Zhuojin Pipe Fitting Co., LtdEfficient Linear Elastic Iterative Analysis of Square CFST Frame Ultimate Capacity Under Second-Order Effects
Literature Overview
The paper by Yang Lufeng, Zhao Yufeng, and Xie Weiwei, published in the China Civil Engineering Journal (Vol. 55, Issue 2, 2022), presents an efficient linear elastic iterative method for analyzing the ultimate bearing capacity of square concrete-filled steel tube (CFST) frames considering second-order effects. The authors develop a homogeneous generalized yield function for square CFST members under combined compression and bending with second-order effects, and based on this function, they establish an Elastic Modulus Reduction Method (EMRM) for ultimate capacity analysis through linear elastic iterative procedures. This work addresses a significant computational challenge in the analysis of CFST structures: the efficient and accurate determination of ultimate capacity while accounting for geometric nonlinearity (P-Δ and P-δ effects).
Core Technical Approach
The method is built upon three key components:
- A quadratic polynomial influence function relating the confining coefficient and the stability coefficient for square CFST members.
- A homogeneous generalized yield function established through regression analysis using a first-order polynomial with fractional exponential power.
- A strategic reduction of the elastic modulus of high-capacity units to simulate structural stiffness degradation, followed by linear elastic iterative analysis to determine the ultimate bearing capacity.
The homogeneous generalized yield function is a critical innovation. It provides a unified mathematical description of the yield surface for square CFST members under combined loading, incorporating the effects of both material confinement and geometric nonlinearity. The fractional exponential power in the polynomial formulation allows for a more accurate representation of the curved yield surface, which is essential for capturing the complex interaction between axial force and bending moment.
| Method Component | Mathematical Form | Purpose |
|---|---|---|
| Influence function | Quadratic polynomial | Relate confining and stability coefficients |
| Generalized yield function | First-order polynomial with fractional exponential power | Describe yield surface |
| Elastic modulus reduction | Strategic reduction factor | Simulate stiffness degradation |
| Analysis procedure | Linear elastic iteration | Determine ultimate capacity |
Interpretation of Technical Points
The traditional elastic modulus adjustment method has difficulty accounting for second-order effects because it treats the structure as purely elastic throughout the analysis. The proposed EMRM overcomes this limitation by strategically reducing the elastic modulus of units that are approaching their yield capacity. As the elastic modulus of a unit is reduced, the stiffness of that unit decreases, causing a redistribution of internal forces to other units. This redistribution continues iteratively until the structure reaches its ultimate capacity, at which point the elastic modulus of the critical unit is reduced to zero, indicating that the structure can no longer sustain additional load.
The concept of the confining coefficient is central to the yield function. For square CFST members, the confinement effect of the steel tube on the concrete core is characterized by the confining coefficient, which depends on the steel tube thickness, width, and the concrete strength. The stability coefficient accounts for the second-order effects, which depend on the axial load, member length, and lateral stiffness. The influence function relates these two coefficients to the ultimate capacity of the member.
The fractional exponential power in the yield function is a mathematical device that allows the polynomial to capture the nonlinear interaction between axial force and bending moment more accurately than a simple linear or quadratic polynomial. The fractional power effectively "stretches" or "compresses" the yield surface in a controlled manner, allowing for a better fit to the actual yield behavior of square CFST members.
Comparison with Conventional Methods
The proposed method is validated against experimental test data and conventional incremental nonlinear finite element (FE) methods. The results demonstrate that the method has strong applicability, good computational accuracy, and high computational efficiency. Compared to incremental nonlinear FE analysis, the linear elastic iterative method is significantly faster because it avoids the need for material nonlinearity modeling, geometric nonlinearity modeling, and the associated convergence difficulties.
| Method | Computational Efficiency | Accuracy | Complexity |
|---|---|---|---|
| Proposed EMRM | High | Good | Low |
| Incremental nonlinear FE | Low | High | High |
| Traditional elastic modulus method | High | Poor (ignores 2nd order) | Low |
| Direct nonlinear analysis | Very low | Very high | Very high |
The computational efficiency advantage is particularly significant for large-scale CFST frame structures, where the number of elements and nodes can be very large. For practical engineering applications, the ability to perform ultimate capacity analysis using linear elastic iterative procedures makes the method accessible to engineers who may not have access to advanced nonlinear FE software or the expertise to set up and run nonlinear analyses.
Second-Order Effect Analysis
The study systematically analyzes the influence of second-order effects on the ultimate bearing capacity of square CFST frames under different loading conditions and geometric parameters. The results show that second-order effects have a non-negligible influence on both long CFST columns and CFST frames. For long columns, the P-δ effect (member-level second-order effect) is dominant, while for frames, the P-Δ effect (story-level second-order effect) is more significant.
The influence of second-order effects on the ultimate capacity is quantified by comparing the results with and without second-order effects. For slender CFST columns, the reduction in ultimate capacity due to second-order effects can be substantial, potentially exceeding 20% for high slenderness ratios. For CFST frames, the reduction depends on the overall frame geometry, loading pattern, and the relative stiffness of columns and beams.
Engineering Practice Implications
From a practical standpoint, the proposed method offers engineers a powerful tool for the preliminary design and capacity assessment of CFST frames. The method can be implemented in existing linear elastic structural analysis software by modifying the elastic modulus of elements based on the yield function and the iterative procedure. This avoids the need for specialized nonlinear analysis software and reduces the computational burden significantly.
The method is particularly valuable for the assessment of existing CFST structures, where the ultimate capacity needs to be determined for retrofitting or strengthening decisions. The ability to perform rapid ultimate capacity analyses for multiple loading scenarios and geometric configurations enables engineers to identify critical members and develop targeted strengthening strategies.
For fabrication and welding quality control, the method provides a framework for evaluating the structural impact of fabrication tolerances and welding defects. If a CFST member has a reduced effective cross-section due to wall thickness variation or weld defects, the method can be used to assess the impact on the member's ultimate capacity and, by extension, on the overall frame capacity.
Key Questions and Reflections
The method assumes that the yield function can be accurately represented by the proposed mathematical form. While the regression analysis provides a good fit to the available data, the extrapolation of the yield function to parameter ranges beyond those used in the regression may introduce errors. Engineers should be cautious when applying the method to CFST members with unusual geometric proportions or material combinations that fall outside the range of the calibration data.
Another consideration is the treatment of the elastic modulus reduction. The method reduces the elastic modulus of high-capacity units, but the specific reduction schedule is based on engineering judgment and calibration against FE results. Different reduction schedules may yield slightly different ultimate capacity predictions, and the sensitivity of the results to the reduction schedule should be evaluated for critical applications.
Study Insights and Implications
The principal contribution of this research is the development of a practical, efficient, and reasonably accurate method for ultimate capacity analysis of square CFST frames that accounts for second-order effects. The method bridges the gap between the simplicity of linear elastic analysis and the accuracy of nonlinear analysis, providing engineers with a tool that is both computationally efficient and structurally meaningful. The homogeneous generalized yield function, with its fractional exponential power formulation, represents a significant advancement in the mathematical description of CFST member behavior under combined loading. For the steel pipe and welding industry, this method provides a valuable analytical tool for evaluating the structural performance of CFST structures and for optimizing the design of steel tube members to achieve the desired balance between capacity and economy.