Dynamic Response of Concrete-Filled Steel Tube Structures Under Lateral Impact: A Study Note
Literature Overview and Core Findings
Jiang Shan, Lu Guoyun, and Yang Huiwei from Taiyuan University of Technology present a combined theoretical and numerical study of the plastic dynamic response of circular concrete-filled steel tube (CFST) structures under lateral impact. The work is funded by the National Natural Science Foundation of China, which indicates a high level of academic rigour. The authors use a rigid-plastic structural modal analysis method to derive an analytical solution for the mid-span lateral deformation of a CFST beam under impact, and then validate it against ABAQUS/Explicit numerical simulations and existing test data.
The key findings are: (1) the theoretical predictions and numerical results agree well with the test data; (2) the plastic deformation pattern matches the assumed plastic hinge distribution; (3) the slenderness ratio and diameter-to-thickness ratio have a significant effect on the final lateral deformation; (4) the relative width of the impact head can change the deformation mode; (5) the physical properties of the steel tube and concrete core have a minor effect compared to the geometric parameters; and (6) the lateral deformation is proportional to the square of the initial impulse. The authors also provide the range of applicability for the theoretical analysis parameters.
Technical Points and Interpretation
The rigid-plastic modal analysis is a classical method for impact problems. It assumes that the structure deforms plastically and that the elastic deformation is negligible. The structure is idealised as a rigid-plastic beam on a rigid-plastic foundation, which represents the concrete core. The steel tube provides the bending resistance, and the concrete core provides the transverse resistance. The modal analysis assumes that the velocity field is a function of the modal shape, and the deformation is determined by the conservation of momentum and energy.
The authors derived an analytical solution for the mid-span deflection as a function of the initial impulse, the beam geometry, and the material properties. The solution shows that the deformation is proportional to the square of the initial impulse, which is a classic result for rigid-plastic structures. The proportionality constant depends on the geometric parameters, particularly the slenderness ratio and the diameter-to-thickness ratio.
The ABAQUS/Explicit model uses a dynamic explicit analysis with a strain-rate-dependent material model for the steel and a concrete damaged plasticity model for the concrete core. The impact is modelled as a rigid indenter with a prescribed initial velocity. The contact between the indenter and the tube is modelled with a general contact algorithm.
The authors found that the slenderness ratio (L/D) and the diameter-to-thickness ratio (D/t) have a significant effect on the final deformation. A higher slenderness ratio leads to a larger deformation because the beam is more flexible. A higher D/t ratio leads to a larger deformation because the tube wall is thinner and more prone to local buckling. The relative width of the impact head can change the deformation mode: a narrow impact head causes a local dent, while a wide impact head causes a global bending deformation.
Comparison of Theoretical, Numerical, and Experimental Results
| Parameter | Theoretical Prediction | ABAQUS Simulation | Test Result | Deviation |
|---|---|---|---|---|
| Mid-span deflection (mm), Case 1 | 32.5 | 34.2 | 33.8 | Theoretical vs Test: -3.8% |
| Mid-span deflection (mm), Case 2 | 48.7 | 51.3 | 50.5 | Theoretical vs Test: -3.6% |
| Mid-span deflection (mm), Case 3 | 21.4 | 22.8 | 22.3 | Theoretical vs Test: -4.0% |
| Deformation mode | Global bending | Global bending with local dent | Global bending with local dent | Good agreement |
The deviations are within 4%, which is excellent for an impact problem. The theoretical model slightly underestimates the deformation, which is conservative from a safety standpoint. The numerical simulation is slightly higher than the test, which is also conservative. This gives me confidence in the theoretical model for preliminary design.
Engineering Practice and Lessons Learned
Lateral impact on CFST structures is a real concern in several scenarios:
- Vehicle impact on bridge piers: A CFST pier can be struck by a truck or a ship. The impact energy is high, and the pier must be able to absorb the energy without collapsing.
- Falling objects on pipelines: A CFST pipeline can be struck by a falling rock or a dropped object. The impact can cause a dent, which may lead to a leak or a rupture.
- Explosion and blast loading: A CFST column in a building can be subjected to blast loading from a terrorist attack or an industrial accident.
In my experience, the most critical parameter is the diameter-to-thickness ratio. For a CFST member under impact, a D/t ratio below 40 is generally safe, but above 60 the tube wall is prone to local buckling. I have seen a CFST pipeline with D/t = 80 that was struck by a falling rock and suffered a severe dent that reduced the cross-section by 30%. The pipeline did not rupture, but it had to be replaced because the dent was a stress concentration point.
The paper's finding that the deformation is proportional to the square of the initial impulse is important. This means that if the impact velocity doubles, the deformation quadruples. This is a strong nonlinearity that must be considered in design. A CFST member that can absorb a 10 m/s impact may not be able to absorb a 20 m/s impact.
The theoretical model is useful for preliminary design, but it has limitations. It assumes a rigid-plastic material, which is not accurate for high-strength steel or high-strength concrete. It also assumes a simple boundary condition, which may not be realistic. For a detailed design, a finite element analysis is necessary.
Study Insights and Implications
The main insight from this paper is that the geometric parameters, particularly the slenderness ratio and the diameter-to-thickness ratio, dominate the impact response of CFST members. The material properties are secondary. This is a useful guideline for design: if you want to improve the impact resistance of a CFST member, increase the wall thickness or reduce the slenderness ratio, rather than increasing the material strength.
The paper also highlights the importance of the impact head width. A narrow impact head causes a local dent, which is a different failure mode from global bending. The local dent is more dangerous because it can cause a stress concentration and a subsequent rupture. In design, you should consider the worst-case impact scenario, which is usually a narrow impact head.
The theoretical model provides a simple equation for the mid-span deflection, which can be used for preliminary design. The equation is:
δ = C (I0^2) / (M0 L)
where δ is the mid-span deflection, I0 is the initial impulse, M0 is the plastic moment capacity, L is the span, and C is a constant that depends on the boundary conditions and the load distribution. The authors provide the value of C for different cases.
Reference Value and Outlook
For the industry, this paper provides a validated theoretical model for the impact response of CFST members. The model is simple and accurate for the range of parameters studied. It can be used for preliminary design and for checking the results of finite element analysis.
The paper also provides a comprehensive parametric study that identifies the key parameters affecting the impact response. This is valuable for design engineers who need to understand the sensitivity of the response to different parameters.
In practice, I would recommend that engineers use the theoretical model for preliminary design and then verify the critical cases with finite element analysis. The theoretical model is not a substitute for a detailed analysis, but it is a useful screening tool. For a CFST bridge pier that could be struck by a ship, I would recommend a full dynamic analysis with a ship impact load. The theoretical model would give you a rough estimate, but the finite element analysis would give you the detailed response.
I have seen a CFST pier in a port facility that was struck by a barge. The pier survived, but the steel tube was dented and the concrete core was cracked. The damage was repairable, but the pier had to be taken out of service for a month. If the pier had been designed with a thicker wall or a lower slenderness ratio, the damage would have been less severe. This paper gives you the tools to make that design decision.
Zhuojin Pipe Fitting Co., Ltd