Sea state estimation using monitoring data by convolutional neural network (CNN)

Sea state identification using machine learning—A comparative study based on in-service data from a container vessel

“…The displayed results show sufficient accuracy for the prediction of the sea state parameters under forward speed conditions. Kawai et al [21] present a simulation-based study of a container carrier in the frequency domain by extracting sequences of spectral values from a set of cross response spectra. The convolutional neural network predicts the parameters for an Ochi-Hubble type spectrum, i.e.…”

“…It is noted that 25 min are chosen as the maximum sample length since recordings close to the 30 min threshold were frequently missing. In addition, the focus of the herein presented work is on using smaller time frames, noting that other studies use longer durations, say, from 30 min in [23] to 60 min in [21]. The synchronization step decreased the size of the dataset further to 5182 samples.…”

“…Thus, the number of discrete frequencies 𝛺 is part of a second sensitivity study, in which also the effect of the sample length on prediction accuracy will be investigated. For reference, 71 and 100 discrete spectral ordinates were extracted from individual cross response spectra in the work of [21,23], respectively. Lastly, the shape of the input matrix for the machine learning model in the spectral domain has a shape of 𝛺 × 𝑀, where 𝑀 takes a maximum value of 43.…”

“…For comparison of ResNet and Inception to state of the art literature, two models are adopted as benchmarks: The convolutional neural network (CNN) from Kawai et al [21] and the multichannel convolutional LSTM (MLSTM) network as proposed by Düz et al [20]. The former network was applied in the frequency domain on cross spectral sequences, whereas the latter one used raw time series as input for the prediction of sea state parameters.…”

Health-Monitoring Systems for Marine Structures: A Review

“…Different research studies reflect the importance of this aspect in structural health monitoring. Kawai et al [ 15 ] estimated the encountered sea state using machine learning from measurement data of ocean-going 14,000 TEU container ships. Karvelis et al [ 16 ] reported a novel data-driven method for the localization of acoustic emissions in complex structures of ship hulls, combining insights from the fields of signal processing.…”