摘要：To ensure structural safety of large container ships, advanced application of monitoring data is pursued in the structural health monitoring project. In this study, we estimated statistical sea state that the ship actually encountered using Bayesian updating of the sea state data estimated from the measured hull stresses. To achieve reliable real-time short-term statistical estimation of the sea state, we introduced forgetting rate to the past data in the Bayesian model. In addition, abrupt change of the sea state due to the ship operation was detected by Bayesian analysis of the data to obtain probable time history of the sea state by Bayesian update. As a result, good tracking performance of the real time sea state was demonstrated.
