Research On Optimal Layout Of Bridge Sensor Measuring Points Based On Improved Artificial Bee Colony Algorithm

As the scale of bridge structure is getting larger and larger,the stress characteristics tend to be complicated,so the establishment of bridge health monitoring system can monitor its working state to the maximum extent.Among them,the optimal arrangement of sensor measuring points is directly related to the collection effect of monitoring system information.How to meet the needs of bridge health monitoring with limited measuring points is the key link of bridge monitoring system information collection.As a member of swarm intelligence algorithm,ABC algorithm has a wide range of applications and is easy to implement.It can be used to deal with bridge sensor optimization problems.However,it has some shortcomings such as "premature" and easy to fall into local optimization.In this paper,the standard artificial bee colony algorithm(ABC algorithm)is improved,its various performances are tested by numerical experiments,and the sensor optimization layout of actual bridge engineering is carried out.The main research is as follows:This paper introduces the research status of bridge health monitoring system and optimal sensor placement of bridge structures,and makes a detailed study on the principles of optimal sensor placement methods: particle swarm algorithm,genetic algorithm and standard ABC algorithm.In order to overcome the problem that the standard ABC algorithm is easy to fall into local optimization,improve its convergence efficiency and enhance its optimization ability,the orthogonal initialization method is introduced to improve the uniformity of the initial solution distribution of the algorithm,and a new search strategy following bees is proposed to expand the search range of the algorithm.Four kinds of soft computing: particle swarm optimization algorithm,genetic algorithm,standard ABC algorithm and improved ABC algorithm are programmed by MATLAB program.Taking four functions such as Shpere,Rosenbrock,Step and Eggholder as numerical experimental platforms,the four algorithms are respectively used for comparative experimental analysis to analyze their optimized performance and applicable scope.The experimental results show that the improved ABC algorithm has better robustness and convergence.It can find the global optimal solution for these four functions,and can effectively deal with continuous or discontinuous,unimodal or multimodal function problems.For Qingshui River Bridge with a total length of 540 m,Midas Civil2019 is adopted to establish a finite element model,then modal analysis is carried out to extract relevant modal information,and then particle swarm optimization algorithm,genetic algorithm and improved ABC algorithm are respectively adopted to optimize the arrangement of longitudinal displacement measuring points of main girder and longitudinal inclination measuring points of pylon.According to the optimization results,the comprehensive layout scheme is determined.In a word,the standard ABC algorithm is improved to improve its convergence and global,and has better robustness.It can be used to solve continuous or discontinuous,unimodal or multimodal function problems.It provides a new and effective method for the arrangement of bridge sensor measuring points.