郭華,祝逢春,豆仁福,翟樹峰,李敬玉
引信步進(jìn)應(yīng)力加速壽命試驗無失效數(shù)據(jù)情況貯存壽命評估
郭華,祝逢春,豆仁福,翟樹峰,李敬玉
(北京航空工程技術(shù)研究中心,南京 210028)
針對某引信加速壽命試驗出現(xiàn)無失效數(shù)據(jù)情況無評估方法問題,開展貯存壽命評估研究。根據(jù)某引信特點和樣品條件,選取采用步進(jìn)應(yīng)力加速壽命試驗。因本次樣品量過少,不能進(jìn)行摸底試驗來確定合理的應(yīng)力和步長,加速壽命試驗只能采用其預(yù)估值,因而試驗結(jié)果出現(xiàn)了無失效情況。為此提出先對無失效數(shù)據(jù)按貝葉斯統(tǒng)計方法將零失效比率(失效數(shù)/樣本數(shù))數(shù)據(jù)轉(zhuǎn)換為非降序失效比率的失效數(shù)據(jù),再按有失效數(shù)據(jù)的處理方法評估貯存壽命。以此方法編制計算程序,對某引信無失效試驗數(shù)據(jù)進(jìn)行處理,評估其貯存壽命為20.1 a。采用貝葉斯統(tǒng)計規(guī)律處理無失效數(shù)據(jù)的方法有效可行,解決了引信步進(jìn)應(yīng)力加速壽命試驗無失效數(shù)據(jù)情況下的貯存壽命評估難題。
引信;步進(jìn)應(yīng)力;加速壽命試驗;無失效數(shù)據(jù);不可靠度;貯存壽命;壽命評估
引信貯存壽命評估有基于自然貯存樣品試驗數(shù)據(jù)統(tǒng)計評估法和加速壽命試驗數(shù)據(jù)評估法。對于新產(chǎn)品和生產(chǎn)批次少的產(chǎn)品,通常只能采用加速壽命試驗來進(jìn)行評估。加速壽命試驗是用以研究產(chǎn)品可靠貯存壽命的一種有效方法。按應(yīng)力施加方式不同,加速壽命試驗可以分為恒定應(yīng)力加速壽命試驗、步進(jìn)(階梯)應(yīng)力加速壽命試驗和序進(jìn)應(yīng)力加速壽命試驗[1],其中步進(jìn)應(yīng)力加速壽命試驗(簡稱步加試驗)是一種常用方法。在正式進(jìn)行加速試驗前,往往需要進(jìn)行科學(xué)合理的試驗設(shè)計,以便獲得理想的試驗數(shù)據(jù),取得好的試驗效果。良好的試驗設(shè)計要求每個測試階段都有產(chǎn)品失效,為達(dá)到產(chǎn)品出現(xiàn)失效的要求,最有效的方法是進(jìn)行摸底試驗。對于樣品不足的產(chǎn)品,又無條件進(jìn)行摸底試驗,只能靠理論分析或憑借研究經(jīng)驗來確定。這給試驗設(shè)計帶來一定盲目性,因而很可能出現(xiàn)無失效或過多失效現(xiàn)象。在進(jìn)行某引信步加試驗時,由于樣品量的不足,無法進(jìn)行摸底試驗來確定合適的加速應(yīng)力和步長,因而采用預(yù)估的應(yīng)力水平和步長加速,結(jié)果出現(xiàn)了無失效或零失效現(xiàn)象。針對這種無失效數(shù)據(jù)情況的貯存壽命評估,查找了大量的加速壽命試驗數(shù)據(jù)處理方法文獻(xiàn),最相關(guān)的有文獻(xiàn)[2-19],但這些文獻(xiàn)均不涉及利用無失效試驗數(shù)據(jù)評估貯存壽命的方法。為此,本文引入貝葉斯統(tǒng)計方法,優(yōu)選先驗分布密度模型,有效地利用無失效數(shù)據(jù)評估了某引信的貯存壽命。該貯存壽命評估結(jié)果被某研究項目采納,后經(jīng)驗證符合工程實際。
正常的步加試驗數(shù)據(jù)的處理或貯存壽命評估方法[20-21]較為成熟。本文方法所不同的是針對無失效數(shù)據(jù)情況的數(shù)據(jù)進(jìn)行處理。本文采用貝葉斯方法,優(yōu)選先驗分布密度模型,將無失效數(shù)據(jù)情況轉(zhuǎn)換為有失效數(shù)據(jù)情況,從而按正常數(shù)據(jù)處理方法來處理數(shù)據(jù),評估貯存壽命。
有關(guān)數(shù)據(jù)處理研究表明,引信、彈藥貯存壽命符合威布爾分布[20-21]。假設(shè)在溫度應(yīng)力T下,產(chǎn)品的貯存壽命服從二參數(shù)威布爾分布,即:
式中:m為形狀參數(shù);η為特征壽命(真尺度參數(shù))。
在不同溫度應(yīng)力水平T下,存在不同的特征壽命η,η與T之間符合阿侖尼斯模型,即:
式中:和為待定參數(shù)。
累計得出應(yīng)力T下產(chǎn)品的實際試驗時間為:
1)似然函數(shù):
式中:為測試子樣序數(shù),=1,2,…,;n為每一測試點處樣本量;r為每一測試點處的失效數(shù);為威布爾分布形狀參數(shù);τ為每一測試點的等效試驗時間;η為特征壽命。式中η和τ分別由式(2)、(5)確定。
2)對數(shù)似然函數(shù)。對式(6)兩邊取對數(shù),則:
3)似然方程組。對式(7)取偏微分,得似然方程組:
采用貝葉斯統(tǒng)計方法處理無失效數(shù)據(jù),對每個檢測點失效比率(不可靠度點估計)作非零非降序處理。對于試驗數(shù)據(jù),記為:
式中:n、r和c(c=n?r)分別為t時刻的樣本數(shù)、失效數(shù)和成功數(shù);為檢測點總數(shù)。
當(dāng)r=0(=1,2,…,)時,可令:S=n+…+1,(1=1),并令不可靠度1的先驗密度[22]為:
式中:S、R和C分別為處理后的樣本數(shù)、失效數(shù)和成功數(shù)。
通過式(2)得出溫度應(yīng)力0時的特征壽命0的估計為:
水源熱泵項目取用地下水水資源論證技術(shù)要點分析……………………………………………… 江 劍,董殿偉(3.50)
參照式(1),正常貯存條件下的壽命分布函數(shù)[23]為:
對于給定置信度=1?和可靠貯存下限L,則對貯存壽命進(jìn)行區(qū)間估計為:
其中:
式中:μ為的下側(cè)分位點;為n或S中的最小者。
為評估某引信貯存壽命,筆者進(jìn)行了引信步進(jìn)應(yīng)力加速壽命試驗。由于樣品嚴(yán)重不足,無法進(jìn)行摸底試驗來確定合理的加速應(yīng)力和步長,因此通過分析進(jìn)行了試驗設(shè)計。在333 K的預(yù)估溫度應(yīng)力下,按5 K預(yù)估溫度步長進(jìn)行了步加壽命試驗,獲得某引信步加壽命試驗數(shù)據(jù),見表1。從表1看出,每個時間測試點失效數(shù)均為0(稱“零失效”或無失效),即未出現(xiàn)1發(fā)失效數(shù)據(jù)。
表1 引信步加試驗數(shù)據(jù)
Tab.1 Fuze step-stress accelerated life test data
因該引信產(chǎn)品的可靠性較高,預(yù)估的加速應(yīng)力和步長不夠合理,加速試驗出現(xiàn)了無失效數(shù)據(jù)情況,因而在正式進(jìn)行數(shù)據(jù)處理前,必須對現(xiàn)有無失效數(shù)據(jù)作預(yù)處理。由于作預(yù)處理計算時需要起始不可靠度數(shù)據(jù),故由某批104枚樣本試驗零失效的數(shù)據(jù)按二項分布求出不可靠度上限為0.022(置信水平0.9)。采用1.4節(jié)無失效數(shù)據(jù)情況的數(shù)據(jù)處理方法,得出無失效數(shù)據(jù)的預(yù)處理結(jié)果,見表2。從表2看出,經(jīng)過預(yù)處理的數(shù)據(jù)變?yōu)槭П嚷史墙敌虻恼?shù)據(jù)。
表2 無失效數(shù)據(jù)預(yù)處理結(jié)果
Tab.2 Processing results of zero-failure data
可靠貯存壽命是指一定可靠度下的壽命,因而評估可靠貯存壽命時,首先需要給定可靠度下限,然后再計算此可靠度下限條件下的可靠貯存壽命下限。根據(jù)引信的重要性、產(chǎn)品相關(guān)要求及用戶意見,確定引信可靠度下限L為0.975。根據(jù)1.5節(jié)的方法,編制Matlab程序,結(jié)合式(15)、(19),求解方程(18),對于給定可靠度下限L,計算得出可靠貯存壽命下限L=20.1 a。
通過上述數(shù)據(jù)處理方法和某引信可靠貯存壽命評估實例研究,可得出如下結(jié)論:
1)采用貝葉斯方法對無失效數(shù)據(jù)進(jìn)行預(yù)處理,推斷出相應(yīng)的有失效數(shù)據(jù),再按有失效數(shù)據(jù)的處理方法進(jìn)行處理,得出可靠度函數(shù),從而可推斷出正常應(yīng)力環(huán)境下產(chǎn)品的可靠度及其可靠貯存壽命下限,有效地解決了加速試驗無失效數(shù)據(jù)情況的數(shù)據(jù)處理難題。
2)采用上述無失效數(shù)據(jù)情況的數(shù)據(jù)處理方法,評估某引信可靠度下限為0.975時的可靠貯存壽命下限為20.1 a。這既滿足當(dāng)時引信延壽研究項目的要求,經(jīng)驗證后,又符合工程實際。
貯存壽命結(jié)論推斷的可信性依賴于試驗數(shù)據(jù)的準(zhǔn)確性和加速失效機(jī)理的不變性,同時加速模型和失效分布模型正確與否也影響推斷結(jié)果。因此,在進(jìn)行加速試驗時,施加應(yīng)力要適度,不能過大而改變失效機(jī)理,也不宜過小使試驗時間太長或出現(xiàn)無失效情況。本文的數(shù)據(jù)處理方法只是一種補救方法,有條件要盡量避免無失效數(shù)據(jù)的情況。在進(jìn)行數(shù)據(jù)處理時,也要盡量選定合適的加速模型和失效分布模型。
本文采用的步進(jìn)應(yīng)力加速壽命試驗,施加應(yīng)力較小,以至于無一發(fā)失效,故不存在改變機(jī)理問題。采用的加速模型和失效分布模型是彈藥引信常用的,無失效數(shù)據(jù)處理中采用的先驗分布密度模型[25]也是經(jīng)過驗證有效的。因此,可以認(rèn)為本文的評估方法正確,評估的結(jié)果風(fēng)險有限,結(jié)論可信。
[1] 劉松. 武器系統(tǒng)可靠性工程手冊[M]. 北京: 國防工業(yè)出版社, 1992.
LIU Song. Weapon System Reliability Engineering Manual[M]. Beijing: National Defense Industry Press, 1992.
[2] 胡恩平, 羅興柏, 艾志利. 三參數(shù)威布爾分布條件下的無線電引信步進(jìn)應(yīng)力加速壽命試驗與數(shù)據(jù)處理[J]. 探測與控制學(xué)報, 2000, 22(2): 37-40.
HU En-ping, LUO Xing-bai, AI Zhi-li. The Step Stress ALT and Data Processing of the Radio Fuze under the Condition of Three-Parameter Weibull Distribution[J]. Journal of Detection & Control, 2000, 22(2): 37-40.
[3] 周虎兒, 張湘平, 徐琰. 基于Powell方法的引信步進(jìn)應(yīng)力加速壽命試驗數(shù)據(jù)處理[J]. 探測與控制學(xué)報, 2007, 29(2): 27-29.
ZHOU Hu-er, ZHANG Xiang-ping, XU Yan. Data Processing in Fuze Step-Stress Accelerated Life Test Based on Powell Method[J]. Journal of Detection & Control, 2007, 29(2): 27-29.
[4] 趙河明, 張亞, 董少峰. 無線電引信電子頭部件長貯加速壽命試驗數(shù)據(jù)處理方法探討[J]. 探測與控制學(xué)報, 1999, 21(4): 33-36.
ZHAO He-ming, ZHANG Ya, DONG Shao-feng. The Study on the Data Processing Method for a Long Term Storage Accelerated Life Test of the Electronic Head Assembly of a Radio Fuze[J]. Journal of Detection & Control, 1999, 21(4): 33-36.
[5] 李道清, 王德元. 某無線電引信加速壽命試驗研究[J]. 探測與控制學(xué)報, 2000, 22(4): 57-61.
LI Dao-qing, WANG De-yuan. Study on the Long Term Storage Stepping Stress Accelerated Life Test of a Radio Fuze[J]. Journal of Detection & Control, 2000, 22(4): 57-61.
[6] 張生鵬, 李宏民, 趙朋飛. 導(dǎo)彈裝備貯存壽命加速試驗技術(shù)體系探討[J]. 裝備環(huán)境工程, 2018, 15(2): 92-96.
ZHANG Sheng-peng, LI Hong-min, ZHAO Peng-fei. Accelerated Testing Technology System for Storage Life of Missile Equipment[J]. Equipment Environmental Engineering, 2018, 15(2): 92-96.
[7] 李敏偉, 傅耘, 王麗, 等. 加速貯存壽命試驗設(shè)計方法研究[J]. 裝備環(huán)境工程, 2014, 11(4): 58-64.
LI Min-wei, FU Yun, WANG Li, et al. Study of the Design Method of the Accelerated Storage Life Testing[J]. Equipment Environmental Engineering, 2014, 11(4): 58-64.
[8] 趙方超, 羅天元, 李超, 等. 某型固體云爆劑加速老化試驗與貯存壽命預(yù)測研究[J]. 裝備環(huán)境工程, 2011, 8(6): 34-37.
ZHAO Fang-chao, LUO Tian-yuan, LI Chao, et al. Study on the Accelerated Aging Test and Storage Life Prediction of SE-FAE[J]. Equipment Environmental Engineering, 2011, 8(6): 34-37.
[9] 祝逢春, 胡瑜, 游培寒, 等. 基于實際儲存使用信息的制導(dǎo)彈藥可靠貯存壽命預(yù)測[J]. 彈箭與制導(dǎo)學(xué)報, 2013, 33(6): 197-199.
ZHU Feng-chun, HU Yu, YOU Pei-han, et al. A Reliable Storage Life Estimation of Guided Munitions Based on Storage and Application Information[J]. Journal of Projectiles, Rockets, Missiles and Guidance, 2013, 33(6): 197-199.
[10] 張清爽, 胡雙啟. X彈頭B部件裝藥貯存壽命研究[J]. 火工品, 2003(1): 26-28.
ZHANG Qing-shuang, HU Shuang-qi. Study on the Storage Life of Explosive Part B in X Warhead[J]. Initiators & Pyrotechnics, 2003(1): 26-28.
[11] 張仕念, 易當(dāng)詳, 宋亞男, 等. 固體推進(jìn)劑多失效模式相關(guān)的貯存可靠性評估[J]. 固體火箭技術(shù), 2007, 30(6): 525-528.
ZHANG Shi-nian, YI Dang-xiang, SONG Ya-nan, et al. Evaluation on Storage Reliability of Solid Propellant Based on Correlative Failure Modes[J]. Journal of Solid Rocket Technology, 2007, 30(6): 525-528.
[12] 李田科, 李建華, 劉煒, 等. 一種基于測試數(shù)據(jù)的單枚導(dǎo)彈貯存壽命評估方法[J]. 裝備環(huán)境工程, 2013, 10(6): 88-92.
LI Tian-ke, LI Jian-hua, LIU Wei, et al. A Test Data Based Analysis Method for Single Missile Storage Life Evaluation[J]. Equipment Environmental Engineering, 2013, 10(6): 88-92.
[13] 張福光, 崔旭濤, 洪亮. 導(dǎo)彈火工品貯存壽命的影響因素分析[J]. 裝備環(huán)境工程, 2011, 8(6): 24-27.
ZHANG Fu-guang, CUI Xu-tao, HONG Liang. Influencing Factor Analysis of Storage Life of Missile Initiating Explosive Devices[J]. Equipment Environmental Engineering, 2011, 8(6): 24-27.
[14] 趙東華, 張懷智, 郭勝強, 等. 基于灰色模型的某型末制導(dǎo)彈藥貯存壽命預(yù)測[J]. 裝備環(huán)境工程, 2011, 8(6): 28-30.
ZHAO Dong-hua, ZHANG Huai-zhi, GUO Sheng-qiang, et al. Prediction of Terminal Guided Projectile Storage Life Based on Gray Model[J]. Equipment Environmental Engineering, 2011, 8(6): 28-30.
[15] 黃強, 廉振國, 方延平. 預(yù)測彈藥裝藥壽命常用方法的分析[J]. 彈箭與制導(dǎo)學(xué)報, 1997, 17(2): 60-62.
HUANG Qiang, LIAN Zhen-guo, FANG Yan-ping. Analysis of Common Methods for Predicting Ammunition Charge Life[J]. Journal of Projectiles, Rockets, Missiles and Guidance, 1997, 17(2): 60-62.
[16] 盧明章, 左英英, 劉所恩. 某型導(dǎo)彈彈射發(fā)動機(jī)裝藥的安全貯存壽命預(yù)估[J]. 火炸藥學(xué)報, 2008, 31(1): 72-74.
LU Ming-zhang, ZUO Ying-ying, LIU Suo-en. Safe Storage Life Estimation of a Certain Catapult Missile Motor Charge[J]. Chinese Journal of Explosives & Propellants, 2008, 31(1): 72-74.
[17] 任國周. 固體火箭發(fā)動機(jī)裝藥壽命預(yù)示方法試驗研究[J]. 推進(jìn)技術(shù), 1996, 17(2): 23-26.
REN Guo-zhou. The Research of Forecast Method for Grain Life in Sold Rocket Motors[J]. Journal of Propulsion Technology, 1996, 17(2): 23-26.
[18] 祝逢春, 劉恒春, 游培寒, 等. TNT基炸藥儲存壽命評估研究[J]. 爆破器材, 2016, 45(3): 36-41.
ZHU Feng-chun, LIU Heng-chun, YOU Pei-han, et al. Study on Storage Life Assessment of a Composite Explosive Based on TNT[J]. Explosive Materials, 2016, 45(3): 36-41.
[19] 余文力, 董三強, 朱滿林, 等. 導(dǎo)彈戰(zhàn)斗部炸藥裝藥的貯存可靠性研究[J]. 空軍工程大學(xué)學(xué)報(自然科學(xué)版), 2005, 6(2): 43-45.
YU Wen-li, DONG San-qiang, ZHU Man-lin, et al. Research on Storage Reliability of Missile Warhead Explosive Charge[J]. Journal of Air Force Engineering University (Natural Science Edition), 2005, 6(2): 43-45.
[20] GJB 5103—2004, 彈藥元件加速壽命試驗方法[S].
GJB 5103—2004, Accelerated Life Test Method of Ammunition Components[S].
[21] 盧秋紅, 董少峰, 張亞. 彈藥步進(jìn)應(yīng)力加速壽命試驗數(shù)據(jù)處理方法探討[J]. 探測與控制學(xué)報, 2000, 22(1): 47-50.
LU Qiu-hong, DONG Shao-feng, ZHANG Ya. The Study on the Data-Processing Method of Ammunition Stepstress Accelerated Life Tests[J]. Journal of Detection & Control, 2000, 22(1): 47-50.
[22] 徐振相, 秦士嘉. 火工品可靠性技術(shù): 基礎(chǔ)與管理[M]. 北京: 兵器工業(yè)出版社, 1996.
XU Zhen-xiang, QIN Shi-jia. Reliability Technology of Initiating Explosive Device: Foundation and Management[M]. Beijing: The Publishing House of Ordnance Industry, 1996.
[23] 李明倫, 李東陽, 鄭波. 彈藥儲存可靠性[M]. 北京: 國防工業(yè)出版社, 1997.
LI Ming-lun, LI Dong-yang, ZHENG Bo. Ammunition Storage Reliability[M]. Beijing: National Defense Industry Press, 1997.
[24] 鄭波, 葛廣平. 基于步進(jìn)應(yīng)力加速壽命試驗的引信貯存壽命評估[J]. 北京理工大學(xué)學(xué)報, 2003, 23(5): 545-547.
ZHENG Bo, GE Guang-ping. Estimation of Fuze Storage Life Based on Stepped Stress Accelerated Life Testing[J]. Journal of Beijing Institute of Technology, 2003, 23(5): 545-547.
[25] 祝逢春, 王曉鳴, 崔大偉, 等. 彈藥可靠性試驗?zāi)嫘驍?shù)據(jù)的統(tǒng)計修正[J]. 彈箭與制導(dǎo)學(xué)報, 2006, 26(3): 135-136.
ZHU Feng-chun, WANG Xiao-ming, CUI Da-wei, et al. Statistic Amendable Method of Reverse Order Data of Ammunition Reliability Test[J]. Journal of Projectiles, Rockets, Missiles and Guidance, 2006, 26(3): 135-136.
Evaluation of Storage Life of Fuze via Step-stress Accelerated Life Test with Zero-failure Data
GUO Hua, ZHU Feng-chun, DOU Ren-fu, ZHAI Shu-feng, LI Jing-yu
(Beijing Aeronautical Technology Research Center, Nanjing 210028, China)
The work aims to study the storage life evaluation to solve problem of zero-failure data and no evaluation method in accelerated life test of a fuze. According to the characteristics and sample conditions of a certain fuze, step-stress accelerated life test was performed. Because the sample size was too small, it was impossible to carry out a fundamental test to determine the rational stress and step length, and the accelerated life test can only use its estimated value, so the test result showed zero-failure. In order to solve this problem, the data zero-failure ratio (number of failures/number of samples) was converted into failure data of non-descending failure ratio according to the Bayesian statistics method. Then the storage life was evaluated according to the processing method with failure data. The calculation program was compiled with this method to process the zero-failure test data of a fuze, and the estimated storage life was 20.1 years. The results of a fuze storage life evaluation show that the method of using Bayesian statistical law to deal with zero-failure data is effective and feasible, and solves the problem of the fuze storage life evaluation under the condition of zero-failure data in step-stress accelerated life test.
fuze; step-stress; accelerated life test; zero-failure data; unreliability; storage life; life evaluation
2020-01-22;
2022-05-05
GUO Hua (1978-), Female, Doctor.
TJ410.6
A
1672-9242(2023)02-0026-06
10.7643/ issn.1672-9242.2023.02.004
2022–01–22;
2022–05–05
郭華(1978—),女,博士。
祝逢春(1964—),男,博士。
Corresponding author:ZHU Feng-chun (1964-), Male, Doctor.
郭華, 祝逢春, 豆仁福, 等. 引信步進(jìn)應(yīng)力加速壽命試驗無失效數(shù)據(jù)情況貯存壽命評估[J]. 裝備環(huán)境工程, 2023, 20(2): 026-031.
GUO Hua, ZHU Feng-chun, DOU Ren-fu, et al.Evaluation of Storage Life of Fuze via Step-stress Accelerated Life Test with Zero-failure Data[J]. Equipment Environmental Engineering, 2023, 20(2): 026-031.
責(zé)任編輯:劉世忠