English

Optimization of the Woodcock Particle Tracking Method Using Neural Network

Computational Physics 2025-02-20 v1

Abstract

The acceptance rate in Woodcock tracking algorithm is generalized to an arbitrary position-dependent variable q(x)q(x). A neural network is used to optimize q(x)q(x), and the FOM value is used as the loss function. This idea comes from physics informed neural network(PINN), where a neural network is used to represent the solution of differential equations. Here the neural network q(x)q(x) should solve the functional equations that optimize FOM. For a 1d transmission problem with Gaussian absorption cross section, we observe a significant improvement of the FOM value compared to the constant qq case and the original Woodcock method. Generalizations of the neural network Woodcock(NNW) method to 3d voxel models are waiting to be explored.

Keywords

Cite

@article{arxiv.2502.13620,
  title  = {Optimization of the Woodcock Particle Tracking Method Using Neural Network},
  author = {Bingnan Zhang},
  journal= {arXiv preprint arXiv:2502.13620},
  year   = {2025}
}
R2 v1 2026-06-28T21:49:54.193Z