Artificial neural network based breakwater damage estimation considering tidal level variation

•Wave height at breakwater is predicted by using ANN considering water level.•Wave prediction ANN is combined with the damage estimation procedure.•Water level change affects breakwater damage.•Lifetime damage is obtained with water level being treated as random variable

A new approach to damage estimation of breakwater armor blocks was developed by incorporating a wave height prediction artificial neural network (ANN) into a Monte Carlo simulation (MCS). The ANN was used to predict the wave height in front of a breakwater, with both the deep water wave heights and tidal level being input to the ANN. The waves predicted by the ANN were comparable to those from a wave transform analysis. Using an ANN in wave prediction makes it possible to very simply and quickly obtain numerous waves near the breakwater. Eventually, the analysis time for the expected damage can be reduced. In addition, the effect of the tidal level on the expected damage was revealed by numerical examples. In these numerical examples, it was found that the tidal variation should be taken into account when estimating the expected breakwater damage.