Cone penetration tests (CPTs) have been performed on very soft marsh and foundation soils in support of the dike design of two containment areas for a liquefied natural gas (LNG) terminal in Georgia.
Due to the inherent variability in layer thickness and soil classification within a CPT sounding profile, an algorithm has been written to automate the CPT sounding analysis. The algorithm was initially written to automate the data interpretation process (i.e. soil classification and strength evaluation) and was later expanded to (i) evaluate strength gain and (ii) estimate settlement using CPT profiles. The algorithm uses CPT profiles (obtained at the same location during different time frames) that are analyzed through signal processing techniques incorporated in the code. The signal processing techniques used include high frequency filters and a series of cross-correlation techniques. The algorithm also incorporates a series of curve fitting techniques to evaluate the profiles as a function of depth. This paper provides an overview of the techniques used in the algorithm along with strength gain plots.Comparison between settlements estimated using the CPT soundings through signal processing and settlements calculated using conventional theories are also presented. The comparison shows favorable agreement between the two techniques.
Read More: http://ascelibrary.org/doi/abs/10.1061/41165(397)236