Damage Inference of Truss Structure Based on Bayesian Updating

Abstract

In the long-term use of civil engineering structures, affected by natural disasters and man-made disasters, its performance gradually deteriorated, so it is very important to find structural damage in time. This paper discusses the problem of damage inference for truss structures, and uses Bayesian updating theory as a solution. By combining Bayesian updating with structural damage inference method, Monte Carlo sampling method is used for repeated variable experiment and analysis. The experimental results show that in a specific truss structure, when the load point is applied at a specific position, the experimental effect is the best. Moreover, with the increase of load and measuring points, the accuracy of experimental results is gradually improved. This method obviously improves the existing problems of the traditional method, and proves its effectiveness and accuracy. The combination of structural damage inference and Bayesian updating will bring more intelligent, reliable and efficient structural health monitoring and maintenance methods to the field of engineering construction. Realizing real-time reliability update and damage assessment of structures has a positive impact on improving the safety, sustainability and longevity of structures.