“ALGERIA” VIADUCT

The experimental investigation was devoted to the study of the wind-induced response of long-span steel viaducts during the launching phase, a temporary but particularly delicate construction stage in which the structural configuration differs substantially from the final one.
The tested structure is part of the Oued Tlélat – Tlemcen high-speed railway line, where long viaducts are erected by incremental launching over complex terrain.

During launching, the most aerodynamically and dynamically sensitive element is the launching nose, a lightweight steel truss temporarily attached to the front of the deck to reduce cantilever bending moments. Owing to its reduced mass, limited stiffness and elongated geometry, the launching nose may exhibit a wind-induced response markedly different from that of the completed bridge.

The main objective of the experimental study was therefore to characterize the aerodynamic forces acting on the launching nose and to investigate its aeroelastic behaviour, with particular emphasis on wind-induced vibrations in the vertical direction and on the potential occurrence of vortex-excited response during the launching operations.

Sectional model concept and design rationale

The experimental campaign was based on a sectional model representative of the most aerodynamically critical portion of the launching nose. The selected cross-section corresponds to the configuration characterized by solid lateral surfaces, identified as the most prone to wind-induced excitation.

The model was conceived as a rigid, non-monolithic assembly, allowing accurate control of its mass and stiffness properties. Special care was devoted to reproducing the geometric features governing flow separation and vortex formation, while ensuring that the dynamic parameters relevant to aeroelastic phenomena could be properly scaled.

[/vc_column_text]

Experimental phases and testing strategy

The experimental programme was articulated into two main phases, designed to progressively build up the understanding of the wind–structure interaction.

In the first phase, static aerodynamic tests were performed on the rigidly mounted sectional model. These tests allowed the characterization of the mean aerodynamic forces acting on the launching nose, the identification of the vortex shedding process and the assessment of the sensitivity of the aerodynamic coefficients to wind speed and flow incidence.

In the second phase, aeroelastic tests were carried out by elastically supporting the same sectional model. The suspension system was designed to allow motion only in the vertical direction, thus reproducing the dominant vibration mode expected during the launching phase, while suppressing all other degrees of freedom. The stiffness of the system was assigned through calibrated leaf springs, enabling precise control of the natural frequency and damping of the model.

This approach ensured full consistency between static and dynamic tests, allowing direct interpretation of the aeroelastic response in light of the previously measured aerodynamic characteristics.

Aeroelastic behaviour and interpretation of results

The aeroelastic tests highlighted a wind-induced response governed by leading-edge vortex interaction with the launching nose geometry. The excitation mechanism is associated with vortices impinging on the windward edges of the section, generating a periodic aerodynamic forcing capable of sustaining vertical oscillations.

A distinctive outcome of the investigation is that the response associated with a higher-order vortex-shedding mechanism proved to be more significant than that linked to the fundamental shedding process. This regime was characterized by comparatively larger vibration amplitudes, a broader range of wind conditions over which the oscillations developed, and displacement time histories that were nearly harmonic. Although not commonly encountered, this behaviour is consistent with known responses of certain bluff sections with sharp edges and confirms the sensitivity of the launching nose geometry to leading-edge flow phenomena.

Specific test configurations were introduced to assess the potential onset of galloping instability. Both aeroelastic and static results consistently indicate a stable behaviour with respect to galloping over the range of conditions relevant to the construction stage. No divergent response or progressive amplification of oscillations was observed.

Throughout the experimental programme, a deliberately conservative level of structural damping was adopted, lower than that expected at full scale. This choice provides a safety margin in the interpretation of the results and supports the robustness of the observed stability.

Torsional motion was not explicitly included in the aeroelastic investigation. Based on the overall aerodynamic behaviour of the section and the absence of critical features in the static response, torsional instabilities are expected only at wind conditions well beyond those relevant for the launching phase and are therefore not considered critical for the execution of the operation.

[/vc_column_text]

Engineering relevance

The experimental results show that, for the investigated configuration, the launching nose does not exhibit critical aeroelastic instabilities under wind action relevant to the construction stage. While vortex-induced vibrations may occur, their amplitude remains limited and self-contained, and no conditions leading to unsafe behaviour during launching were identified.

The study highlights the importance of dedicated experimental analyses even for temporary construction configurations.