Adaptive Elastic Modulus Reduction Method for Ultimate Bearing Capacity of CFST Arch Bridges

Literature Overview

This paper by Yang Lufeng, Zheng Jian, Zhang Wei, Ou Wei, and Liu Huijuan (2017), published in the China Journal of Highway and Transport, proposes an adaptive elastic modulus reduction method for analyzing the ultimate bearing capacity of concrete-filled steel tube (CFST) arch bridges. The research is funded by the National Natural Science Foundation of China (Projects 51478125 and 51469004) and the Guangxi Natural Science Foundation Major Project (2012GXNSFEA053002). The authors address the limitations of conventional incremental nonlinear finite element methods, which are computationally complex and inefficient, by developing a novel adaptive approach based on elastic modulus reduction and homogeneous generalized yield functions.

Core Technical Content

The conventional incremental nonlinear finite element method for CFST ultimate bearing capacity analysis relies on the elastoplastic constitutive relationships of CFST materials to establish nonlinear iterative calculation formulas. This approach is computationally intensive and often impractical for engineering design and analysis requirements. The authors propose an adaptive elastic modulus reduction method that addresses these limitations through several key innovations.

First, for circular cross-section CFST members with significant material property differences under various loading conditions, the authors establish a generalized yield function based on unified theory and bearing capacity correlation equations. A full factorial experimental design is then used to set configuration points on the generalized yield surface, with denser point placement near extremum points. Regression analysis is employed to establish a homogeneous generalized yield function.

Second, the authors define unit bearing ratio, bearing ratio uniformity, and reference bearing ratio, and propose an adaptive identification criterion for high-bearing-capacity elements in CFST arch bridges. Third, an elastic modulus adjustment strategy is established based on the principle of strain energy conservation, which adaptively reduces the elastic modulus of high-bearing-capacity elements to simulate stiffness damage during the loading process. The ultimate bearing capacity is then calculated using linear elastic iterative analysis.

Methodology and Key Parameters

The method is validated against model test results and incremental nonlinear finite element calculations, demonstrating high computational accuracy and efficiency. The authors also discuss the influence of discrete element quantity, load distribution mode, and the homogeneity of the generalized yield function on the results.

Engineering Practice Implications

For bridge engineering practice, this adaptive method offers a practical alternative to the computationally expensive incremental nonlinear finite element analysis. The method can capture the adaptive adjustment capability of different material fibers in CFST during the loading and deformation process. The adaptive simulation of structural damage evolution in arch bridges achieves both high computational accuracy and efficiency.

The approach is particularly valuable for the preliminary design and optimization stages of CFST arch bridges, where rapid iterative analysis is essential. The homogeneous generalized yield function provides a simplified yet accurate representation of the complex material behavior, enabling efficient calculation of the ultimate bearing capacity without the need for full elastoplastic constitutive modeling.

Study Insights and Reflections

The adaptive elastic modulus reduction method represents a significant methodological advance in CFST structural analysis. By converting a nonlinear problem into a series of linear elastic analyses with adaptive stiffness reduction, the computational burden is substantially reduced while maintaining accuracy. The use of a homogeneous generalized yield function is a clever approach that captures the essential material behavior without the complexity of full constitutive modeling.

However, the method's accuracy depends heavily on the quality of the generalized yield function and the appropriateness of the adaptive identification criterion. The homogeneity assumption of the yield function should be verified for different loading conditions and material combinations. For engineering practice, the method provides a valuable tool for rapid assessment of CFST arch bridge ultimate capacity, particularly during the design optimization phase where multiple load cases and geometric configurations need to be evaluated.