to accurately detect damage.In this paper,particle swarm optimization(PSO)was used to detect the structural damage,and two methods were proposed to improve the inertia weight and the change of the particle position for the standard particle swarm algorithm which is easy to fall into the local minimum.The results of five test functions for optimization problems show that the performance of the improved PSO is better than that of the standard particle swarm optimization.At the same time,in order to improve the accuracy of identification,the objective function was established by using the frequency,vibration mode and flexibility matrix.Through the simulation of the damage condition of simply supported beam,the sensitivity of the three objective functions was compared,and it is found that the improved objective function is more sensitive to the damage.The frequency was added to the modal criterion in the form of a correction coefficient,taking into account the influence of frequency and preventing the objective function from being dominated by frequency.Considering there is a large number of elements for the real structure,the accuracy of calculation is affected.The phased method based on the improved PSO algorithm was used to detect damage in this paper.Firstly,calculate two damage indexes of the element damage variable and the element stiffness reduction factor,then use D-S evidence theory to fuse the two indexes for damage localization.Finally by using improved PSO algorithm,the localization results and the damage extent are detected precisely.The influence of the damage location,the number of damage elements,the damage extent and the noise level on the damage identification effect was studied by using Matlab software to simulate the finite element of simply supported beam.The identification results of the multi-damage case were compared with those of the standard PSO algorithm and the traditional objective function.The effectiveness of the improved algorithm and the improved objective function was verified.The vibration tests were made on a reinforced concrete beam with rectangular section and an I-steel beam.The first three-order modal parameters of the structure were extracted from the measured signal by using the random subspace identification method,and then the identification method based on improved PSO algorithm was used to detect the damage.The results show that the precision of this method is better for the case of large damage extent;while it has slightly larger identification error when the damage extent is small and the number of the damaged elements is large.