Hybrid Framework for Forecasting Circular Excavation Collapse: Combining Physics-Based and Data-Driven Modeling

The use of supporting fluids to stabilize excavations is a common technique adopted in the construction industry. Rapid detection incipient collapse for deep and timely decision making are crucial ensure safety during construction. This paper explores hybrid framework forecasting fluid-supported circular by combining physics-based data-driven modeling. Finite-element limit analysis first used develop numerical database stability numbers both unsupported excavations. parameters considered modeling include excavation geometry, soil strength profile, support fluid properties. A algorithm learn results fast surrogate amenable integration within real-time monitoring systems. By way example, proposed strategy retrospectively applied recent field case history where observational method update input surrogate.