Study On Correlation Between Shear-wave Velocity And Liquefaction Resistance Of Saturated Sand

In saturated sand liquefaction evaluation, the correlation between shear-wave velocity and liquefaction resistance of saturated sand is an important question that has not been thoroughly understood until now. In this dissertation, the following work will be carried out.First, the principle of shear-wave velocity measurement by torsion-vibration is studied. According to this theory, two sets of technique realizing torsion-vibration by piezoelectric-ceramic are proposed. Then two new kinds of shear-wave velocity transducer are developed which can be used in cyclic triaxial cell for shear-wave velocity measurement of saturated sand samples. By jointing the shear-wave velocity transducer and the cyclic triaxial test apparatus, combined test equipment is developed on which shear-wave velocity measurement and cyclic liquefaction test can be carried out simultaneously.Second, by using large amount of undisturbed and remolded samples of 12 different kinds of typical liquefiable granular soil, the correlation between the shear-wave velocity and cyclic-liquefaction strength of saturated is studied. The soil obtained can be classified as fine sand, silty fine sand, sandy silt, and silt. During the procedure of testing, an effective method of controlling the initial shear-wave velocity of soil samples in cyclic triaxial cell is proposed. From the results we found that shear-wave velocity correlates well with liquefaction resistance of saturated sand within a certain failure strain criterion. This conclusion provides a valuable way to estimate the in-situ liquefaction resistance of granular soil stratum. Finally, a relational expression between shear-wave velocity and liquefaction resistance of saturated is derived based on laboratory test results, in which such factors as in-situ stress state, fine content, and seismic level are considered which may influence liquefaction resistance. The credibility of the relational expression is checked using 174 liquefaction and non-liquefaction case histories, which proves the expression has a correct rate of 71.2%. At the same time, an Error-Back-Propagation artificial neural network model for saturated sand liquefaction evaluation is established, which has four input parameters such as shear-wave velocity, consolidation stress, fine ontent and number of cycles. By comparison, The BP network has a higher correct rate than the former expression.