约束空间近似正交的空间填充试验设计方法

doi:10.12305/j.issn.1001-506X.2021.07.14

Abstract

Abstract:

The existence of constraints in experimental design space is a widespread practical engineering problem. In view of the problems in the existing experimental design methods for constrained spaces, such as inflexibility in generating design points, long optimization time and low efficiency, limited applicability for different types of constraints, single design criteria, and so on, an experimental design method of nearly-orthogonal space-filling for constrained spaces is proposed. The experimental design criterion is constructed based on the distance between design points and the correlation coefficient, and the scheme is solved by the improved stochastic coordinate-exchange algorithm. The proposed algorithm is suitable for convex constraints, non-convex constraints, analytical constraints, non-analytical constraints and other types of constraints, and it is suitable for multi-dimensional irregular experimental design space. The analysis of the example shows that the algorithm proposed in this paper has better space-filling characteristics and orthogonality compared with the existing methods.

Key words: orthogonality, space-filling, constrained spaces, coordinate-exchange

CLC Number:

Yang YOU, Guang JIN, Zhengqiang PAN, Rui GUO. Nearly-orthogonal space-filling experimental design method for constrained spaces[J]. Systems Engineering and Electronics, 2021, 43(7): 1831-1837.

Figures/Tables 7

Table 1

Fig.1

Fig.2

Fig.3

Table 2

Fig.4

Table 3

References 30

1	MONTGOMERY D C . Experimental designs for constrained r egions[J]. Quality Engineering, 2002, 14 (4): 587- 601. doi: 10.1081/QEN-120003561
2	AUFFRAY Y , BARBILLON P , MARIN J M . Maximin design on nonhypercube domains and kernel interpolation[J]. Statistics and Computing, 2012, 22 (3): 703- 712. doi: 10.1007/s11222-011-9273-9
3	LEKIVETZ R , JONES B . Fast flexible space-filling designs for nonrectangular regions[J]. Quality and Reliability Engineering International, 2015, 31 (5): 829- 837. doi: 10.1002/qre.1640
4	LEKIVETZ R , JONES B . Fast flexible space-filling designs with nominal factors for nonrectangular regions[J]. Quality and Reliability Engineering International, 2019, 35 (2): 677- 684. doi: 10.1002/qre.2429
5	MAK S , JOSEPH V R . Minimax and minimax projection designs using clustering[J]. Journal of Computational and Grap hical Statistics, 2018, 27 (1): 166- 178. doi: 10.1080/10618600.2017.1302881
6	PERRIN G , CANNAMELA C . A repulsion-based method for the definition and the enrichment of optimized space filling designs in constrained input spaces[J]. Journal of the French Statistical Society, 2017, 158 (1): 37- 67.
7	FUERLE F , SIENZ J . Formulation of the audze-Eglais uniform Latin hypercube design of experiments for constrained design spaces[J]. Advances in Engineering Software, 2011, 42 (9): 680- 689. doi: 10.1016/j.advengsoft.2011.05.004
8	BEAL M. CLAEYS B M . Constructing space-filling designs using an adaptive WSP algorithm for spaces with constraints[J]. Chemometrics and Intelligent Laboratory Systems, 2014, 133, 84- 91. doi: 10.1016/j.chemolab.2013.11.009
9	KANG L L . Stochastic coordinate-exchange optimal designs with complex constraints[J]. Quality Engineering, 2019, 31 (3): 401- 416. doi: 10.1080/08982112.2018.1508695
10	YANG D S, DONG M Z. A novel criterion and searching algorithm for constructing Latin hypercube designs[C]//Proc. of the International Conference on Advanced Robotics and Mechatronics. Macau, 2016.
11	ZHANG X R , LIU M Q , ZHOU Y . Orthogonal uniform composite designs[J]. Journal of Statistical Planning and Inference, 2020, 206, 100- 110. doi: 10.1016/j.jspi.2019.08.007
12	王东辉, 武泽平. 约束域优化拉丁超立方设计方法[P]. 中国: CN106202667A, 2016-12-07.
	WANG D H, WU Z P. Optimized latin hypercube design method for constrained spaces[P]. China: CN106202667A, 2016-12-07.
13	WU Z P , WANG D H . Space-filling experimental designs for constrained design spaces[J]. Engineering Optimization, 2019, 51 (9): 1495- 1508. doi: 10.1080/0305215X.2018.1542691
14	JOSEPH V R , YING H . Orthogonal-maximin latin hypercube designs[J]. Statistica Sinica, 2008, 18 (1): 171- 186.
15	刘新亮, 郭波. 基于改进ESE算法的多目标优化试验设计方法[J]. 系统工程与电子技术, 2011, 32 (2): 410- 414.
	LIU X L , GUO B . Multi-objective experimentation design optimization based on modified ESE algorithms[J]. Systems Engineering and Electronics, 2011, 32 (2): 410- 414.
16	张昆仑, 郭波. 基于改进ILS算法的多目标优化试验设计[J]. 计算机工程, 2011, 37 (5): 273- 275. doi: 10.3969/j.issn.1000-3428.2011.05.093
	ZHANG K L , GUO B . Experimental design of multi-objective optimization based on modified ILS algorithm[J]. Computer Engineering, 2011, 37 (5): 273- 275. doi: 10.3969/j.issn.1000-3428.2011.05.093
17	WANG C Y , YANG J Y , LIU M Q . Construction of space-filling orthogonal designs[J]. Journal of Statistical Planning and Inf erence, 2021, 213, 130- 141. doi: 10.1016/j.jspi.2020.11.009
18	WU Z P , WANG D H , OKOLO P N , et al. Efficient space-filling and near-orthogonality sequential latin hypercube for computer experiments[J]. Computer Methods in Applied Mechanics & Engineering, 2017, 324 (1): 348- 365.
19	SU Z R , WANG Y P , ZHOU Y C . On maximin distance and nearly orthogonal Latin hypercube designs[J]. Statistics & Probability Letters, 2020, 166, 108876.
20	杨雪, 周琦. 一类正交空间填充设计的构造[J]. 系统科学与数学, 2020, 40 (2): 289- 297.
	YANG X , ZHOU Q . The construction of orthogonal space filling designs[J]. Journal of Systems Science and Mathematical Sciences, 2020, 40 (2): 289- 297.
21	牟唯嫣, 崔栋利. 计算机实验的正交空间填充设计[J]. 系统科学与数学, 2015, 35 (12): 1457- 1462.
	MU W Y , CUI D L . Orthogonal space-filling designs for computer experiments[J]. Journal of Systems Science and Mathematical Sciences, 2015, 35 (12): 1457- 1462.
22	JIN R C , CHEN W , SUDJIANTO A . An efficient algorithm for constructing optimal design of computer experiments[J]. Journal of Statistical Planning and Inference, 2005, 134 (1): 268- 287. doi: 10.1016/j.jspi.2004.02.014
23	谢描. 近似正交设计[D]. 北京: 清华大学, 2004.
	XIE M. Nearly orthogonal designs[D]. Beijing: Tsinghua University, 2004.
24	JOSEPH V R , GUL E , BA S . Maximum projection designs for computer experiments[J]. Biometrika, 2015, 102 (2): 371- 380. doi: 10.1093/biomet/asv002
25	MORRIS M D , MITCHELL T J . Exploratory designs for computational experiments[J]. Journal of Statistical Planning and Inference, 1995, 43 (3): 381- 402. doi: 10.1016/0378-3758(94)00035-T
26	IMAN R L , CONOVER W J . Small sample sensitivity analysis techniques for computer models, with an application to risk assessment[J]. The American Statistician Communications in Statistics: Theory Methods, 1980, 9 (17): 1749- 1845. doi: 10.1080/03610928008827996
27	CONOVER W J . Practical nonparametric statistics[M]. New York: John Wiley and Sons Incorporated, 1999.
28	OWEN A B . Controlling correlations in Latin hypercube samples[J]. The American Statistician Journal of the American Statistical Association: Theory and Methods, 1994, 23 (10): 1517- 1522.
29	李为民, 陈刚, 黄仁全. 一种近正交试验设计方法[J]. 空军工程大学学报(自然科学版), 2010, 11 (3): 84- 88. doi: 10.3969/j.issn.1009-3516.2010.03.019
	LI W M , CHEN G , HUANG R Q . A method of nearly or thogonal experimental design[J]. Journal of Air Force Engineering University (Natural Science Edition), 2010, 11 (3): 84- 88. doi: 10.3969/j.issn.1009-3516.2010.03.019
30	KAI H , ALEXANDER A . The point in polygon problem for arbitrary polygons[J]. Computational Geometry, 2001, 20 (3): 131- 144. doi: 10.1016/S0925-7721(01)00012-8

设计结果对比	设计方法
设计结果对比	MO-SCE		SF-SCE
实验设计矩阵	-0.118 085 4	0.388 962 66	0.684 245 17	-0.729 252 04
	-0.481 210 4	-0.088 931 35	0.780 969 88	0.624 568 69
	-0.986 807 1	0.161 900 34	-0.115 731 25	-0.453 720 89
	0.125 649 5	0.992 074 69	-0.995 879 04	-0.090 691 45
	0.088 457 5	-0.369 740 72	-0.443 171 48	-0.896 436 86
	-0.802 788 3	0.596 264 16	0.171 774 83	-0.985 136 24
	0.981 662 7	-0.190 626 06	0.408 751 58	-0.133 227 01
	0.411 611 0	-0.911 359 65	-0.777 281 80	-0.629 152 60
	-0.394 724 4	0.918 799 56	-0.562 111 53	0.827 061 44
	0.455 593 3	0.091 650 22	-0.890 377 84	0.455 222 26
	-0.561 403 0	-0.827 542 53	0.938 115 09	-0.346 323 66
	0.899 504 3	0.436 911 91	0.445 711 37	0.895 176 73
	0.608 468 6	0.793 577 95	0.989 691 68	0.143 214 47
	0.780 627 3	-0.624 996 86	0.141 466 27	0.392 416 61
	-0.143 843 8	-0.989 258 13	-0.383 351 87	0.132 873 13
	-0.851 157 8	-0.516 591 76	-0.043 186 79	0.999 067 02
相关性准则	2.9×10^-4		1.5×10^-2
空间填充准则ϕ₂	9.24		9.43

设计点数	MO-SCE		SF-SCE
设计点数	相关性准则	空间填充准则ϕ₁₀	相关性准则	空间填充准则ϕ₁₀
5	1.2×10^-4	0.83	1.6×10^-1	0.82
16	3.7×10^-6	1.98	1.4×10^-3	1.94
30	6.3×10^-6	3.14	2.1×10^-2	3.13
100	1.2×10^-3	8.44	3.2×10^-2	7.88

设计结果对比	设计方法
设计结果对比	MO-SCE		SF-SCE
试验设计矩阵	0.089 168 36	0.999 752 4	0.088 967 29	0.999 436 4
	0.088 971 11	0.999 400 0	0.088 947 07	0.999 725 4
	0.089 008 87	0.999 823 3	0.088 863 19	0.999 630 5
	0.088 740 56	0.999 664 1	0.088 802 91	0.999 765 1
	0.088 886 33	0.999 758 7	0.089 166 36	0.999 711 1
	0.088 904 22	0.999 997 0	0.088 761 26	0.999 891 7
	0.089 096 84	0.999 620 6	0.089 019 38	0.999 544 5
	0.088 666 74	0.999 756 1	0.089 103 33	0.999 806 6
	0.088 945 48	0.999 555 6	0.088 925 99	0.999 973 4
	0.089 111 61	0.999 872 6	0.089 061 94	0.999 663 8
	0.088 840 55	0.999 898 6	0.088 692 22	0.999 803 6
	0.088 837 72	0.999 604 0	0.088 872 64	0.999 846 2
	0.088 983 41	0.999 682 2	0.088 741 01	0.999 663 2
	0.088 735 77	0.999 848 4	0.089 002 55	0.999 858 1
相关性准则	0.09		0.21
空间填充准则ϕ₂₀	3.88		4.02

Nearly-orthogonal space-filling experimental design method for constrained spaces

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 7

References 30

Related Articles 1

Recommended Articles

Metrics

Comments