ZHUOJIN-LOGOZhuojin Pipe Fitting Co., Ltd
Zhuojin Pipe Fitting Co., Ltd
STEEL PIPE · FITTING · WELDING  |  NOTES FROM A 30-YEAR INDUSTRY VETERAN

Nonlinear Effects in Erection of a Very Long Span CFST Arch Rib

Why This Paper Deserves a Second Read

I have spent thirty years shipping steel pipe to bridge yards, and I can tell you that the erection phase of a concrete-filled steel tube (CFST) arch is where reputations get made or broken. The paper by Hao, Gu and Zhang on the Bosideng Bridge is not just an academic exercise; it addresses a question that has cost real money on real projects: how much does nonlinearity actually matter when you are lifting a huge steel arch rib into place?

The authors set out to compare two boundary conditions at the arch foot — hinged and fixed — and to quantify how geometric nonlinearity affects the rib alignment during hoisting. They derived closed-form formulas for rib alignment under different structural systems, then verified their findings against the actual construction of the Bosideng Bridge. The conclusion is clear and, to my mind, correct: nonlinearity matters far more when the arch foot is hinged than when it is fixed.

The Core Technical Findings

The paper's central insight is that the sensitivity of the rib alignment to geometric nonlinearity depends strongly on the structural system at the arch foot. When the rib is fixed at the base, the nonlinear effects are relatively small. When it is hinged, the effects become significant. The reason, as the authors explain, is that geometric nonlinearity changes the angle at which the stay cables act on the rib, and a hinged system is far more sensitive to that change than a fixed one.

Let me put this in terms of what actually happens on site. When you are erecting a large steel arch rib, you are typically using temporary stay cables (often called "cable-stayed" systems) to control the rib's position as you assemble it segment by segment. The force in each stay cable has a magnitude and a direction. If the rib deflects under its own weight or under cable tension, the angle of the cable changes, which changes the effective force component acting on the rib. In a fixed-base system, the rib is stiff enough that these deflections are small, so the nonlinear correction is minor. In a hinged system, the rib can rotate more freely, and the same cable force produces a larger displacement, which in turn changes the cable angle more, creating a feedback loop that must be accounted for.

The authors' contribution is to quantify this effect and to show that it cannot be ignored for very long spans. For the Bosideng Bridge, they demonstrated that a linear analysis would have predicted a different final alignment than the one actually achieved, and that the nonlinear correction was essential for matching the measured results.

What This Means for the Pipe Supplier and the Erection Team

From my side of the table, this paper raises a practical question: what does the erection team need from the pipe manufacturer to make this work? The answer is threefold.

First, the pipe sections must be delivered with accurate geometric tolerances. If the rib is being erected with a hinged base, the nonlinear effects will amplify any initial imperfection in the pipe geometry. A pipe that is out of round by more than the specified tolerance will cause the rib to deviate from its theoretical alignment, and the nonlinear analysis will not converge to the expected result. I have seen this happen. The erector blames the pipe, the pipe mill blames the erector, and the project manager has to make a decision that nobody is happy with.

Second, the stay cable forces must be applied with precision. The paper shows that the cable angle is a critical parameter. If the cable is anchored at a slightly different position than assumed in the analysis, the nonlinear effect will be different from the prediction. This is not a theoretical concern; it is a daily reality on site. The erector must have a clear procedure for setting cable tensions and for measuring the resulting rib positions, and the analysis must be updated as the actual geometry is revealed.

Third, the welding sequence and the temporary bracing system must be designed with the nonlinear behavior in mind. When you weld a segment of the rib to the already-erected portion, the heat input will cause local distortion. In a hinged system, this distortion can cause the rib to move in a way that the linear analysis would not predict. The paper does not address welding distortion directly, but the principle is the same: anything that changes the geometry of the rib will change the nonlinear response.

A Practical Table for the Erection Engineer

Parameter Fixed-Base Arch Foot Hinged-Base Arch Foot
Sensitivity to geometric nonlinearity Low High
Effect of cable angle change on rib alignment Minor Significant
Required accuracy of cable force application Moderate High
Sensitivity to initial pipe geometric imperfections Low High
Recommended analysis approach Linear with safety factor Full nonlinear analysis
Typical span range where linear is acceptable Under 200 m Under 100 m
Typical span range where nonlinear is required Over 200 m Over 100 m

My Reflection on the Broader Lesson

What I take from this paper is not just a technical formula but a management lesson. The erection of a large steel arch is a system, not a collection of parts. The pipe, the cable, the welding, the foundation, and the analysis are all interconnected. If you change one element — say, you decide to use a hinged base instead of a fixed base — you change the entire behavior of the system, and every other element must be re-evaluated.

I have seen projects where the design team changed the boundary condition at the arch foot late in the process, without re-running the full nonlinear analysis. The result was a rib that did not meet the specified alignment, and the correction required additional temporary bracing and re-welding that cost weeks of schedule. The paper by Hao, Gu and Zhang would have prevented that mistake if it had been applied early enough.

The Bosideng Bridge is a good example of a project where the nonlinear analysis was done properly, and the result was a successful erection. The authors' verification against actual construction data gives me confidence that their approach is sound. I would recommend that any engineer responsible for erecting a CFST arch of more than 200 m span read this paper carefully, and that they insist on a full nonlinear analysis for any hinged-base system, regardless of span.

Final Word

The practical takeaway is simple: do not trust a linear analysis for a hinged-base CFST arch rib of significant span. The nonlinear effects are real, they are significant, and they will show up in the final alignment. Plan for them, design for them, and verify them on site. The pipe you ship will be exactly as good as the analysis that supports its erection.