Predicting Soil Parameters by Modelling Dynamic Compaction Induced Subsidence

Abstract

It is common practice in Dynamic Compaction to carry out a calibration programme before production and execution of actual ground improvement works to optimize the design parameters. In the calibration the ground is initially tested. Then Dynamic Compaction is carried out on a predefined grid size. One or two heave and penetration tests may be carried out during the calibration. Upon completion of Dynamic Compaction the ground will be tested again to ensure that the desired parameters were achieved. Occasionally, a number of patterns may be tried in the calibration to provide the engineer with more design options and sometimes poor test results force the repetition of the calibration with alternative patterns.Obviously testing requires time, and it can be understood that it could be beneficial to be able to predict the improvement in the ground and to take possible remedial measures if the assessment is able to demonstrate that the tests will not meet the design criteria. This may be achieved in Dynamic Compaction by developing a relation between the induced ground subsidence and the improvement of Pressuremeter Test (PMT) limit pressure. In this approach Dynamic Compaction induced ground subsidence is assumed to be the accumulation of vertical strains down to the depth of improvement according to a Rayleigh distribution. The Dynamic Compaction induced strain for each level is correlated to the increase of limit pressure with the assumption that limit pressure will double for every 3% of strain [1].