SimBlue Marine Logo

Digital Twin Solutions

Physics-based AI/ML solutions for high-fidelity simulations, underwater acoustics, and custom digital twin platforms.

Overview

SimBlue Marine's Digital Twin technology leverages cutting-edge physics-based machine learning and artificial intelligence to create high-fidelity virtual replicas of marine systems. Our advanced AI/ML solutions provide intelligent answers for underwater noise prediction, high-fidelity fluid dynamics simulations, acoustic propagation modeling, and custom digital twin platforms. We specialize in developing tailored solutions, custom platforms, and integrated pipelines that combine deep learning with fundamental physics to deliver unprecedented accuracy in marine engineering applications.

Core Capabilities

Advanced graph neural networks and physics-informed machine learning for high-fidelity fluid dynamics, acoustic propagation, and multi-physics simulations with unprecedented accuracy.
AI-driven underwater acoustic modeling, transmission loss prediction, and noise propagation analysis using conditional convolutional neural networks and graph-based approaches.
Spatio-temporal prediction of cavitating flows, multiphase fluid dynamics, and complex marine hydrodynamics using advanced deep learning architectures and physics constraints.
Tailored digital twin solutions, custom platforms, and integrated pipelines designed for specific marine applications with real-time data processing and predictive analytics.
AI-powered optimization framework for fluid-acoustic shape optimization, balancing performance, noise reduction, and efficiency using graph neural network surrogate models.
  • Computational speed-up of three orders of magnitude compared to traditional methods.
  • Median relative error of 1-2% in pressure and velocity predictions.
  • Multi-objective optimization for noise reduction and aerodynamic performance.
  • Demonstrated 13.9% reduction in sound pressure level with 7.2% increase in lift coefficient.
Reference: Computer Methods in Applied Mechanics and Engineering, 2025

Our AI/ML Digital Twin Solutions Include

Featured Galleries

Research Profile

View Google Scholar Profile

Latest Publications on Physics-Based AI/ML & Digital Twins

"A Graph Neural Network Surrogate Model for Multi-Objective Fluid-Acoustic Shape Optimization"
F. Hadizadeh, W. Mallik, R. K. Jaiman, Computer Methods in Applied Mechanics and Engineering 441, 117921, 2025
View Paper →
"Towards spatio-temporal prediction of cavitating fluid flow with graph neural networks"
R. Gao, S. Heydari, R. K. Jaiman, International Journal of Multiphase Flow 177, 104858, 2024
View Paper →
"Predicting wave propagation for varying bathymetry using conditional convolutional autoencoder network"
I. K. Deo, A. Venkateshwaran, R. Jaiman, ASME OMAE Conference 87844, V006T08A038, 2024
View Paper →