New method tracks non-spherical particles in complex flows
A new mathematical framework uses moments to represent particle size, shape, and velocity efficiently.
Researchers developed a conditional quadrature method of moments (CQMOM) for oblate spheroidal particles. This method tracks particles within a five-dimensional phase space. It includes particle size, aspect ratio, and velocity.
Industrial processes often involve non-spherical particles. In these systems, particle size, shape, and velocity influence transport, breakage, and classification dynamics. These properties evolve within a high-dimensional phase space.
One way to model these flows is the Euler-Lagrange method. This method tracks every individual particle and its properties. While accurate, it is prohibitively expensive for industrial-scale systems.
Another approach is the Euler-Euler method. This method treats the disperse phase as a continuum. It is computationally efficient because costs are largely independent of the number of particles. However, classical Euler-Euler methods often assume particles are monodisperse and monokinetic.
The Population Balance Equation (PBE) describes polydisperse populations using a number density function (NDF). The quadrature method of moments (QMOM) provides an efficient way to solve PBEs. Instead of the NDF, QMOM transports a small set of low-order moments. These moments provide a compact statistical description of the population.
CQMOM represents a multidimensional joint NDF as a hierarchy of conditional distributions. This allows the model to preserve correlations between variables. For example, it can show how velocity depends on particle size.
The researchers tested the CQMOM framework against Monte Carlo simulations. They tested cases involving drag, breakage, recoil, and attrition at different Stokes numbers. With two nodes per coordinate, the particle number, mean size, and mean aspect ratio matched the Monte Carlo solution within statistical uncertainty. Velocity statistics matched within 7%.
The CQMOM method was more than 60 times faster than a single Monte Carlo realization. The study found that the conditional velocity quadrature is the main source of remaining deviations. The authors discuss potential remedies for these deviations.
Key facts
- The CQMOM method is more than 60 times faster than a single Monte Carlo realization.
- The method tracks particles in a five-dimensional phase space including size, aspect ratio, and velocity.
- Velocity statistics in the model agree with Monte Carlo solutions within 7%.