University of Thessaly (UTH)
Smoothed particle hydrodynamics για ροές με ελεύθερη επιφάνεια
Abstract
dc:descriptionThis thesis presents a comprehensive investigation into the capabilities, limitations, and advancement of the Smoothed Particle Hydrodynamics (SPH) method for simulating fluid flows in hydraulic and environmental engineering contexts. As a mesh-free, Lagrangian approach, SPH offers distinct advantages in handling complex geometries, free surfaces, and multiphase flows, making it a compelling alternative to traditional mesh-based Computational Fluid Dynamics (CFD) techniques. However, the method still faces several challenges, including numerical consistency, boundary condition implementation, turbulence modeling, and computational efficiency.The research begins with an extensive theoretical and computational review of SPH, identifying its grand challenges and critically assessing existing formulations and kernel functions. Two widely used open-source SPH solvers, LAMMPS and DualSPHysics, are employed to simulate a broad set of test cases, including laminar and turbulent flows in both 2D and 3D configurations, under closed and open channel conditions. Validation is performed against analytical solutions, numerical benchmarks, and experimental data, with results demonstrate that SPH can produce accurate and physically consistent predictions when properly configured. A significant area of focus is the treatment of boundary conditions, a critical factor affecting SPH stability and accuracy. The study evaluates multiple boundary handling strategies, with particular emphasis on the Modified Dynamic Boundary Conditions (mDBC) approach, which provides superior performance in reducing pressure artifacts and enhancing particle behavior near solid interfaces. The effectiveness of this approach is confirmed through geometrically complex flow simulations, including expansions, contractions, and simulations of flow control systems such as sluice gates and weirs.The key contribution of this thesis is the implementation of a Dynamic Smagorinsky turbulence model into the DualSPHysics source code. This approach eliminates the need for empirical calibration of turbulence parameters, offering an adaptive, locally responsive model that improves the accuracy of SPH in simulating turbulent regimes. Comparative results show notable improvements in velocity profile predictions, particularly in near-wall regions, and demonstrate closer alignment with theoretical log-law behavior.The practical applicability of SPH is further demonstrated through simulations of 3D open channel flow over submerged vegetation, a scenario with direct relevance to ecohydraulics and environmental modeling. The SPH results show strong agreement with experimental and finite volume data, affirming the method’s capability to resolve complex fluid–structure interactions in natural flow environments. The modified SPH model outperformed both the traditional grid-based numerical method and the existing SPH model.Overall, this thesis contributes to the advancement of SPH as a robust, flexible, and scalable CFD tool. Through rigorous validation, model development, and practical application, it sets the foundation for future research and industrial deployment of particle-based flow solvers. It also highlights avenues for further innovation, including adaptive resolution methods, machine learning integration, and hybrid numerical frameworks.
Degree
thesis:*- Grantor dc:publisher
- University of Thessaly (UTH)
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Chatzoglou, Epstathios
- Χατζόγλου, Ευστάθιος
Subjects
dc:subject × 24- SPH
- Ροή με ελεύθερη επιφάνεια
- Μοντελοποίηση Τυρβώδους Ροής
- Υπολογιστική ρευστομηχανική (CFD)
- Dynamic Smagorinsky model
- Αριθμητική προσομοίωση
- Υδραυλική Μηχανική
- Περιβαλλοντικές ροές
- Smoothed Particle Hydrodynamics (SPH)
- Free Surface Flow
- Turbulence Modeling
- Computational fluid dynamics (CFD)
- Boundary Condition Handling
- Numerical simulation
- Hydraulic Engineering
- Environmental flows
- Επιστήμη Πολιτικού Μηχανικού
- Επιστήμες Μηχανικού και Τεχνολογία
- Πολιτική μηχανική
- Επιστήμη Μηχανολόγου Μηχανικού
- Μηχανολογική μηχανική
- Civil Engineering
- Engineering and Technology
- Mechanical Engineering
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/59902