Prediction of Safe Sea-State for Pile Driving
In this paper vertical piles have been studied with a view to identifying the range of sea-states suitable for the safe pile driving operation. Pile configuration, including the non-linear foundation and the gap between the pile and the pile sleeve shims have been modeled using the finite elements (FE) analysis facilities within ABAQUS. Dynamic analyses of the system for various sea-states characterized by significant wave heights and mean zero upcrossing periods and modeled as a combination of several wave components, have been performed. Repeating the above procedure can generate a table of safe and unsafe sea-states. If the prediction is repeated N times from which n times proved to be safe, then it can be said that the predicted sea-state is safe with a probability of 100(n/N)%. The significant wave height (Hs) and mean zero upcrossing period (Tz) of a future sea-state of a location in NE Pacific (near 46° N 131° W) were generated using the artificial neural networks (ANNs) already trained for this purpose — the location of US National Oceanographic Data Center (NODC) Buoy 46005 is used in this study. The Hs and Tz of some future sea-states were generated from their corresponding conditional 7-parameter probability density functions (pdf’s) given some information including a number of previously measured Hs’s and Tz’s. The parameters of the pdf’s have been estimated from the outputs of 2 different 7-network sets of trained ANNs. This gives a predicted sea-state for a specific time in future. The methodology explained in this paper can identify all control parameters and offer possible solution strategy. The finding reveals that how slight changes in the design configuration can be beneficially exploited to limit the pile response.