Response of lightly overconsolidated clay under irregular cyclic loading and comparison with predictions from the strain accumulation procedure

Offshore wind turbine foundations are designed to withstand environmental loads from the wind and waves, both of which are cyclic in nature. The current design methods consider site-specific cyclic load histories and typically require these to be translated from a time-load history with irregular characteristics to an idealised history of parcels of cycles with uniform amplitude and constant average load. The Rainflow counting method is typically used for this translation. The idealised history is then applied in a design method, for example the strain accumulation method. These methods assume that the idealised load history will give approximately the same effect on the soil as the irregular time history. This paper investigates this assumption by a laboratory test programme where the soil response under realistic irregular loading is compared with the response under idealised loading. The laboratory programme consists of cyclic direct simple shear tests on lightly overconsolidated North Sea clay. For most problems, the test results suggest that the assumption is reasonable and represents a convenient simplification for practical design. However, for load histories with large average load, prediction of permanent strain based on idealised histories may underestimate the strain observed in tests with irregular time histories.