欢迎来到天天文库
浏览记录
ID:46781705
大小:632.11 KB
页数:4页
时间:2019-11-27
《飞行器健康状态的灰色预测方法》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、ComputerEngineeringandApplications计算机工程与应用2010,46(26)223飞行器健康状态的灰色预测方法1,2,313313崔建国,宋德胜,李明,陈希成,李忠海,徐长君1,2,313313CUIJian-guo,SONGDe-sheng,LIMing,CHENXi-cheng,LIZhong-hai,XUChang-jun1.沈阳航空工业学院自动化学院,沈阳1101362.东北大学信息科学与工程学院,沈阳1100043.沈阳飞机设计研究所,沈阳1100351.AutomatizationCollege,ShenyangInstituteofAe
2、ronauticalEngineering,Shenyang110136,China2.SchoolofInformationScience&Engineering,NortheasternUniversity,Shenyang110004,China3.ShenyangAircraftDesign&ResearchInstitute,Shenyang110035,ChinaE-mail:gordon_cjg@163.comCUIJian-guo,SONGDe-sheng,LIMing,etal.Greyforecastmethodofaircrafthealthconditio
3、ns.ComputerEngineeringandApplications,2010,46(26):223-226.Abstract:Anewkindofhealthconditionforecastmethodfortheaircraft,basedonthewaveletpackettransformandadap-tiveMulti-VariableGreyForecastModel(MVGFM)ispresentedinthispapertorealizetheveraciousforecasttothehealthstatusofaircraft.Theadvanced
4、AcousticEmission(AE)techniqueisusedtomonitortheaircraftstabilizerhealthconditionandgettheAEinformation.TheoriginalAEsignalsaredecomposedwiththedb4waveletpacket,andtheMaximumofEner-gy(ME),MaximumofVariance(MV),MaximumofNorm(MN)ofthethirdlayerwaveletpacketdecompositionstructurearerespectivelyex
5、tractedtoformeigenvectors.ThentheadaptiveMVGFM(1,n,β)isestablishedbytheeigenvectors.Thepa-rameterβwillberectifiedbytheerrorsbetweentheforecastvaluesandtheactualones.Sotheforecastprecisioncanbeadaptivelyimproved.ExperimentsshowthattheMVGFM(1,n,β)canforecasttheaircraftstabilizerfatiguecrackmore
6、accu-ratelythantheGM(1,1).AndthevalidityoftheMVGFM(1,n,β)isvalidated.Keywords:aircraft;adaptive;waveletpacket;multi-variablegreyforecastmodel摘要:针对飞行器健康状况难以准确预测的问题,提出了一种基于小波包变换与自适应多变量灰色预测模型对飞行器健康状况进行预测的新方法。采用先进的声发射技术监测飞行器关键部件的健康状态,运用小波包对由声发射监测系统募集到的飞行器关键部件原始声发射信号进行三级小波包分解,分别提取其第三级小波包分解中八个频段分解系
7、数的能量最大值、方差最大值和范数最大值作为特征向量,并以此构建三变量MVGFM(1,n,β)模型。运用该模型对飞行器关键部件的健康状态进行预测研究,并通过该模型预测值与特征真实值之间的相对偏差来修正模型中参数β,以提高模型的下一步预测精度。实验结果表明,提出的自适应MVGFM方法可以动态实现对飞行器关键部件裂纹故障的准确预测,其预测准确度明显高于GM(1,1)模型,从而证实了该方法的有效性。关键词:飞行器;自适应;小波包;多变量灰色预测模型DOI:10.3778/j.issn.1
此文档下载收益归作者所有