In this paper we consider a new generalized algorithm for the efficient calculation of component object volumes given their equivalent constructive solid geometry (CSG) definition. The new method relies on domain decomposition to recursively subdivide the original component into smaller pieces with volumes that can be computed analytically or stochastically, if needed. Unlike simpler brute-force approaches, the proposed decomposition scheme is guaranteed to be robust and accurate to within a user-defined tolerance. The new algorithm is also fully general and can handle any valid CSG component definition, without the need for additional input from the user. The new technique has been specifically optimized to calculate volumes of component definitions commonly found in models used for Monte Carlo particle transport simulations for criticality safety and reactor analysis applications. However, the algorithm can be easily extended to any application which uses CSG representations for component objects. The paper provides a complete description of the novel volume calculation algorithm, along with a discussion of the conjectured error bounds on volumes calculated within the method. In addition, numerical results comparing the new algorithm with a standard stochastic volume calculation algorithm are presented for a series of problems spanning a range of representative component sizesmore » and complexities. (authors)« less
[1]
E. B. Wilson.
Probable Inference, the Law of Succession, and Statistical Inference
,
1927
.
[2]
D. C. Carpenter,et al.
The MC21 Monte Carlo Transport Code
,
2007
.
[3]
ARISTIDES A. G. REQUICHA,et al.
Representations for Rigid Solids: Theory, Methods, and Systems
,
1980,
CSUR.
[4]
Hanan Samet,et al.
The Quadtree and Related Hierarchical Data Structures
,
1984,
CSUR.
[5]
Joshua Levin,et al.
A parametric algorithm for drawing pictures of solid objects composed of quadric surfaces
,
1976,
CACM.
[6]
J. F. Briesmeister.
MCNP-A General Monte Carlo N-Particle Transport Code
,
1993
.