A novel framework for rapid short-term prediction of ship motions using an enhanced gradient boosting decision tree with dynamical evolution sequence

Abstract Accurate short-term prediction of ship motions is essential for selecting operational areas and time windows, which are critical for maritime safety and operational efficiency. Traditional physics-based methods provide high accuracy but are computationally expensive, limiting their use in onboard applications. To address this challenge, this study proposes a rapid short-term prediction method for ship motions. The method utilizes a dynamical evolution sequence (DES) to generate a more uniform sample set, replacing the conventional Latin hypercube sampling (LHS) in the gradient boosting decision tree (GBDT), thus enhancing model accuracy through improved input sampling. First, four test functions are used to validate and compare the original and improved GBDT models. Results show that the DES-based GBDT model consistently outperforms the traditional LHS-based GBDT model across all functions. Based on this, the enhanced model is used to develop a comprehensive set of surrogate models for all hydrodynamic coefficients required for six-degree-of-freedom (6-DOF) motion prediction, including added mass, radiation damping, and wave excitation force, comprising 84 submodels in total. This innovation reduces computation time from hours to approximately 10 s per operating condition. The feasibility of the proposed method is validated under two loading conditions for an offshore support vessel, with prediction results compared to those obtained from the commercial software Hydrostar. The validation demonstrates that the 6-DOF motion amplitudes predicted by the proposed method align closely with those from the commercial software, achieving both rapid and accurate predictions. This method provides a viable alternative for rapid short-term ship motion prediction in engineering practice.

Authors

Publication Details

Journal
Intelligent Marine Technology and Systems
Published
2026-11-16
DOI
https://doi.org/10.1007/s44295-026-00119-1
Primary Topic
Ship Hydrodynamics and Maneuverability
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A novel framework for rapid short-term prediction of ship motions using an enhanced gradient boosting decision tree with dynamical evolution sequence

Kang Zou, Hao Li, Manhai Gui, Li Zhu et al.
Intelligent Marine Technology and Systems
Ship Hydrodynamics and Maneuverability
article

A novel framework for rapid short-term prediction of ship motions using an enhanced gradient boosting decision tree with dynamical evolution sequence

Kang Zou, Hao Li, Manhai Gui, Li Zhu, Feng Wu
article en

Abstract

Abstract Accurate short-term prediction of ship motions is essential for selecting operational areas and time windows, which are critical for maritime safety and operational efficiency. Traditional physics-based methods provide high accuracy but are computationally expensive, limiting their use in onboard applications. To address this challenge, this study proposes a rapid short-term prediction method for ship motions. The method utilizes a dynamical evolution sequence (DES) to generate a more uniform sample set, replacing the conventional Latin hypercube sampling (LHS) in the gradient boosting decision tree (GBDT), thus enhancing model accuracy through improved input sampling. First, four test functions are used to validate and compare the original and improved GBDT models. Results show that the DES-based GBDT model consistently outperforms the traditional LHS-based GBDT model across all functions. Based on this, the enhanced model is used to develop a comprehensive set of surrogate models for all hydrodynamic coefficients required for six-degree-of-freedom (6-DOF) motion prediction, including added mass, radiation damping, and wave excitation force, comprising 84 submodels in total. This innovation reduces computation time from hours to approximately 10 s per operating condition. The feasibility of the proposed method is validated under two loading conditions for an offshore support vessel, with prediction results compared to those obtained from the commercial software Hydrostar. The validation demonstrates that the 6-DOF motion amplitudes predicted by the proposed method align closely with those from the commercial software, achieving both rapid and accurate predictions. This method provides a viable alternative for rapid short-term ship motion prediction in engineering practice.

Intelligent Marine Technology and SystemsVol. 4(1)
Ship Hydrodynamics and Maneuverability
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon
A novel framework for rapid short-term prediction of ship motions using an enhanced gradient boosting decision tree with dynamical evolution sequence — Kang Zou, Hao Li, et al. · Intelligent Marine Technology and Systems (2026) | TGRS Research Map | TGRS