考虑专家信息的威布尔型产品定时截尾可靠性验收试验方案设计

doi:10.12305/j.issn.1001-506X.2022.04.40

系统工程与电子技术 ›› 2022, Vol. 44 ›› Issue (4): 1409-1416.doi: 10.12305/j.issn.1001-506X.2022.04.40

考虑专家信息的威布尔型产品定时截尾可靠性验收试验方案设计

谭尧¹, 赵骞², 王文峰³, 郭波^1,*, 蒋平¹

1. 国防科技大学系统工程学院, 湖南长沙 410005
2. 国防科技大学信息通信学院, 陕西西安 710106
3. 中国人民解放军93209部队, 北京 100085

收稿日期:2021-02-07 出版日期:2022-04-01 发布日期:2022-04-01
通讯作者: 郭波
作者简介:谭尧(1997—), 男, 硕士研究生, 主要研究方向为可靠性评估、装备综合保障|赵骞(1994—), 男, 助教, 硕士, 主要研究方向为系统可靠性分析、装备试验鉴定|王文峰(1978—), 男, 高级工程师, 博士, 主要研究方向为雷达装备综合论证与保障|郭波(1962—), 男, 教授, 博士研究生导师, 博士, 主要研究方向为可靠性、项目管理|蒋平(1976—), 男, 副教授, 博士, 主要研究方向为系统可靠性
基金资助:
国家自然科学基金(71871218);国家自然科学基金(72071208)

Type I censored reliability acceptence test plan for Weibull distributed products by considering expert information

Yao TAN¹, Qian ZHAO², Wenfeng WANG³, Bo GUO^1,*, Ping JIANG¹

1. College of Systems Engineering, National University of Defense Technology, Changsha 410005, China
2. College of Information Communication, National University of Defense Technology, Xi'an 710106, China
3. Unit 93209 of the PLA, Beijing 100085, China

Received:2021-02-07 Online:2022-04-01 Published:2022-04-01
Contact: Bo GUO

摘要/Abstract

摘要：

可靠性验收试验是检验产品是否达到可靠性要求而进行的试验。试验方案的确定需要保证使用方风险和生产方风险可控。当前GJB899A中定时截尾试验方案往往需要较长的试验时间, 短时试验方案又存在两类风险较大的问题, 实际应用颇受诟病。根本原因在于GJB899A的试验方案确定只考虑了系统试验数据, 而忽视了其多源信息。因此, 本文针对复杂系统验收试验前存在多种形式专家信息, 考虑系统寿命服从威布尔分布的情况, 结合专家信息来计算产品在给定试验方案下的两类风险, 进一步推导得到试验时间较短、风险可控的试验方案。文章最后给出了案例分析, 通过对比, 验证了所提方法的有效性。

关键词: 威布尔分布, 专家信息, 生产方风险, 使用方风险, 可靠性验收试验

Abstract:

The reliability acceptance test is to check whether the product meets the reliability requirement. It is essential to choose a test plan with controllable customer's and producer's risks. However, the reliability acceptance test adopted from GJB899A oftentimes leads to either long test duration or high customer's and producer's risks, which hinds its real-world application. The primary reason lies in that, only the test data collected from the systems are taken into account, whereas the information from experts is ignored. Considering there are various types of expert information and the system life follows the Weibull distribution, the two types of risks of the product can be calculated. Then the test plan with shorter test time and controllable risks can be determined. The proposed method is applied to a numerical case study and the results verify the effectiveness of the method.

Key words: Weibull distribution, expert information, producer's risk, customer's risk, reliability acceptance test

中图分类号:

TN306

谭尧, 赵骞, 王文峰, 郭波, 蒋平. 考虑专家信息的威布尔型产品定时截尾可靠性验收试验方案设计[J]. 系统工程与电子技术, 2022, 44(4): 1409-1416.

Yao TAN, Qian ZHAO, Wenfeng WANG, Bo GUO, Ping JIANG. Type I censored reliability acceptence test plan for Weibull distributed products by considering expert information[J]. Systems Engineering and Electronics, 2022, 44(4): 1409-1416.

图/表 5

图1

图2

表1

表2

表3

参考文献 21

1	WASSERMAN G . Reliability verification, testing, and analysis in engineering design[M]. New York: CRC Press, 2002.
2	祝耀昌. MIL-HDBK-781美国军用手册简介[J]. 航空标准化与质量, 1986, (3): 52- 56.
	ZHU Y C . An introduction of the MIL-HDBK-781 US military manual[J]. Aeronautic Standardization & Quality, 1986, (3): 52- 56.
3	GJB899A-2009. 可靠性鉴定和验收试验[S]. 北京: 中国人民解放军总装备部, 2009.
	GJB899A-2009. Reliability testing for qualification and production[S]. Beijing: General Reserve Department of PLA, 1987.
4	孙扩, 赵波, 杨航. 基于三参数威布尔可靠性模型的航空装备故障预测研究[J]. 航空维修与工程, 2020, 352 (10): 51- 54.
	SUN K , ZHAO B , YANG H . Research on aviation equipment failure prediction based on three parameter Weibull reliability model[J]. Aviation Maintenance and Engineering, 2020, 352 (10): 51- 54.
5	JUN C H , BALAMURALI S , LEE S H . Variables sampling plans for Weibull distributed lifetimes under sudden death testing[J]. IEEE Trans.on Reliability, 2006, 55 (1): 53- 58. doi: 10.1109/TR.2005.863802
6	TSAI T R , LU Y T , WU S J . Reliability sampling plans for Weibull distribution with limited capacity of test facility[J]. Computers & Industrial Engineering, 2008, 55 (3): 721- 728.
7	CHEN J W , LI K H , LAM Y . Bayesian single and double variable sampling plans for the Weibull distribution with censoring[J]. European Journal of Operational Research, 2007, 177 (2): 1062- 1073. doi: 10.1016/j.ejor.2005.11.023
8	BALAMURALI S , JEYADURGA P , USHA M . Optimal design of repetitive group sampling plans for Weibull and Gamma distributions with applications and comparison to the Birnbaum-Saunders distribution[J]. Journal of Applied Statistics, 2018, 45 (14): 2499- 2520. doi: 10.1080/02664763.2018.1426740
9	杨军, 武小悦, 马溧梅. 可靠性试验评定中专家信息融合[J]. 航空计算技术, 2007, 38 (5): 14- 17. doi: 10.3969/j.issn.1671-654X.2007.05.004
	YANG J , WU X Y , MA S M . Aggregation of expert judgments in reliability test and estimation[J]. Aeronautical Computing Technique, 2007, 38 (5): 14- 17. doi: 10.3969/j.issn.1671-654X.2007.05.004
10	明志茂, 张云安, 陶俊勇, 等. 二项分布Bayes鉴定试验次数的选择与分析[J]. 系统工程与电子技术, 2008, 30 (12): 2512- 2515. doi: 10.3321/j.issn:1001-506X.2008.12.054
	MING Z M , ZHANG Y A , TAO J Y , et al. Analysis of test scheme for a Bayes plan of qualification test in binomial case[J]. Systems Engineering and Electronics, 2008, 30 (12): 2512- 2515. doi: 10.3321/j.issn:1001-506X.2008.12.054
11	ZHANG X G , MAHADEVAN S , DENG X Y . Reliability analysis with linguistic data: an evidential network approach[J]. Reliability Engineering & System Safety, 2017, 162 (C): 111- 121.
12	ZHAO Q , JIA X , CHENG Z J , et al. Bayesian estimation of resi-dual life for Weibull-distributed components of on-orbit satellites based on multi-source information fusion[J]. Applied Sciences, 2019, 9 (15): 3017. doi: 10.3390/app9153017
13	ZHAO Q, JIA X, GUO B. Bayesian estimation of residual life for Weibull distributed products based on the fusion of different forms of expert information[C]//Proc. of the IEEE 19th International Conference on Software Quality, Reliability and Secu-rity Companion, 2019.
14	ZHAO Q, JIA X, SONG Z L, et al. Bayesian estimation of residual life for Weibull distributed components by fusing expert knowledge[C]//Proc. of the International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, 2019: 9021172.
15	ZHU T , YAN Z Z , PENG X . A Weibull failure model to the study of the hierarchical Bayesian reliability[J]. Eksploatacja i Niezawodnosc-Maintenance and Reliability, 2016, 18 (4): 501- 506. doi: 10.17531/ein.2016.4.4
16	周晓东, 汤银才, 费鹤良. 删失数据威布尔分布参数的贝叶斯统计分析[J]. 上海师范大学学报(自然科学版), 2008, 37 (1): 28- 34. doi: 10.3969/j.issn.1000-5137.2008.01.005
	ZHOU X D , TANG Y C , FEI H L . Bayesian estimation of the parameters of the Weibull distribution based on censored samples[J]. Journal of Shanghai Normal University(Natural Sciences), 2008, 37 (1): 28- 34. doi: 10.3969/j.issn.1000-5137.2008.01.005
17	韩峰, 陆希成, 刘钰, 等. 二项分布中成功概率的贝叶斯序贯检验方法[J]. 航空动力学报, 2013, 28 (2): 270- 274.
	HAN F , LU X C , LIU Y , et al. Bayesian sequential test method for probability of success of binomial distribution based on posterior probability[J]. Journal of Aerospace Power, 2013, 28 (2): 270- 274.
18	GUO J , LI Z J , KEYSER T . A Bayesian approach for integrating multilevel priors and data for aerospace system reliability assessment[J]. Chinese Journal of Aeronautics, 2018, 31 (1): 41- 53. doi: 10.1016/j.cja.2017.08.020
19	ADAMCZAK R , BEDNORZ W . Some remarks on MCMC estimation of spectra of integral operators[J]. Bernoulli Official Journal of the Bernoulli Society for Mathematical Statistics & Probability, 2015, 21 (4): 2073- 2092.
20	李龙享, 曾声奎. 小样本下指数分布代替威布尔分布的研究[J]. 北京航空航天大学学报, 1995, 21 (4): 72- 77.
	LI L X , ZENG S K . Study on substituting the exponential distribution for the Weibull distribution under the condition of small sample[J]. Journal of Beijing University of Aeronautics and Astronautics, 1995, 21 (4): 72- 77.
21	于春雨, 苏子美, 孙永全, 等. 指数型产品多阶段可靠性增长的Bayes模型[J]. 中国机械工程, 2015, 26 (23): 3146- 3149. 3146-3149, 3155 doi: 10.3969/j.issn.1004-132X.2015.23.005
	YU C Y , SU Z M , SUN Y Q , et al. Bayes model of exponential product multi-stage reliability growth[J]. China Mechanical Engineering, 2015, 26 (23): 3146- 3149. 3146-3149, 3155 doi: 10.3969/j.issn.1004-132X.2015.23.005

验收方案序号	α	β	接收产品最大故障数C	投入试验样本数n	T/h	α+β
GJB899A	0.100 0	0.100 0	14	15	6 787	0.200 0
1	0.018 9	0.005 8	8	20	7 220	0.024 7
2	0.014 4	0.009 1	8	19	6 859	0.023 5
3	0.010 5	0.014 1	8	18	6 498	0.024 6
4	0.019 3	0.009 2	7	17	6 137	0.028 5
5	0.014 7	0.014 2	7	16	5 776	0.028 9
6	0.011 2	0.021 7	7	15	5 415	0.032 9
7	0.020 7	0.014 1	6	14	5 054	0.034 8
8	0.015 6	0.022 4	6	13	4 693	0.038 0
9	0.031 9	0.013 5	5	12	4 332	0.045 4
10	0.023 1	0.022 8	5	11	3 971	0.045 9
11	0.048 8	0.012 7	4	10	3 610	0.061 5
12	0.034 6	0.022 9	4	9	3 249	0.057 5
13	0.024 4	0.040 8	4	8	2 888	0.065 2
14	0.059 2	0.023 1	3	7	2 527	0.082 3
15	0.039 5	0.043 8	3	6	2 166	0.083 3
16	0.024 2	0.084 5	3	5	1 805	0.108 7
17	0.079 0	0.046 2	2	4	1 444	0.125 2

验收方案序号	α	β	接收产品最大故障数C	投入试验样本数n	T/h	α+β
GJB899A	0.100 0	0.100 0	6	7	2 232	0.200 0
1	0.012 2	0.031 2	4	10	2 400	0.043 4
2	0.033 7	0.014 6	3	9	2 160	0.048 3
3	0.025 3	0.027 7	3	8	1 920	0.053 0
4	0.017 9	0.051 9	3	7	1 680	0.069 8
5	0.066 4	0.022 8	2	6	1 440	0.089 2
6	0.047 1	0.047 3	2	5	1 200	0.094 4
7	0.031 2	0.098 2	2	4	960	0.129 4

专家信息	试验方案(T, n, C)	α	β
GJB899A	(9 930, 20, 19)	0.066 7	0.057 5
0.956 8^*	(7 220, 20, 11)	0.001 1	0.055 1
0.956 8^*	(7 220, 20, 10)	0.003 1	0.030 0
0.956 8^*	(7 220, 20, 9)	0.008 1	0.014 1
0.956 8^*	(7 220, 20, 8)	0.018 9	0.005 8
0.97	(7 220, 20, 9)	0.012 2	0.009 8
0.9	(7 220, 20, 9)	0.015 4	0.007 1
0.8	(7 220, 20, 10)	0.008 1	0.007 7
0.7	(7 220, 20, 10)	0.008 2	0.005 1
0.6	(7 220, 20, 10)	0.008 5	0.003 6
0.55	(7 220, 20, 11)	0.004 8	0.004 4

[1]	刘鸿彬, 赵骞, 贾祥, 郭波. 失效信息已知时r/n(G)表决系统的剩余寿命预测[J]. 系统工程与电子技术, 2022, 44(5): 1738-1746.
[2]	张怀强, 李铁成, 杜军岗. 舰载导弹系统故障率及保修期模型优化[J]. 系统工程与电子技术, 2022, 44(1): 347-356.
[3]	赵骞, 贾祥, 程志君, 郭波. 部件寿命服从威布尔分布时典型系统的寿命与剩余寿命估计[J]. 系统工程与电子技术, 2019, 41(7): 1665-1671.
[4]	邢云燕, 蒋平. 基于顺序Dirichlet分布的Bayes可靠性增长评估方法[J]. 系统工程与电子技术, 2017, 39(5): 1178-1182.
[5]	黄文平, 周经伦, 宁菊红, 金光. 基于竞争失效数据的Lindley分布参数估计[J]. 系统工程与电子技术, 2016, 38(2): 464-469.
[6]	翟亚利，张志华，李大伟. 基于冲击理论的性能可靠性评估研究[J]. 系统工程与电子技术, 2014, 36(10): 2108-2112.
[7]	李凌，徐伟. 威布尔产品加速寿命试验的可靠性分析[J]. Journal of Systems Engineering and Electronics, 2010, 32(7): 1544-1548.
[8]	钟强晖, 张志华, 吴和声. 基于退化数据的可靠性评估方法探讨[J]. Journal of Systems Engineering and Electronics, 2009, 31(9): 2280-2284.
[9]	罗航, 龙兵, 黄建国, 王厚军. 一种搜索威布尔MLE形状参数的PSO联合方法[J]. Journal of Systems Engineering and Electronics, 2009, 31(4): 988-993.

考虑专家信息的威布尔型产品定时截尾可靠性验收试验方案设计

Type I censored reliability acceptence test plan for Weibull distributed products by considering expert information

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 5

参考文献 21

相关文章 9

编辑推荐

Metrics

本文评价