Damage assessment of a tunnel-type structure to protect submarine power cables during anchor collisions

• We calculated drag coefficients and terminal velocities of five anchors.
• We considered four parameters for anchor collision analyses.
• We obtained dynamic characteristics of tunnel-type structure during anchor collision.
• We provided a procedure for damage assessment and recommendation.
• We verified numerical analysis results by a field test.

The dynamic characteristics of a tunnel structure used to protect underwater power cables, the so-called A-duct, were determined for anchor collisions to provide a procedure for damage assessment and recommendations. The required physical quantities of five target anchors, including the drag coefficient, were obtained using an element-based finite-volume method and ANSYS-CFX software. The terminal velocities of the anchors were then calculated to maximize the colliding kinetic energy. For collision analysis, four parameters (anchor type, ground condition, collision velocity, and collision point) were considered, and the A-duct was modeled based on the Riedel–Hiermaier–Thoma concrete model using ANSYS-Autodyn software. Our analysis results indicated severe damage (D = 1) for most of the gauge points; the damaged area and level increased with the anchor weight. The results showed that the damage was concentrated in the collision area for stock anchors; however, for stockless anchors, damage was also evident in adjacent areas (i.e., damage propagation) due to the anchor head shape as well as the transfer mechanism provided by its reinforcing nets. Accordingly, the 2-ton stock anchor caused more damage at the gauge points near the collision location than the 2-ton stockless anchor. Second, regardless of the ground conditions and rotation angle of the anchor heads with respect to the vertical axis, the damage levels were almost identical. Fixed boundary conditions and non-rotational angle were sufficient for the model used. Third, the damaged areas became smaller when the anchor collision locations deviated from the reference gauge point (P1), i.e., the center of the A-duct. Finally, a comparison of the field-test results to equivalent numerical collision simulations indicated that the size of the predicted and experimentally observed damaged areas were in agreement within 7%.