Design and testing of a decision tree algorithm for early failure detection in steel truss bridges

Barros, Brais; Conde, Borja; Cabaleiro, Manuel; Riveiro, Belén

doi:10.1016/j.engstruct.2023.116243

Cited by 14 publications

(2 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the use of civil engineering buildings and structures, fatigue of materials and damage to members cannot be avoided. For the damage identification of trusses, scholars have proposed methods based on deep neural networks, decision tree algorithms, statistical steady-state strain eigenfunctions [1][2][3]. Bayesian updating is an emerging method in civil engineering in recent years.…”

Section: Introductionmentioning

confidence: 99%

Study on the truss damage measurement based on Bayesian updating

Zhang

2024

ACE

View full text Add to dashboard Cite

Bayesian updating can integrate new observations in real time. This online updating property makes the Bayesian approach particularly suitable for dynamic monitoring of structural damage. In this paper, the node displacements are obtained by the finite element software OpenSees, assuming that the bars are damaged and the stiffness is reduced. Rejection Sampling method is used to determine the stiffness drop interval. During the sampling process, different load sizes, load locations, errors and types are selected to realize the updating of the stiffness of the top chord bar, bottom chord bar, vertical web bar, and diagonal web bar. A Pratt-type truss is used as an example, combined with displacement data. The proposed stiffness damage measurement method is demonstrated by updating the stiffness of four types of bars separately. The results shows that damage in structures can be accurately identified by Bayesian updating methods. This method not only pinpoints the location of the damage, but also quantifies the extent of the damage.

show abstract

Section: Introductionmentioning

confidence: 99%

Study on the truss damage measurement based on Bayesian updating

Zhang

2024

ACE

View full text Add to dashboard Cite

show abstract

“…In addition, there have been efforts to combine fuzzy theory with neural network technology to establish fuzzy neural network assessment models [16][17][18][19][20][21][22][23][24][25]. Additionally, methods that integrate genetic algorithms with neural networks have been developed to create bridge damage fuzzy assessment expert systems [26][27][28][29][30]. Furthermore, bridge safety assessment systems have been developed utilizing the powerful database management capability of the GIS [31][32][33][34][35].…”

Section: Introductionmentioning

confidence: 99%

Safety Assessment Method of Concrete-Filled Steel Tubular Arch Bridge by Fuzzy Analytic Hierarchy Process

Shen,

Chen,

et al. 2023

Buildings

View full text Add to dashboard Cite

The concrete-filled steel tubular (CFST) arch bridge has achieved significant development in recent years due to its unique mechanical performance and technical advantages. However, due to the lagging theoretical research compared to engineering practice, many problems have been exposed in the existing bridges, resulting in adverse social impacts and enormous economic losses. With the increasing prominence of safety issues in CFST arch bridges, it is necessary to assess their safety condition in service. This paper establishes a safety assessment index system for CFST arch bridges using the fuzzy analytic hierarchy process (AHP) based on an exponential scale. The assessment method proposed includes the following main points: (1) Bridge safety assessment is closely related to the load-bearing capacity of components. This study proposes an assessment index that comprehensively considers both the defect conditions and the design load-bearing capacity of components for the safety assessment. (2) The exponential scale method is introduced to safety assessment for the first time, and the AHP based on an exponential scale is applied to calculate the component weights. (3) Considering the specific structural characteristics of CFST arch bridges, this study provides a detailed division of component types and calculates the component weights. By combining the component assessment indexes, a comprehensive safety assessment index system is established, and a safety assessment method for CFST arch bridges is proposed. (4) Taking the Jiantiao Bridge in Zhejiang Province as an engineering case, the load-bearing capacity of components is calculated using finite element software ANSYS 19.1. Based on the established safety assessment index system, the safety of the bridge is assessed by integrating the inspection results. (5) Software for the safety assessment of a CFST arch bridge is developed using Visual Basic, and the assessment results align well with the actual condition of the bridge.

show abstract