基于共轭离散纵标的减方差方法

郑征; 丁谦学; 周岩

doi:10.11884/HPLPB201830.170223

基于共轭离散纵标的减方差方法

doi: 10.11884/HPLPB201830.170223

上海核工程研究设计院有限公司, 上海 200233

详细信息

作者简介:
郑征(1985—), 男，工程师，从事核电厂屏蔽设计和分析工作; zhengzheng@snerdi.com.cn

中图分类号: TL328
计量
- 文章访问数: 800
- HTML全文浏览量: 154
- PDF下载量: 84
- 被引次数: 0
出版历程
- 收稿日期: 2017-06-20
- 修回日期: 2017-08-20
- 刊出日期: 2018-02-15

Variance reduction method based on adjoint discrete ordinate

Shanghai Nuclear Engineering Research and Design Institute Ltd Co, Shanghai 200233, China

摘要

摘要: 对于深穿透类型的屏蔽计算，为了得到较为可信的统计结果，蒙特卡罗方法(MC方法)需要模拟大量的粒子，巨大的计算时间是其存在的主要问题。源偏倚和权窗技巧能够有效降低深穿透问题的计数误差。开展了基于共轭离散纵标(SN)的MC减方差方法研究，根据SN方法的共轭注量率计算并生成了源偏倚和权窗参数，编写了JMCT程序的源抽样子程序，并且在秦山一期测量值基础上进行了验证，成功应用到CAP1400压力容器快中子注量率和堆腔中子和光子剂量率计算中。数值结果表明，对于深穿透屏蔽计算问题，和无偏的MC方法相比，基于共轭SN的MC减方差方法能够在保证结果精度的前提下，提高计算效率1~2个量级。
- 深穿透 /
- 离散纵标方法 /
- 蒙特卡罗方法 /
- 源偏倚 /
- 权窗
Abstract: For deep-penetration shielding calculation, Monte Carlo method (MC method) requires modeling a great number of particles to obtain reliable results, thus huge computation time is the main problem of the MC method. Source biasing and weight window technique effectively decrease the tally error of deep penetration problem. This paper studies the variance reduction (VR) method based on adjoint Discrete Ordinate (SN), generates source biasing factors and weight window parameters for the MC method by using the adjoint fluence rates of the SN method, and develops source sampling subroutine for JMCT. The VR method was verified at phaseⅠ of Qinshan Nuclear Power Plant measurements. It was applied to CAP1400 pressure vessel fast neutron fluence rate and cavity neutron and photon dose rate calculations. Numerical results show that the VR method based on SN increases calculation efficiency by 1~2 orders for deep-penetration shielding calculation with high precision compared with unbiased MC method.
- deep-penetration /
- discrete ordinate method /
- Monte Carlo method /
- source biasing /
- weight window

HTML全文

图 1 基于共轭SN的MC减方差方法流程图

Figure 1. Flowchart of MC variance reduction method based on adjoint SN

下载: 全尺寸图片幻灯片

图 2 秦山一期JMCT计算模型示意图

Figure 2. JMCT model for phase Ⅰ of Qianshan Nuclear Power Plant

下载: 全尺寸图片幻灯片

图 3 秦山一期堆芯中平面JSNT网格划分示意图

Figure 3. JSNT mesh for core mid-plane of phase Ⅰ of Qianshan Nuclear Power Plant

下载: 全尺寸图片幻灯片

图 4 秦山一期前100万无偏和偏倚源粒子分布示意图

Figure 4. First million unbiased and biased source particles for phase Ⅰ of Qianshan Nuclear Power Plant

下载: 全尺寸图片幻灯片

图 5 CAP1400压力容器内表面快中子注量率JMCT和JSNT计算模型示意图

Figure 5. JMCT and JSNT models for CAP1400 RPV calculation

下载: 全尺寸图片幻灯片

图 6 CAP1400压力容器内表面JSNT共轭源和JMCT前100万源粒子示意图

Figure 6. Adjoint source of JSNT and first million biased source particles of JMCT

下载: 全尺寸图片幻灯片

图 7 CAP1400压力容器内表面快中子注量率(E≥1.0 MeV)分布

Figure 7. FNFR distribution (E≥1.0 MeV) on inner RPV of CAP1400

下载: 全尺寸图片幻灯片

图 8 CAP1400堆腔剂量率JMCT和JSNT计算模型示意图

Figure 8. JMCT and JSNT models for CAP1400 cavity dose rate calculation

下载: 全尺寸图片幻灯片

表 1 秦山一期快中子注量率(E≥1.0 MeV)计数误差小于0.05时计算总时间和加速比

Table 1. Calculation time and speedup for different cases with fast neutron fluence rate (FNFR) tally error less than 0.05

case	geometry type	number of mesh	fast neutron fluence rate/(10⁹ n·cm^-2·s^-1)	JSNT time/min	JMCT time/min	total time T/min	speedup (T_bias/ T_unbias)
measure	-	-	7.41	-	-	-	-
1(unbiased)	-	-	6.08	-	19 814.12	19 814.12	1
2	xyz	393 120	6.48	19.66	3 575.56	3 595.22	6
3	xyz	393 120	6.35	19.66	197.92	217.58	91
4	xyz	393 120	6.41	19.66	30.00	49.66	399
5	rtz	402 948	5.92	80.83	100.78	181.61	109
6	rtz	823 368	6.44	161.66	429.62	591.28	34
7	rtz	1 857 492	6.53	397.59	185.07	582.66	34

下载: 导出CSV

表 2 CAP1400压力容器内表面快中子注量率各个计数误差区间内的网格数目

Table 2. Number of mesh in each tally error bin on inner RPV of CAP1400

tally error		number of mesh
upper	lower	unbiased	biased
0.00	0.01	0	0
0.01	0.02	0	1
0.02	0.03	0	311
0.03	0.04	0	350
0.04	0.05	343	119
0.05	0.06	322	48
0.06	0.07	114	16
0.07	0.08	50	8
0.08	0.09	21	1
0.09	0.10	7	1
0.10	1.00	3	5

下载: 导出CSV

表 3 CAP1400堆腔剂量率中子和光子剂量率及加速比

Table 3. Neutron dose rate, photon dose rate and speedup for CAP1400 cavity dose rate calculation

		dose rate/(mSv·h^-1)	tally error/10^-2	sum of time/min	M/10^-2	JMCT speedup(M_biased/M_ref)
case1: particle type	neutron	2.82×10⁶	7.14	10 612.57	1.85	1
biased source (reference)	photon	6.82×10⁴	5.35	10 612.57	3.29	1
case2: neutron and photon	neutron	2.80×10⁶	1.30	10 384.53	57	31
fluence rate response function	photon	6.50×10⁴	3.15	10 384.53	9.70	3
case3: neutron and photon	neutron	2.82×10⁶	1.15	10 064.27	75.1	41
dose rate response function	photon	6.27×10⁴	1.74	10 064.27	32.8	10

下载: 导出CSV

参考文献(9)

[1]	Wagner J C. Acceleration of Monte Carlo shielding calculations with an automated variance reduction technique and parallel processing[R]. University Park, 1997.
[2]	Wagner J C, Haghighat A. Automated variance reduction of Monte Carlo shielding calculations using the discrete ordinates adjoint function[J]. Nuclear Science and Engineering, 1998, 128: 186-208. doi: 10.13182/NSE98-2
[3]	Haghighat A, Wagner J C. Monte Carlo variance reduction with deterministic importance functions[J]. Progress in Nuclear Energy, 2003, 42(1): 25-53. doi: 10.1016/S0149-1970(02)00002-1
[4]	Blakeman E D, Peplow D E, Wagner J C, et al. PWR facility dose modeling using MCNP5 and the CADIS/ADVANTG variance-reduction methodology[R]. ORNL/TM-2007/133, 2007.
[5]	Peplow D E, Evans T M, Wagner J C. Simultaneous optimization of tallies in difficult shielding problems[J]. Nuclear Technology, 2008, 168: 785-792.
[6]	Wagner J C, Peplow P E, Evans T M. Automated variance reduction applied to nuclear well-logging problems[J]. Nuclear Technology, 2009, 168: 799-809. doi: 10.13182/NT09-A9309
[7]	Vasiliev A, Ferroukhi H, Zimmermann M A, et al. Development of a CASMO-4/SIMULATE-3/MCNPX calculation scheme for PWR fast neutron fluence analysis and validation against RPV scraping test data[J]. Annals of Nuclear Energy, 2007, 34: 615-627. doi: 10.1016/j.anucene.2007.02.020
[8]	Fero A H. SORCERY user manual[R]. LTR- R EA-06-74, 2006.
[9]	X-5 Monte Carlo Team, MCNP - Ageneral Monte Carlo N-Particle transport code[R]. LA-UR-03-1987, 2003.