Comparative Study on Modal Identification of Steel Tube Concrete Frame Structure Models

Literature Overview

This paper by Du Guofeng, Xu Chengxiang, and Fan Hong (2007), published in the Journal of Lanzhou University of Technology, presents a comparative study on modal identification methods for steel tube concrete (STC) frame structure models. The research employs three distinct approaches: computational modal analysis using ANSYS 8.1 finite element software, single-input single-output (SISO) experimental modal analysis, and environmental excitation (ambient vibration) modal analysis. The study is supported by the Hubei Provincial Natural Science Foundation and provides valuable insights into the applicability and accuracy of different modal identification techniques for composite structural systems.

Methodology and Experimental Setup

The research investigates a physical model of an STC frame structure, which represents a simplified version of a multi-story building or industrial frame where the primary structural members are composite steel tube concrete columns and beams. The model is instrumented with accelerometers at strategic locations to capture the dynamic response under both controlled and ambient excitation conditions.

Comparative Results and Analysis

The study finds that all three methods can effectively identify the fundamental frequencies of the STC frame structure model, demonstrating that modal analysis is feasible for composite steel tube concrete structures. However, the accuracy and completeness of the identified modes differ among the methods.

The computational modal analysis provides the most complete set of natural frequencies and mode shapes, including higher-order modes that may be difficult to excite experimentally. However, the accuracy of the computational results depends critically on the fidelity of the finite element model, particularly the representation of the steel-concrete interface behavior, boundary conditions, and material property assumptions. Any discrepancy between the assumed and actual boundary conditions or material properties will introduce errors into the predicted frequencies.

The SISO experimental method, typically involving impact hammer testing or shaker excitation, provides high-quality frequency response functions with excellent signal-to-noise ratios. This method is particularly effective for identifying the lower-order modes of the structure, which are most relevant for seismic design and serviceability assessment. However, SISO testing requires controlled excitation, which may not be feasible for full-scale structures or for structures that cannot be isolated from their foundation.

The environmental excitation method, which relies on natural ambient vibrations such as wind, traffic, and microseismic activity, offers the advantage of non-intrusive measurement that does not require shutting down or isolating the structure. The challenge with this method is the low signal-to-noise ratio and the difficulty of separating structural modes from environmental noise. Modern system identification techniques, such as the stochastic subspace identification (SSI) method and the extended least squares (ELS) method, have significantly improved the accuracy of environmental excitation modal analysis, making it a viable option for in-service structural assessment.

Engineering Practice Implications

The comparative study has several important implications for the structural assessment and monitoring of STC frame structures in engineering practice.

First, for the design phase of new STC structures, computational modal analysis should be performed to establish the expected dynamic characteristics, including natural frequencies, mode shapes, and damping ratios. These computed values serve as the baseline against which measured values from post-construction testing can be compared. Any significant deviation between computed and measured frequencies may indicate modeling errors, construction defects, or unexpected boundary conditions that require investigation.

Second, for the construction quality verification phase, SISO experimental modal analysis is recommended as the primary testing method. The high signal-to-noise ratio and clear mode identification make it suitable for verifying that the as-built structure meets the design requirements. The measured frequencies should fall within a specified tolerance band (typically +/- 10% of the computed values) to confirm that the structural stiffness and mass properties are as designed.

Third, for the long-term structural health monitoring of in-service STC structures, environmental excitation modal analysis is the most practical approach. Periodic measurements of the structure's natural frequencies under ambient conditions can detect changes in structural stiffness caused by damage, material degradation, or foundation settlement. A significant reduction in natural frequency (typically exceeding 5% of the baseline value) should trigger a detailed inspection to identify the cause of the stiffness reduction.

Study Insights and Outlook

This research contributes to the growing body of knowledge on the dynamic characterization of composite steel tube concrete structures. The finding that all three modal identification methods can effectively identify the fundamental frequencies of STC frame structures validates the use of modal analysis as a tool for structural assessment in this composite system. The comparative analysis also highlights the complementary nature of the three methods, suggesting that a multi-method approach is the most robust strategy for comprehensive structural dynamic characterization.

From a materials science perspective, the dynamic response of STC structures is influenced by the composite action between the steel tube and concrete core, which introduces complex interfacial behavior that is difficult to model accurately in finite element analyses. The steel-concrete interface can exhibit slip, debonding, and partial composite action depending on the loading conditions and the quality of the interface preparation. These factors can cause discrepancies between computed and measured frequencies, particularly for higher-order modes where the interfacial behavior has a more pronounced effect.

Future research should focus on several areas. First, the development of more sophisticated finite element models that accurately capture the steel-concrete interface behavior, including bond-slip relationships and partial composite action, is essential for improving the accuracy of computational modal analysis. Second, the application of advanced signal processing techniques, such as wavelet transform and empirical mode decomposition, to environmental excitation data can improve the identification of closely spaced modes and transient dynamic events. Third, the integration of modal analysis with structural health monitoring systems that incorporate distributed fiber optic sensors and wireless sensor networks can enable real-time dynamic monitoring of STC structures, providing continuous assessment of structural integrity throughout the service life. The ultimate goal is to establish a comprehensive modal-based structural health monitoring framework for STC structures that combines computational, experimental, and in-situ measurement approaches for optimal performance.