摘要：Core Ideas Specimen preparation methods have a significant influence on hydraulic conductivity. The difference caused by different methods can be large as one order of magnitude. Soil pore structure should be considered in predicting hydraulic conductivity. A pore‐information‐based model is presented to predict hydraulic conductivity. The new model is more accurate than traditional particle information based models. A series of laboratory tests were performed to investigate the influences of specimen preparation on pore size distribution of soil and saturated hydraulic conductivity ( K s ). Nuclear magnetic resonance technology was used to measure the pore size distribution of the saturated samples of silty soil, which were prepared by three different kinds of methods: Proctor compaction, static compaction, and the consolidation method. The K s of the samples was measured by the falling head permeability test. The results show that the difference in K s caused by different specimen preparations can be large as one order of magnitude, as the measured K s varied from 3.09 × 10 −3 to 3.36 × 10 −4 cm s −1 . The consolidated specimen tended to have the greatest K s value, followed by those prepared by Proctor compaction and static compaction. The observed difference highlights the importance of pore structure in determining K s . This study also presents a pore‐information‐based theoretical approach for predicting K s . A comparison of measured data shows that the proposed model performs better than the traditional void‐ratio‐based models.
关键词：FID; free induction decay; K–C; Kozeny–Carman; NMR; nuclear magnetic resonance
