[1]王 烨,左洪福,蔡 景,等.基于Bayesian推断和LS-SVM的 发动机在翼寿命预测模型[J].南京理工大学学报(自然科学版),2013,37(06):955-959.
 Wang Ye,Zuo Hongfu,Cai Jing,et al.Forecasting model of engine life on wing based on LS-SVM and Bayesian inference[J].Journal of Nanjing University of Science and Technology,2013,37(06):955-959.
点击复制

基于Bayesian推断和LS-SVM的 发动机在翼寿命预测模型
分享到:

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

卷:
37卷
期数:
2013年06期
页码:
955-959
栏目:
出版日期:
2013-12-31

文章信息/Info

Title:
Forecasting model of engine life on wing based on LS-SVM and Bayesian inference
作者:
王 烨左洪福蔡 景戎 翔
南京航空航天大学 民航学院,江苏 南京 210016
Author(s):
Wang YeZuo HongfuCai JingRong Xiang
College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China
关键词:
贝叶斯推断 最小二乘支持向量机 发动机 在翼寿命 预测
Keywords:
Bayesian inference least squares support vector machine engine life on wing prediction
分类号:
V235; V37
摘要:
为解决发动机的寿命预测精度问题,该文将贝叶斯(Bayesian)推断应用于最小二乘支持向量机(LS-SVM)模型参数的选择,建立了发动机在翼寿命的非线性预测模型。分析了影响发动机在翼寿命的性能参数,建立了用于机器学习的预测模型训练集,构建了基于LS-SVM的发动机在翼寿命预测模型。采用Bayesian推断理论优化LS-SVM建模,获取最优建模参数。通过某型发动机在翼寿命数据集训练模型,对在翼寿命进行预测。与几种常用的算法相比较,该文模型预测精度能够提高4.58%至9.51%,较好地解决了小样本下的预测问题,具有良好的泛化能力和预测精度。
Abstract:
To resolve the problem of engine life forecasting accuracy,a nonlinear forecasting model for engine life on wing is established by applying Bayesian inference to the choices of model parameters of least squares support vector machine(LS-SVM).The performance parameters affecting engine life on wing are analyzed,a forecasting model training set for machine study is established,and a forecasting model of engine life on wing is established based on the LS-SVM.The LS-SVM model is optimized by using Bayesian inference,and the best modeling parameters are obtained.The engine life on wing is forecasted by using a data set training model of a certain engine life on wing.Compared with several common algorithms,the forecasting accuracies of the model proposed here increase by 4.58%-9.51%,which solves the problem of forecasting of small samples,and performs well in generalization ability and forecasting precision.

参考文献/References:

[1] Vittal S,Hajela P,Joshi A.Review of approaches to gas turbine life management[A].10th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference[C].New York,USA:American Institute for Aeronautics & Astronautics,2004:1-11.
[2]Wang W B.A model to predict the residual life of aircraft engines based upon oil analysis data[J].Naval Research Logistics,2005,52(3):276-284.
[3]Xu D,Zhao W B.Reliability prediction using multivariate degradation data[A].Proceedings of the Annual Reliability and Maintainability Symposium[C].Virginia,USA:IEEE,2005:337-341.
[4]谭东宁,谭东汉,小样本机器学习理论:统计学习理论[J].南京理工大学学报,2001,25(1):108-112.
Tan Dongning,Tan Donghan.Small-sample machine learning theory:statistical learning theory[J].Journal of Nanjing University of Science and Technology,2001,25(1):108-112.
[5]Vapnik V N.The nature of statistical learning theory[M].New York,USA:Springer,1999.
[6]余正涛,邹俊杰,赵兴,等.基于主动学习的最小二乘支持向量机稀疏化研究[J].南京理工大学学报,2012,36(1):12-17.
Yu Zhengtao,Zou Junjie,Zhao Xing,et al.Sparseness of least squares support vector machines based on active learning[J].Journal of Nanjing University of Science and Technology,2012,36(1):12-17.
[7]Suykens J A K.Least squares support vector machines for classification and nonlinear modeling[J].Neural Network World,2000,10(1):29-48.
[8]Nello C,John S.An introduction to support vector machines and other kernel-based learning methods[M].Beijing:Publishing House of Electronics Industry,2004.
[9]Kowk J T.The evidence framework applied to support vector machines[J].IEEE Trans on Neural Network,2009,11(5):1162-1173.
[10]MacKay D J C.Bayesian interpolation[J].Neural Computation,1992,4(3):415-447.
[11]Robinson M E,Crowder M J.Bayesian methods for a growth-curve degradation model with repeated measures[J].Lifetime Data Analysis,2000(6):357-374.
[12]邢永忠,吴晓蓓,徐志良.基于矢量基学习的自适应迭代最小二乘支持向量机回归算法[J].南京理工大学学报,2011,35(3):328-333.
Xing Yongzhong,Wu Xiaobei,Xu Zhiliang.Adaptive iterative LS-SVM regression algorithm based on vector base learning[J].Journal of Nanjing University of Science and Technology,2011,35(3):328-333.
[13]University of California Irvine[EB/OL].http://archive.ics.uci.edu/ml/index.html,2011-05-01.

相似文献/References:

[1]王庆超,张健中.基于Hammerstein模型的连续搅拌反应釜非线性预测控制[J].南京理工大学学报(自然科学版),2010,(05):618.
 WANG Qing-chao,ZHANG Jian-zhong.Nonlinear Predictive Control for Continuous Stirred-tank Reactor Using Hammerstein Model[J].Journal of Nanjing University of Science and Technology,2010,(06):618.
[2]朱慧明,韩玉启.基于随机参数的贝叶斯统计质量控制模型[J].南京理工大学学报(自然科学版),2004,(04):445.
 ZHU Hui ming,HAN Yu qi.Bayesian Statistical Quality Control Models Based on the Random Parameters[J].Journal of Nanjing University of Science and Technology,2004,(06):445.
[3]韩玉启,朱慧明.多元质量特性的贝叶斯均值向量控制图[J].南京理工大学学报(自然科学版),2003,(05):561.
 HanYuqi ZhuHuiming.Bayesian Mean Vector Control Chart for Multiple Quality Characteristics[J].Journal of Nanjing University of Science and Technology,2003,(06):561.
[4]林 棋,张 宏,李千目.一种基于MA-LSSVM的封装式特征选择算法[J].南京理工大学学报(自然科学版),2016,40(01):10.
 Lin Qi,Zhang Hong,Li Qianmu.Wrapper feature selection algorithm based on MA-LSSVM[J].Journal of Nanjing University of Science and Technology,2016,40(06):10.
[5]徐 琴.基于误差补偿的物流需求混沌预测模型[J].南京理工大学学报(自然科学版),2018,42(01):126.[doi:10.14177/j.cnki.32-1397n.2018.42.01.019]
 Xu Qin.Logistics demand chaotic prediction model by error compensation[J].Journal of Nanjing University of Science and Technology,2018,42(06):126.[doi:10.14177/j.cnki.32-1397n.2018.42.01.019]

备注/Memo

备注/Memo:
收稿日期:2008-08-12 修回日期:2009-02-11
基金项目:国家“863”计划资助项目(2006AA04Z427); 国家自然科学基金(61079013; 61179066)
作者简介:王烨(1979-),男,博士生,主要研究方向:可靠性分析评估及发动机寿命预测,E-mail:wangye@nuaa.edu.cn; 通讯作者:左洪福(1959-),男,教授,博士生导师,主要研究方向:可靠性分析评估、维修分析评估、发动机健康管理和寿命预测,E-mail:rms@nuaa.edu.cn。
引文格式:王烨,左洪福,蔡景,等.基于Bayesian推断和LS-SVM的发动机在翼寿命预测模型[J].南京理工大学学报,2013,37(6):955-959.
投稿网址:http://njlgdxxb.paperonce.org
更新日期/Last Update: 2013-12-31