Numerical Analysis of Impact Resistance of CFRP-Reinforced Damaged Hollow Sandwich Steel Tubular Concrete T-Joints

Overview of the Research Topic

This study investigates the impact behavior of hollow sandwich steel tubular concrete (HS-SC) T-joints that have been pre-damaged and subsequently repaired using carbon fiber reinforced polymer (CFRP) wraps. The numerical analysis employs finite element (FE) modeling to simulate both the damage phase and the impact loading phase, providing insights into the residual load-carrying capacity of critical connection nodes in offshore and marine structures. The research addresses a significant gap in structural engineering: the assessment of repaired connections under dynamic loading conditions, which is essential for the integrity of jacket platforms, subsea pipelines, and bridge piers exposed to collision or blast events.

Core Technical Points and Modeling Methodology

The hollow sandwich configuration refers to a double-tube steel section filled with concrete, where the inner and outer tubes are connected by shear keys or dowels. The T-joint geometry involves a branch member intersecting the chord member at 90 degrees, creating complex stress concentrations at the intersection welds and the hollow interface.

The FE model typically employs shell elements (such as S4R or S4RS) for the steel tubes and solid elements (such as C3D8R) for the concrete core. The material models are critical to capturing nonlinear behavior:

Component	Element Type	Material Model	Key Parameters
Outer steel tube	Shell (S4R)	Johnson-Cook or Von Mises with isotropic hardening	Yield strength 345–460 MPa, Young's modulus 206 GPa
Inner steel tube	Shell (S4R)	Same as outer	Same grade or slightly lower
Concrete core	Solid (C3D8R)	Concrete Damaged Plasticity (CDP)	f'c 30–60 MPa, tensile strength 3–5 MPa
CFRP wrap	Shell (S4R)	Orthotropic linear elastic with failure criteria	E1 230 GPa, E2 10–20 GPa, ultimate strain 1.5–2.5%
Weld zones	Shell or solid	Isotropic hardening with reduced yield	0.8–0.9 × base metal yield

The damage phase simulation typically involves a static push or pull load applied to the branch member until local buckling or weld fracture occurs. The resulting plastic deformation, strain distribution, and any crack initiation sites are then transferred to the second phase model. CFRP wraps are then applied to the damaged regions, and an impact load (simulated via mass-spring systems or prescribed velocity boundary conditions) is applied to evaluate the post-repair performance.

Key Findings and Engineering Implications

The numerical results generally demonstrate that CFRP wrapping can recover 60–85% of the original impact resistance of the undamaged joint, depending on the wrap configuration (full wrap, partial wrap, spiral wrap) and the number of layers applied. The hollow sandwich configuration itself provides inherent energy absorption capacity through the concrete core and the interaction between the inner and outer tubes.

Critical observations from such studies include:

The CFRP wrap effectively constrains local buckling at the intersection welds, preventing premature fracture propagation.
The hollow interface shear keys play a crucial role in distributing impact loads between the inner and outer tubes; damage to these keys significantly reduces the effectiveness of CFRP reinforcement.
Impact velocity has a pronounced effect: at higher velocities (>10 m/s), the CFRP contribution diminishes due to the strain-rate sensitivity of steel and the potential for debonding between the CFRP and the steel surface.
The location and extent of pre-damage are more influential than the severity of damage in determining post-repair performance; damage near the weld toe is particularly detrimental.

Practical Recommendations for Engineers

For practical application in offshore and marine structures, the following guidelines emerge from this line of research:

CFRP repair should be applied within 72 hours of damage assessment to prevent corrosion initiation at exposed steel surfaces.
Surface preparation (grinding to bare metal, degreasing) is critical for achieving adequate bond strength; bond strength should be verified by pull-off tests (minimum 20 MPa for steel substrates).
The wrap configuration should prioritize coverage of the weld toe regions and the inner tube intersection zones, where stress concentrations are highest.
For high-impact environments (ship collision zones, blast-prone areas), CFRP alone may be insufficient; hybrid reinforcement with steel plates and CFRP should be considered.
Post-repair inspection using ultrasonic testing (UT) and visual examination should confirm bond integrity before returning the structure to service.

Reflections on Methodological Limitations

The numerical approach, while powerful, has inherent limitations that engineers must acknowledge. The Johnson-Cook material model, commonly used for dynamic analysis, requires strain-rate parameters that are often extrapolated from limited experimental data. The concrete damaged plasticity model does not fully capture the behavior of concrete under impact conditions, particularly the transition from quasi-static to dynamic failure modes. Furthermore, the bond behavior between CFRP and steel is typically modeled using cohesive zone elements with assumed material properties, which may not reflect the actual performance under cyclic or impact loading.

Future research should focus on validating numerical models with full-scale impact tests, developing more sophisticated bond models that account for temperature and moisture effects, and exploring hybrid reinforcement strategies that combine CFRP with traditional steel reinforcement for optimal post-damage performance.

This study reinforces the importance of considering the full lifecycle of structural connections, including damage assessment, repair strategies, and post-repair performance evaluation, rather than focusing solely on the initial design and construction phases.