Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 10 of 10 for “"Physics-Constrained Machine Learning"”.

  1. Physics-Constrained Machine Learning Models: Analysis and Generalisation to Varied Hemodynamic Domains

    … models for accurate predictions. The use of machine learning (ML) to create efficient surrogate models is a promising solution. In this thesis, a physics-constrained ML method was defined to address these challenges. Specifically, a physics-informed neural network (PINN) methodology over a …

    auckland-ms Repository record for Physics-Constrained Machine Learning Models: Analysis and Generalisation to Varied Hemodynamic Domains (opens in a new tab)

  2. Investigation of Process-Structure Relationship for Additive Manufacturing with Multiphysics Simulation and Physics-Constrained Machine Learning

    … measurement. In this research, a mesoscale multiphysics simulation model, called phase-field and thermal lattice Boltzmann method (PF-TLBM), is developed with simultaneous considerations of heterogeneous nucleation, solute transport, heat transfer, and phase transition. The simulation can reveal …

    gatech Repository record for Investigation of Process-Structure Relationship for Additive Manufacturing with Multiphysics Simulation and Physics-Constrained Machine Learning (opens in a new tab)

  3. Subsurface imaging, characterization, and monitoring with time-lapse seismic using physics-guided data-driven methods

    This dissertation develops physics-guided, data-driven methods for subsurface imaging, characterization, and monitoring from seismic data. The central problem is that subsurface structure, elastic properties, and dynamic reservoir evolution cannot be observed directly; instead, they must be …

    texas Repository record for Subsurface imaging, characterization, and monitoring with time-lapse seismic using physics-guided data-driven methods (opens in a new tab)

  4. Advancing Subsurface Flow Modeling for Improved Oil Recovery and CO2 Sequestration Using Multi-Scale Numerical Simulations and Physics-Informed Neural Networks

    … approaches—multi-scale numerical simulations and Physics-Informed Neural Networks (PINNs)—to solve complex sub-surface flow problems with improved accuracy and efficiency. The research spans three major areas: (1) multi-scale simulations for surfactant-based chemical EOR, (2) application of PINNs …

    tamu Repository record for Advancing Subsurface Flow Modeling for Improved Oil Recovery and CO2 Sequestration Using Multi-Scale Numerical Simulations and Physics-Informed Neural Networks (opens in a new tab)

  5. Advancing Subsurface Flow Modeling for Improved Oil Recovery and CO2 Sequestration Using Multi-Scale Numerical Simulations and Physics-Informed Neural Networks

    … approaches—multi-scale numerical simulations and Physics-Informed Neural Networks (PINNs)—to solve complex sub-surface flow problems with improved accuracy and efficiency. The research spans three major areas: (1) multi-scale simulations for surfactant-based chemical EOR, (2) application of PINNs …

    tamu Repository record for Advancing Subsurface Flow Modeling for Improved Oil Recovery and CO2 Sequestration Using Multi-Scale Numerical Simulations and Physics-Informed Neural Networks (opens in a new tab)

  6. Advancing Subsurface Flow Modeling for Improved Oil Recovery and CO2 Sequestration Using Multi-Scale Numerical Simulations and Physics-Informed Neural Networks

    … approaches—multi-scale numerical simulations and Physics-Informed Neural Networks (PINNs)—to solve complex sub-surface flow problems with improved accuracy and efficiency. The research spans three major areas: (1) multi-scale simulations for surfactant-based chemical EOR, (2) application of PINNs …

    tamu Repository record for Advancing Subsurface Flow Modeling for Improved Oil Recovery and CO2 Sequestration Using Multi-Scale Numerical Simulations and Physics-Informed Neural Networks (opens in a new tab)

  7. Advancing Subsurface Flow Modeling for Improved Oil Recovery and CO2 Sequestration Using Multi-Scale Numerical Simulations and Physics-Informed Neural Networks

    … approaches—multi-scale numerical simulations and Physics-Informed Neural Networks (PINNs)—to solve complex sub-surface flow problems with improved accuracy and efficiency. The research spans three major areas: (1) multi-scale simulations for surfactant-based chemical EOR, (2) application of PINNs …

    tamu Repository record for Advancing Subsurface Flow Modeling for Improved Oil Recovery and CO2 Sequestration Using Multi-Scale Numerical Simulations and Physics-Informed Neural Networks (opens in a new tab)

  8. Advancing Subsurface Flow Modeling for Improved Oil Recovery and CO2 Sequestration Using Multi-Scale Numerical Simulations and Physics-Informed Neural Networks

    … approaches—multi-scale numerical simulations and Physics-Informed Neural Networks (PINNs)—to solve complex sub-surface flow problems with improved accuracy and efficiency. The research spans three major areas: (1) multi-scale simulations for surfactant-based chemical EOR, (2) application of PINNs …

    tamu Repository record for Advancing Subsurface Flow Modeling for Improved Oil Recovery and CO2 Sequestration Using Multi-Scale Numerical Simulations and Physics-Informed Neural Networks (opens in a new tab)

  9. Advancing Subsurface Flow Modeling for Improved Oil Recovery and CO2 Sequestration Using Multi-Scale Numerical Simulations and Physics-Informed Neural Networks

    … approaches—multi-scale numerical simulations and Physics-Informed Neural Networks (PINNs)—to solve complex sub-surface flow problems with improved accuracy and efficiency. The research spans three major areas: (1) multi-scale simulations for surfactant-based chemical EOR, (2) application of PINNs …

    tamu Repository record for Advancing Subsurface Flow Modeling for Improved Oil Recovery and CO2 Sequestration Using Multi-Scale Numerical Simulations and Physics-Informed Neural Networks (opens in a new tab)

  10. Physics-Augmented Modeling and Optimization of Complex Systems: Healthcare Applications

    … of this dissertation is to develop innovative physics-augmented methodologies</em></p> <p><em>for modeling, monitoring, and optimizing complex systems. Specifically, the research</em></p> <p><em>focuses on the development of physics-augmented machine learning models that facilitate</em></p> …

    utk-theses Repository record for Physics-Augmented Modeling and Optimization of Complex Systems: Healthcare Applications (opens in a new tab)