非矩形试验区域综合应力加速寿命试验优化设计的分析

《非矩形试验区域综合应力加速寿命试验优化设计的分析》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、浙江理工大学硕士学位论文非矩形试验区域综合应力加速寿命试验优化设计的研究法。第四章以正常应力下寿命估计值的渐近方差最小为目标，以最低应力水平的两应力为设计变量，针对失效概率最大点（最高应力水平点）唯一和不唯一两种情形，给出了应力水平K3、4、5时的最佳应力组合方式和相应的最优试验方案，进一步完善了非矩形区域E-M试验方案设计理论。第五章提出了试验方案的定量评价指标；建立了最优试验方案综合评价方法；通过计算机模拟试验，比较了各优化模型（均匀、正交、E-M）对应的最优试验方案综合性能优劣，确定了综合性能最优的试验方案（EM-5），并归纳出非矩形区域最优试验方案选择

2、准则，对非矩形试验区域最优试验方案的选择提出指导和建议。最后对全文进行了总结，并对下一步的研究方向提出了自己的看法和建议。关键词：均匀设计、正交设计、非矩形试验区域、加速寿命试验、方案模拟评价II万方数据浙江理工大学硕士学位论文非矩形试验区域综合应力加速寿命试验优化设计的研究ABSTRACTMultiplestressesacceleratedlifetestisacommontechniquetoaccessproductliferapidlyinengineering.Twostressconstantacceleratedlifetestisdesigne

3、dmainlyontherectangleregionformedbytwostresses.However,sometimesonestressislimitedbytheotherstressinengineeringpractice,thatwillinducethetwostressesnottoreachthehigheststresslevelssimultaneously,andthetestregionbecomesnon-rectangle.Forthetwostressacceleratedlifetestonthenon-rectangle

4、testregion,thestresslevelpointswilllocateoutsidethetestregionifdesigningtheoptimaltestplanaccordingtotheideasabouttraditionalrectangleregiontestdesign,sowecannotsurelygetafeasibilitytestplan.Thenitisnecessarytoresearchthemethodsaboutoptimaldesignofmultiplestressconstantacceleratedlif

5、etestplanonnon-rectangletestregionandwaystoaccessthecorrespondingoptimaltestplandesign.Aimingatlinear-extremevaluemodel,thepaperproposedoptimaldesignofmultiplestressconstantacceleratedlifetestplanonnon-rectangleregionbasedonuniform,orthogonalandE-Mdesigntheory;gaveoutthecorresponding

6、rationevaluatingindicatoroftestplan.Itobtainedthetestplandesignofoptimalcomprehensiveperformanceonnon-rectangletestregionbycomputersimulationexperimentandsummarizedthetestplanningselectioncriteria.Thisresearchprovidesthetheoryandmethodfortwo-stressoptimalCSALTplanningandassessmentonn

7、on-rectangletestregions.Thisdissertationiscomposedofsixchaptersandthemaincontentofeachchapterisasfollows:Chapteronedescribedthebackgroundandthesignificanceoftheresearch;reviewedtheALTmethodsanditsdevelopmenttrend;introducedthepresentresearchsituationandexistingproblemsofALTstatistica

8、lmodel,ALTop