Operator Guidance Informed by AI-Augmented Simulations

TL;DR

This paper introduces a multi-fidelity, data-adaptive method using LSTM neural networks to accurately estimate ship response statistics in complex bimodal, bidirectional sea conditions, combining low- and high-fidelity simulation data.

Contribution

It presents a novel approach integrating simple low-fidelity simulations with high-fidelity data through LSTM to improve ship response predictions in bimodal seas.

Findings

LSTM effectively predicts ship responses using combined data sources

The approach reduces reliance on high-fidelity simulations

Validation shows good agreement with high-fidelity results

Abstract

This paper will present a multi-fidelity, data-adaptive approach with a Long Short-Term Memory (LSTM) neural network to estimate ship response statistics in bimodal, bidirectional seas. The study will employ a fast low-fidelity, volume-based tool SimpleCode and a higher-fidelity tool known as the Large Amplitude Motion Program (LAMP). SimpleCode and LAMP data were generated by common bi-modal, bi-directional sea conditions in the North Atlantic as training data. After training an LSTM network with LAMP ship motion response data, a sample route was traversed and randomly sampled historical weather was input into SimpleCode and the LSTM network, and compared against the higher fidelity results.