Language
Country/Area
No result found
A brief discussion on the birth of SPH: How did two scientists promote the revolution of fluid mechanics in 1977?
In 1977, the study of fluid mechanics ushered in a major change, when two scientists, Gingold and Monaghan and Lucy, proposed a new computational method, smoothed particle hydrodynamics (SPH). The original purpose of this method was to solve astrophysical problems. However, over time, SPH has been gradually applied to many fields, including ballistics, volcanology, and oceanography. It plays an important role in the simulation of fluid flow. contribute.
SPH is unique in that it is a meshless Lagrangian method, which means that the coordinate system changes as the fluid moves.
The greatest advantage of SPH lies in its ability to adapt to complex boundary dynamics. Traditional grid-based methods are not able to handle free surface flows or large boundary displacements, while SPH can easily simulate fluids and greatly simplify the implementation and parallelization of models in multi-core architecture calculations. . In addition, since the SPH method can flexibly adjust the resolution according to demand and can flexibly set parameters for variables such as density, it is often the first choice in the aforementioned applications.
Advantages and limitations of SPH
Although SPH shows superior performance in fluid dynamics, it still has some limitations. SPH is relatively complex, both in setting boundary conditions and handling fluid inlets and outlets. This is particularly true in simulations of interfacial flows, where particles near the boundary will behave in time.Some experts pointed out that "the processing of boundary conditions is undoubtedly one of the most difficult technical links in the SPH method."
In addition, when certain metrics (e.g., kinetic energy spectrum) are not directly related to density, the computational cost of SPH will be much higher than that of grid-based simulation methods. In such cases, it is more efficient to use traditional meshing methods, especially when simulating constant density flows.
Expansion of application areas
Fluid Dynamics
In recent years, SPH has been increasingly used in simulating fluid motion. Since the SPH method naturally has the property of mass conservation, it can ensure the stability of the fluid during the simulation. Moreover, it does not require solving a system of linear equations to calculate pressure, which allows for greater flexibility and efficiency in the calculation process.
SPH can directly create the free surface of the two-phase fluid interaction, which requires additional tracking of the fluid boundary in traditional meshing technology.
Nevertheless, the SPH rendering process still requires the use of polygonization techniques to generate renderable free-surface geometry. The latest research, such as the one proposed by PCISPH, aims to improve the de-oping constraints and solve the interaction problems between fluids and rigid bodies, further improving the realism and accuracy of fluid simulation.
Astrophysics
In the field of astrophysics, SPH is widely used due to its adaptive resolution and numerical conservatism of physical quantities. From simulating galaxy formation and star birth to supernova explosions and asteroid collisions, SPH has demonstrated its powerful computing power.
Solid Mechanics
As SPH has been further extended to solid mechanics, its applications range from metal forming to crack growth. The mesh-independent nature of this method effectively avoids certain problems in traditional mesh methods, especially stability issues when dealing with local deformations such as cracks.
Conclusion
In summary, SPH has gone through decades of development since its birth in 1977 and has gradually evolved into an important tool in computational fluid dynamics. Its successful practice in various application fields demonstrates its unparalleled flexibility and effectiveness. Of course, every technology has its limitations, and SPH is no exception. In the future, with the improvement of computing power and further improvement of algorithms, will SPH become the mainstream method for fluid simulation?
