Random Finite Element Analysis of Initial Defects in Rectangular CFST Columns

Literature Overview

This paper by Yang Lei, Xue Xiaohong, Jiang Jiguo, Huang Zhengye, and Jiang Ziqi, published in Construction Technology (Chinese and English) in 2022 (Vol. 51, No. 22, pp. 103-108), presents a random finite element analysis of the influence of internal random geometric defects in the core concrete on the axial load-bearing capacity of rectangular concrete-filled steel tube (CFST) columns. The research was supported by the Ministry of Housing and Urban-Rural Development Science and Technology Program (Project No. 2021-K-077) on safety control of temporary support structures for super-high-rise buildings based on intelligent construction theory. The study was conducted at China Railway 20th Engineering Group Co., Ltd. Sixth Engineering Company. The research employed ABAQUS/Python secondary development to establish finite element models incorporating internal random defects in the core concrete and performed multiple cyclic calculations to generate a large dataset of load-bearing capacity values for statistical analysis.

Research Background and Defect Characterization

In practical construction of CFST columns, the core concrete inevitably contains various geometric defects resulting from imperfect concrete placement, inadequate compaction, or construction joints. These defects include voids, honeycombing, segregation zones, and incomplete fill areas. The presence of these defects reduces the effective cross-sectional area of the concrete core and creates stress concentrations that can initiate premature failure. Traditional design approaches typically assume a homogeneous and defect-free concrete core, which may lead to unconservative estimates of the load-bearing capacity.

The research focuses specifically on rectangular CFST columns, which are increasingly used in super-high-rise buildings due to their efficient use of space and favorable bending resistance about both principal axes. The random finite element analysis approach adopted in this study represents a significant advancement over deterministic analysis methods, as it accounts for the inherent variability in defect distribution and provides probabilistic insights into the structural performance.

Methodology and Finite Element Modeling

The finite element models were developed using ABAQUS software with Python-based secondary development to automate the generation of models with different random defect configurations. The defect characterization includes the location, size, and shape of internal voids within the concrete core. Multiple models were generated with varying defect distributions, and each model was subjected to axial compression loading to determine the ultimate load-bearing capacity. The large number of simulation runs enabled the construction of statistical distributions of the load-bearing capacity, revealing the probabilistic nature of defect influence on structural performance.

Key Findings on Defect Influence

The analysis results revealed that the presence of internal defects has a substantial influence on the axial load-bearing capacity of rectangular CFST columns. Statistical analysis of the large dataset of load-bearing capacity values showed that the capacity distribution follows a skewed distribution rather than a normal distribution, indicating that the influence of defects is nonlinear and asymmetric. When the defect rate is very small, the structural load-bearing capacity can still be reduced by more than 40%, with a rapid rate of capacity reduction. As the defect rate increases further, the rate of capacity reduction slows down, suggesting a diminishing marginal effect at higher defect levels.

The location of defects within the cross-section was found to have a significant influence on the load-bearing capacity. Defects located at the center of the cross-section had the greatest adverse effect on the load-bearing capacity, while defects located near the ends of the cross-section, close to the middle region, had the least influence. This finding is physically intuitive, as central defects create the largest stress concentration and reduce the effective load-bearing area most significantly, while peripheral defects are partially accommodated by the confining steel tube.

Quality Control Implications and Engineering Practice

From a quality control perspective, the findings of this research underscore the critical importance of ensuring proper concrete placement and compaction in the core of CFST columns. The potential for a 40% or greater reduction in load-bearing capacity even with small defect rates highlights the need for rigorous construction quality control measures. Engineers and inspectors should pay particular attention to the following quality control aspects:

• Concrete workability and slump control to ensure proper flow into the confined space of the steel tube
• Vibration and compaction procedures to minimize void formation
• Use of non-destructive testing methods such as ultrasonic testing to detect internal voids after concrete hardening
• Implementation of acceptance criteria based on defect detection results
• Consideration of probabilistic design approaches that account for construction variability

The research also has implications for the design of temporary support structures in super-high-rise buildings, which was the original motivation for the study. In such applications, the reliability of CFST columns is critical for worker safety, and the potential for significant capacity reduction due to internal defects must be carefully considered in the design and inspection protocols.

Study Insights and Reflections

This paper represents a significant advancement in the understanding of CFST column behavior by incorporating the realistic imperfections that inevitably occur during construction. The random finite element analysis approach provides a powerful tool for quantifying the influence of construction variability on structural performance, which is essential for reliable and safe design. I find the finding that even small defect rates can cause substantial capacity reductions particularly alarming from a safety perspective and highlight the need for stringent construction quality control. The skewed distribution of load-bearing capacity values suggests that traditional deterministic design methods may be unconservative for CFST columns with internal defects. The research methodology, combining ABAQUS/Python automation with statistical analysis, provides a replicable framework for similar studies on other types of construction imperfections and structural configurations. This work contributes to the broader goal of improving the reliability and safety of CFST structures through a better understanding of construction-induced variability.