微观燃耗方程的求解算法比较与性能分析

Comparison and Performance Analysis of Microscopic Depletion Equation Solving Method

1.
Shanghai Nuclear Engineering Research & Design Institute, Shanghai 200233, China
2.
Shanghai Jiao Tong University, Shanghai 200240, China

摘要: 燃耗方程的求解是燃耗计算的核心。常见的算法包括泰勒方法、Pade方法、子空间方法、切比雪夫有理近似方法和龙格库塔法等。通过数值实验，对每种算法在精度、效率、稳定性方面进行分析比较。结果表明：子空间方法、泰勒方法在计算效率方面具有优势；计算精度及稳定性方面，泰勒方法和Pade方法均占优势。综合考虑，泰勒方法在3个方面均表现突出，可作为燃耗计算的优选算法。
- 燃耗计算 ,
- 指数矩阵 ,
- 燃耗算法
Abstract: The solving of depletion equation is the key to carry burnup calculation out. The common methods are Taylor method, Pade method, Krylov method, Chebyshev rational approximation method (CRAM) and Runge-Kutta method, and so on. By means of numerical experiment, analysis comparison was undertaken from angles of efficiency, accuracy and stability. The results show that Krylov method and Taylor method take the advantage in efficiency, and Taylor mehod and Pade method take advantage in both accuracy and stability. In a word, Taylor method behaves best as a whole, which can be chosen as the candidate for the final depletion method.
- burnup calculation ,
- exponential matrix ,
- depletion method

[1]	HERMANN O W, WESTFALL R M. ORIGEN-S: Scale system module to calculate fuel depletion, actinide transmutation, fission product buildup and decay, and associated radiation source terms, NUREG/CR0200/Rev 6[R]. USA: Oak Ridge National Laboratory, 1998.
[2]	MOLER C, van LOAN C. Nineteen dubious ways to compute the exponential of a matrix, twenty-five years[J]. Society for Industrial and Applied Mathematics Review, 2003, 45（1）: 3-49.
[3]	ARIOLI M, CODENOTTI B, FASSINO C. The Pade method for computing the matrix exponential[J]. Linear Algebra and its Applications, 1996, 240（6）: 111-130.
[4]	YAMAMOTO A, MASAHIRO T, NAOKI S. Numerical solution of stiff burnup equation with short half lived nuclides by the Krylov subspace method[J]. J Nucl Sci Technol, 2007, 44（2）: 147-154.
[5]	PUSA M, LEPPANEN J. Computing the matrix exponential in burnup calculations[J]. Nucl Sci Eng, 2010, 164（2）: 140-150.
[6]	CASTILLO P, SAAD Y. Preconditioning the matrix exponetial operator with applications[J]. Journal of Scinetific Computing, 1998,13（3）: 275-302.