Abstract:Big multi-step ship motion forecasting plays an essential role as ship safety warning in maritime operations. However, big multi-step prediction will weaken the correlation of time series and introduce large error accumulation. Real-time prediction also requires a simple prediction model structure, data input-output relationship in line with the actual application, and a high prediction accuracy. To realize the above requirements, a hybrid big multi-step forecasting model is proposed based on real-time wavelet packet decomposition (RTWPD), outlier robust extreme learning machine (ORELM), boosting algorithm, and least squares support vector machine (LSSVM)-based error correction method. Taking 10-step, 20-step, and 30-step as the research objects of big multi-step prediction, two sets of ship roll and pitch motion data obtained by practical experiment are provided to complete four comparison experiments, including the comparison of different decomposition algorithms, the comparison of different classical forecasting models, the comparison of different error correction methods and the comparison of the proposed model with different benchmark models. Taking the 10-step, 20-step, and 30-step prediction results of the pitch of dataset #2 as an example, the mean absolute errors (MAEs) of the proposed model are 0.1619°, 0.1742°, and 0.1808°, respectively; and the minimum Diebold-Mariano (DM) value is 4.88, which is greater than the upper limit of the 1 % confidence level, indicating that the prediction results of the proposed model differ significantly from those of the other models. And the root mean square error (RMSE) of the proposed model is about 0.20° for different data sets and different ship motions. The experimental results indicate that the proposed RTWPD-ORELM-AdaBoost.MRT-LSSVM (RWOALS) model is stable and feasible in practical applications, and it can provide reliable guidance for maritime operations.

Multi-step-ahead significant wave height prediction using a hybrid model based on an innovative two-layer decomposition framework and LSTM

Multiple-step accurate prediction of wave energy: A hybrid model based on quadratic decomposition, SSA and LSTM

A hybrid VMD-LSTM/GRU model to predict non-stationary and irregular waves on the east coast of China

Improving wave height prediction accuracy with deep learning

Study on prediction of ocean effective wave height based on hybrid artificial intelligence model

Deterministic wave prediction model for irregular long-crested waves with Recurrent Neural Network

A significant wave height prediction method based on deep learning combining the correlation between wind and wind waves

Prediction of significant wave height using a VMD-LSTM-rolling model in the South Sea of China

A hybrid CEEMDAN-VMD-TimesNet model for significant wave height prediction in the South Sea of China

Wind Speed Multi-Step Forecasting Using High-Resolution and Multi-Objective Ensemble Model with Error Correction

Significant wave height prediction based on the local-EMD-WaveNet model

Wave Power Prediction Based on Seasonal and Trend Decomposition Using Locally Weighted Scatterplot Smoothing and Dual-Channel Seq2Seq Model

Short-term prediction of the significant wave height and average wave period based on the variational mode decomposition–temporal convolutional network–long short-term memory (VMD–TCN–LSTM) algorithm

Wave field predictions using a multi-layer perceptron and decision tree model based on physical principles: A case study at the Pearl River Estuary

ConvLSTM-Based Wave Forecasts in the South and East China Seas

Big Multi-Step Ship Motion Forecasting Using a Novel Hybrid Model Based on Real-Time Decomposition, Boosting Algorithm and Error Correction Framework

Multi-step wind speed forecasting based on a hybrid decomposition technique and an improved back-propagation neural network

Multistep-ahead Forecasting of Chlorophyll a Using a Wavelet Nonlinear Autoregressive Network

An integrated methodology for significant wave height forecasting based on multi‐strategy random weighted grey wolf optimizer with swarm intelligence

Multi-Step-Ahead Forecasting of Wave Conditions Based on a Physics-Based Machine Learning (PBML) Model for Marine Operations

ConvGRU-RMWP: A Regional Multi-Step Model for Wave Height Prediction