Development of a framework for automatic quantification of uncertainty in seismic cone penetration testing
Please login to view abstract download link
Accurate quantification of the shear wave velocity, Vs, of soils and rocks is an important consideration in several geotechnical design applications. Offshore wind turbine foundation designers, in particular, require reliable profiles of Vs to calculate the shear modulus Gmax (i) for input into soil-structure-interaction models to calculate the first mode of frequency of the foundation-turbine-system and (ii) as one of several critical inputs into 1D Winkler type or 3D finite element analysis approaches. Seismic velocity testing such as seismic cone penetration test (SCPT) utilise the travel time for a shear wave from its source to one or more receivers along an assumed travel path to calculate Vs. Despite the complexities and uncertainties associated with obtaining vertical Vs and therefore Gmax profiles, the results are often reported to designers as a single deterministic dataset without an intuitive measure of uncertainty that can be incorporated in the design process. This paper reviews standard interpretation methods and outlines the key sources of uncertainty in downhole seismic testing. A rigorous workflow to rapidly obtain uncertainty-quantified profiles from seismic velocity testing using a Bayesian inversion approach combined with an error handling framework is developed. The inversion approach delivers probabilistic uncertainty bounds while the error handling framework suppresses hidden errors which can lead to inaccurate, misleading results – such errors are frequently encountered when applying existing methods to real datasets. The algorithm is demonstrated by application to large database of SCPT data. The results demonstrate that the framework is a robust, rapid and useful tool for practical analyses.