Abstract:Purpose: Linear Boltzmann transport equation (LBTE) is as accurate as the Monte Carlo method (MC) for dose calculation in photon/particle therapy (LBTE is a deterministic and Eulerian formulation and MC is a statistical and Lagrangian description). An advantage of LBTE is that numerous acceleration techniques can be utilized for acceleration. This work is to explore the acceleration of LBTE via diffusion synthetic acceleration (DSA). Methods: For simplicity, two-dimensional, steady-state, and within-group LBTE is considered with two angular dimensions and two spatial dimensions. The discrete ordinate method is developed for solving this integro-differential equation. The angular variables are discretized using a level-symmetric quadrature set on the unit sphere. The spatial variables are discretized on the structured grid based on the diamond scheme. The source-iteration method (SI) is used to solve the discretized system.Since SI is slow in optically thick and highly scattering regime. DSA is developed to accelerate SI. The motivation for DSA is that diffusion equation (DE) is a good approximation of LBTE in the above regime. However, DE is much cheaper than LBTE computationally since DE only involves spatial variables. Thus, in each DSA iteration, DSA adds to the SI step a computationally-negligible DE step, i.e., to first solve DE with the SI residual as source term, and then compensate the SI solution with DE solution. Results: DSA was benchmarked and compared with SI. The difference between two methods was within 0.12% which verifies the accuracy of DSA, while DSA demonstrated the great advantage in speed, e.g., the reduction of iteration number to 6% and 4% respectively for cases with 100 and 1,000 scattering-absorption ratio that commonly occur in clinical dose calculation. Conclusion: DSA has been developed as one of many possible means for accelerating the numerical solver of LBTE for dose calculation. The authors were partially supported by the NSFC (#11405105), the 973 Program (#2015CB856000) and the Shanghai Pujiang Talent Program (#14PJ1404500).

MARCOS, a numerical tool for the simulation of multiple time-dependent non-linear diffusive shock acceleration

Numerical Modeling of Time Dependent Diffusive Shock Acceleration

Data-driven Nonlinear Constitutive Relations for Rarefied Flow Computations

Computation of diffusive shock acceleration using stochastic differential equations

Numerical Studies of Cosmic Ray Injection and Acceleration

One-Dimensional, One-Phase and Two-Phase Eulerian Explicit Shock Tube Simulation Code

Assessment of a high-order shock-capturing central-difference scheme for hypersonic turbulent flow simulations

Diffusive Acceleration of Particles at Oblique, Relativistic, Magnetohydrodynamic Shocks

Efficient numerical methods for Anisotropic Diffusion of Galactic Cosmic Rays

High-order discontinuous Galerkin schemes with subcell finite volume limiter and adaptive mesh refinement for a monolithic first-order BSSNOK formulation of the Einstein-Euler equations

SU-E-T-270: Diffusion Synthetic Acceleration for Linear Boltzmann Transport Equation

Online Optimisation of Machine Learning Collision Models to Accelerate Direct Molecular Simulation of Rarefied Gas Flows

Cosmic ray acceleration at perpendicular shocks in supernova remnants

Building a Hydrodynamics Code with Kinetic Theory

Numerical Simulation of Hypersonic Transitional Flows by Means of a Kinetic Node-Pair Approach

Local time stepping methods and discontinuous Galerkin methods applied to diffusion advection reaction equations

Numerical methods for interface coupling of compressible and almost incompressible media

A MUSTA-FORCE algorithm for solving partial differential equations of relativistic hydrodynamics

Parallelization of Kinetic Theory Simulations

A novel transient-adaptive subcell algorithm with a hybrid application of different collision techniques in direct simulation Monte Carlo (DSMC)