Hybrid and symbolic regression-based modeling of pore water pressure generation in sands under cyclic loading

The increase in pore water pressure induced by cyclic shear stresses in saturated sandy soils is one of the fundamental mechanisms that triggers liquefaction by reducing the effective stress within the soil. Accurate modeling of this process is critical for reliably predicting soil behavior in geotechnical engineering applications. In this doctoral study, empirical pore water pressure models available in the literature were reviewed, and two new models were proposed based on data obtained from stress-controlled cyclic triaxial tests. The developed models were constructed using experimental datasets compiled from the literature and were tested by comparison with laboratory experiments previously conducted within the scope of a master's thesis. The predictive performance of the models was evaluated using statistical methods and benchmarked against existing models from the literature. The results demonstrate that the proposed models offer a reliable and applicable framework for predicting pore pressure generation under various soil conditions.