Application Of Particle Swarm Optimization In Damage Identification Of Grid Structure

In recent years,the country has vigorously developed steel structures.Due to its own advantages,large-span space grid structures have been increasingly used in national key projects and large industrial buildings.If the local structural damage of the grid structure is not detected and processed in a timely manner,it may endanger the safety of the entire structure and increase maintenance costs.At the same time,due to its special functional positioning and important social functions,the large-span space grid structure will once a safety accident occurs,Will endanger the safety of people 's lives and the safety of economic property;therefore,it is of great significance to establish a structural health monitoring system for the structure during construction and service;that is,to deploy sensors and use damage without affecting the use of the building Identification technology for monitoring,safety assessment and maintenance measures.The main research contents of this paper are: damage identification of grid structure and optimal sensor placement.This paper uses the improved particle swarm algorithm based on the simulation model of the grid structure as the background to study the application of the particle swarm algorithm in structural damage detection and optimal sensor placement.Particle swarm optimization is a new type of intelligent optimization algorithm,which has the advantages of outstanding processing optimization ability,easy to understand,and simple operation.Based on the improved particle swarm algorithm,this paper makes the following applications for damage identification and sensor optimization jobs:(1)The particle swarm algorithm is introduced in detail,and the improved particle swarm algorithm with inertial weight is selected as the optimization algorithm in this paper;the corresponding program of the particle swarm algorithm is written in MATLAB,and the implementation process of the particle swarm algorithm is introduced;The optimization performance of the particle swarm algorithm and the superiority of the improved particle swarm algorithm are illustrated.Four function examples are selected to optimize the calculation respectively.The results show that the particle swarm algorithm has a good optimization performance,and the improved particle swarm algorithm is more superior.(2)Optimization of the number of sensors for damage identification.Firstly,the experimental model of the grid structure was simulated.The finite element model was established using ANSYS software,and the modal analysis and modal extraction were performed.The number of target modals of the structure was selected based on the Fisher information matrix 2-norm method,and then determined.The vertical order of the structure is monitored.Finally,the number of sensors to be laid in the grid structure is determined using the MAC optimization criterion.(3)Sensor location optimization based on improved particle swarm algorithm.The fitness function of the particle swarm optimization algorithm is determined,and an optimization calculation is performed according to the selected mode of the monitoring mode order.Firstly,the fitness function based on the modal confidence criterion is defined.The results of multiple optimization calculations after parameter setting show that the algorithm has good stability and efficiency in the optimization of the sensor position.Then we define the adaptation based on the modal deformation energy.Degree function,the results of optimization calculations show that different fitness functions will affect the optimization performance of the algorithm.(4)Damage identification based on improved particle swarm algorithm.This method introduces the idea of model modification,mainly combining ANSYS software and MATLAB software.During the iterative optimization process,the particle swarm algorithm program constantly calls ANSYS program to modify the stiffness of the grid structure,and iterates continuously until it obtains the “false” which is closest to the real working conditions.Operating conditions." In the numerical simulation of damage identification,we set up three single-spot damage conditions for the upper chord,lower chord,and web,three different damage conditions for the same member,and three multiple-position damage conditions.The results show that the improved particle swarm algorithm has a good effect on the damage identification of grid structures.