This study presents the principle of nonlinear energy dissipation using an artificial neural network to assess liquefaction potential. The nonlinear relationship between an increase in pore water pressure and the dissipation of seismic energy observed in test data from triaxial shear testing of saturated sand was used to calculate the hysteresis loop energy required to trigger liquefaction. Data recorded during the 1999 Chi-Chi earthquake in Taiwan were utilized to validate the proposed ANN-based hysteresis loop energy model. Results show that the concept of using hysteresis loop energy and the proposed neural network are capable of effectively assessing liquefaction potential.
All Science Journal Classification (ASJC) codes
- Civil and Structural Engineering
- Building and Construction
- Geotechnical Engineering and Engineering Geology