Litcius/Paper detail

Physics-informed neural network with pretraining optimization for ocean acoustic field prediction

Juncong Tang, Haiqiang Niu

2025The Journal of the Acoustical Society of America6 citationsDOIOpen Access PDF

Abstract

To address the challenge of capturing high-frequency features in ocean acoustic field prediction using Physics-Informed Neural Networks (PINNs), this study introduces an enhanced pretraining optimization approach, termed PreT-OceanPINN, based on a recently proposed OceanPINN framework, a PINN-based method for predicting ocean acoustic pressure fields. By implementing a two-stage strategy-pretraining in a hypothetical environment followed by fine-tuning with real-world data-PreT-OceanPINN significantly improves both prediction accuracy for high-frequency components and training efficiency. In the pretraining stage, the model is trained using envelope signals derived from a simulated environment, allowing it to internalize the underlying physical principles of sound propagation. During the fine-tuning stage, a limited amount of measured data are used to adapt the model to complex real-world conditions. Compared to the standard OceanPINN approach, PreT-OceanPINN delivers more accurate high-frequency predictions without increasing real data dependency, thus demonstrating clear performance advantages. The effectiveness of the proposed method is validated through numerical simulations and experimental data from the SWellEx-96 field experiment.

Topics & Concepts

Artificial neural networkComputer scienceField (mathematics)Experimental dataEnvelope (radar)Artificial intelligenceMachine learningSound pressureDeep neural networksPredictive modellingPattern recognition (psychology)Training setTraining (meteorology)Data miningPerformance predictionData modelingEngineeringUnderwater Acoustics ResearchModel Reduction and Neural NetworksNeural Networks and Reservoir Computing