基于LSTAR的机载燃油泵多阶段退化建模

李娟; 景博; 焦晓璇; 刘晓东; 戴洪德

doi:10.13700/j.bh.1001-5965.2016.0347

基于LSTAR的机载燃油泵多阶段退化建模

doi: 10.13700/j.bh.1001-5965.2016.0347

李娟^{1, 2, ,},
景博¹,
焦晓璇¹,
刘晓东^{3, 4},
戴洪德⁵

1.
空军工程大学航空航天工程学院, 西安 710038
2.
鲁东大学数学与统计科学学院, 烟台 264025
3.
中航工业金城南京机电液压工程研究中心, 南京 210000
4.
航空机电系统综合航空科技重点实验室, 南京 210000
5.
海军航空工程学院控制工程系, 烟台 264001

基金项目:

航空科学基金 201428960221

详细信息

作者简介:
李娟, 女, 博士研究生, 讲师。主要研究方向:故障预测与健康管理

景博, 女, 博士, 教授, 博士生导师。主要研究方向:故障预测与健康管理、可测试性设计、传感器网络、数据融合

通讯作者:
李娟, E-mail:daidaiquanquan123@126.com

中图分类号: V240.2;TP277
计量
- 文章访问数: 771
- HTML全文浏览量: 158
- PDF下载量: 429
- 被引次数: 0
出版历程
- 收稿日期: 2016-04-28
- 录用日期: 2016-07-22
- 网络出版日期: 2017-05-20

Multi-stage degradation modeling for airborne fuel pump based on LSTAR

LI Juan^{1, 2
, ,},
JING Bo¹,
JIAO Xiaoxuan¹,
LIU Xiaodong^{3, 4},
DAI Hongde⁵

1.
College of Aeronautics and Astronautics Engineering, Air Force Engineering University, Xi'an 710038, China
2.
College of Mathematics and Statistics, Ludong University, Yantai 264025, China
3.
China Aviation Industry Jincheng Nanjing Electrical and Hydraulic Engineering Research Center, Nanjing 210000, China
4.
Aviation Science and Technology Key Laboratory of Aviation Mechanical and Electrical System, Nanjing 210000, China
5.
Department of Control Engineering, Naval Aeronautical and Astronautical University, Yantai 264001, China

Funds:

Aeronautical Science Foundation of China 201428960221

More Information

Corresponding author: LI Juan, E-mail:daidaiquanquan123@126.com

摘要

摘要:
机载燃油泵的性能退化呈现出平稳—加速—平稳的非线性、多阶段模式，针对现有退化模型难以准确描述其全寿命周期性能退化的问题，以逻辑平滑转换自回归(LSTAR)模型为工具，对机载燃油泵出口压力传感器信号进行建模。首先，对转换后的压力传感器信号建立自回归(AR)模型，通过非线性检验说明建立LSTAR模型的必要性；然后，应用非线性最小二乘法完成参数估计；最后，在AIC准则最小及拟合优度最大的原则下，选择转换变量，通过残差进行模型的适应性检验与正态性检验。结果表明：基于LSTAR模型的拟合精度明显优于线性自回归模型。本文提出的方法成功解决了机载燃油泵性能退化的多阶段准确建模问题，为机载燃油泵的预测与健康管理(PHM)奠定了坚实的基础。
- 燃油泵 /
- 传感器 /
- 预测与健康管理(PHM) /
- 逻辑平滑转换自回归(LSTAR)模型 /
- 退化建模
Abstract:
The performance degradation of airborne fuel pump is nonlinear and multi-stage with stationary-accelerated-stationary degradation pattern. The existing degradation models are unsuitable for the modeling of this degradation problem in life cycle, so the signal output from the pressure sensor attached to the fuel pump is modeled based on the logistic smooth transition auto-regression (LSTAR) model. First, auto-regressive (AR) model was established for the converted pressure signal, the necessity of the LSTAR model was examined by nonlinear test, and parameters of the model was estimated by nonlinear least square method. The transfer variable was chosen by minimizing the AIC value and maximizing the goodness of fit. Adaptive test and normality test of the model have been done based on residual analysis. The results show that the LSTAR based method is superior to the AR model. The dividing of the degradation stage and the modeling problem are solved by the presented method, which lays better foundation for the prognostics and health management (PHM) of airborne fuel pump.
- fuel pump /
- sensor /
- prognostics and health management (PHM) /
- logistic smooth transition auto-regression (LSTAR) model /
- degradation modeling

HTML全文

图 1 机载燃油泵压力退化曲线

Figure 1. Curves of airborne fuel pump pressure degradation

下载: 全尺寸图片幻灯片

图 2 机载燃油泵压力退化增量值

Figure 2. Incremental value about pressure degradation of airborne fuel pump

下载: 全尺寸图片幻灯片

图 3 转换函数值

Figure 3. Transition function values

下载: 全尺寸图片幻灯片

图 4 机载燃油泵压力退化预测值

Figure 4. Predictive values for airborne fuel pump pressure degradation

下载: 全尺寸图片幻灯片

图 5 LSTAR模型与AR模型残差Q-Q图

Figure 5. Residual Q-Q diagram about LSTAR and AR models

下载: 全尺寸图片幻灯片

表 1 PP检验

Table 1. PP test

压力序列	统计量	显著性概率
y_t	-2.438	0.132 6
x_t	-72.745	0.000 1

下载: 导出CSV

表 2 模型定阶

Table 2. Model order determination

λ_k	偏自相关系数估计值	AIC值
λ₁	-0.405	2.277
λ₂	-0.392	2.117
λ₃	-0.154	2.105
λ₄	-0.234	2.044
λ₅	-0.157	1.950
λ₆	-0.003	1.948

下载: 导出CSV

表 3 拟合优度和AIC值

Table 3. Goodness of fit and AIC value

转换变量	拟合优度	AIC值
x_t－1	0.379	1.969
x_t－2	0.377	1.972
x_t－3	0.435	1.873
x_t－4	0.374	1.976
x_t－5	0.366	1.989

下载: 导出CSV

表 4 白噪声检验

Table 4. White noise test

延迟阶数	LB统计量	显著性概率
1	0.1671	0.683
2	0.8955	0.639
3	1.4921	0.684
4	1.8461	0.764
5	1.8668	0.867

下载: 导出CSV

表 5 LSTAR模型与AR模型对比

Table 5. Comparison of LSTAR and AR models

模型	误差平方和	J-B统计量	显著性概率
LSTAR	66.867	1.305	0.521
AR	82.529	4.619	0.099

下载: 导出CSV

参考文献(23)

[1]	冯威, 于劲松, 袁海文.机载燃油系统在线实时健康管理[J].北京航空航天大学学报, 2013, 39(12):1639-1643. http://bhxb.buaa.edu.cn/CN/abstract/abstract12804.shtml FENG W, YU J S, YUAN H W.Online real-time health management for aerial fuel delivery system[J].Journal of Beijing University of Aeronautics and Astronautics, 2013, 39(12):1639-1643(in Chinese). http://bhxb.buaa.edu.cn/CN/abstract/abstract12804.shtml
[2]	景博, 杨洲, 迟小泉, 等.系统健康管理及其在航空航天领域的应用[M].北京:国防工业出版社, 2014:3-7. JING B, YANG Z, CHI X Q, et.al. System helath management with aerospace application[M].Beijing:National Defence Industry Press, 2014:3-7(in Chinese).
[3]	TSENG S T, HAMADA M S, CHIAO C H.Using degradation data to improve fluorescent lamp reliability[J].Journal of Quality Technology, 1995, 27(4):363-369. http://cat.inist.fr/?aModele=afficheN&cpsidt=11175582
[4]	BAE S J, KVAM P H.A change-point analysis for modeling incomplete burn in for light displays[J].ⅡE Transaction, 2006, 38(6):489-498. doi: 10.1080/074081791009068
[5]	BAE S J, YUAN T, NING S, et al. A Bayesian approach to modeling two-phase degradation using change-point regression[J]. Reliability Engineering and System Safety, 2015, 134(2):66-74.
[6]	黎明. 带随机拐点的退化失效建模与分析方法研究[D]. 长沙: 国防科学技术大学, 2009. LI M.Reserch on the degradation-failure modeling and analysis methods of degradation path with random change-points[D]. Changsha:National University of Defense Technology, 2009(in Chinese).
[7]	鄢伟安, 宋保维, 段桂林, 等.基于两阶段维纳退化过程的液力耦合器可靠性评估[J].系统工程与电子技术, 2014, 36(9):1882-1886. doi: 10.3969/j.issn.1001-506X.2014.09.34 YAN W A, SONG B W, DUAN G L, et al. Reliability evaluation of LCD based on two-phase Wiener degradation process[J].Systems Engineering and Electronics, 2014, 36(9):1882-1886(in Chinese). doi: 10.3969/j.issn.1001-506X.2014.09.34
[8]	刘君强, 谢吉伟, 左洪福, 等.基于随机维纳过程的航空发动机剩余寿命预测[J].航空学报, 2015, 36(2):564-574. http://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201502017.htm LIU J Q, XIE J W, ZUO H F, et al.Residual lifetime prediction for aeroengines based on Wiener process with random effects[J]. Acta Aeronautica et Astronautica Sinica, 2015, 36(2):564-574(in Chinese). http://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201502017.htm
[9]	LU C J, MEEKER W O. Using degradation measures to estimate a time-to-failure distribution[J]. Technometrics, 1993, 35(2):161-174. doi: 10.1080/00401706.1993.10485038
[10]	BAE S J, KVAM P H. A nonlinear random-coefficients model for degradation testing[J]. Technometrics, 2004, 46(4):460-469. doi: 10.1198/004017004000000464
[11]	徐正国, 周东华.基于马尔可夫链蒙特卡罗的实时可靠性预测方法研究[J].机械强度, 2007, 29(5):765-768. http://www.cnki.com.cn/Article/CJFDTOTAL-JXQD200705014.htm XU Z G, ZHOU D H.Real-time reliability prediction method based on Markov chain Monte Carlo[J]. Journal of Mechanical Strength, 2007, 29(5):765-768(in Chinese). http://www.cnki.com.cn/Article/CJFDTOTAL-JXQD200705014.htm
[12]	SI X S, WANG W, HU C H, et al.Remaining useful life estimation based on a nonlinear diffusion degradation process[J]. IEEE Transactions on Reliability, 2012, 61(1):50-67. doi: 10.1109/TR.2011.2182221
[13]	WANG X L, BALAKRISHNAN N, GUO B. Residual life estimation based on a generalized Wiener degradation process[J]. Reliability Engineering & System Safety, 2014, 124(4):13-23.
[14]	TERASVIRTA T.Specification estimation, and evaluation of smooth transition autoregressive models[J]. Journal of the American Statistical Association, 1994, 89(425):208-218.
[15]	TERASVIRTA T, TJØSTHEIM D, GRANGER C W J. Modelling nonlinear economic time series[M]. Oxford:Oxford University Press, 2010:37-40.
[16]	李娟, 景博, 羌小清, 等.基于样本分位数的机载燃油泵故障状态特征提取及实验研究[J].航空学报, 2016, 37(9):2851-2863. http://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201609023.htm LI J, JING B, QIANG X Q, et al.Fault states feature extraction and experimental study for airborne fuel pumps based on sample quantile[J]. Acta Aeronautica et Astronautica Sinica, 2016, 37(9):2851-2863(in Chinese). http://www.cnki.com.cn/Article/CJFDTOTAL-HKXB201609023.htm
[17]	SANDBERG R. Critical values for linearity tests in time-varying smooth transition autoregressive models when data are highly persistent[J]. The Econometrics Journal, 2008, 11(3):638-647. doi: 10.1111/ectj.2008.11.issue-3
[18]	LUNDBERGH S, TERASVIRTA T, VAN DIJK D.Time-varying smooth transition autoregressive models[J]. Journal of Business and Economic Statistics, 2003, 21(1):104-121. doi: 10.1198/073500102288618810
[19]	王燕.应用时间序列分析[M].北京:中国人民大学出版社, 2008:88-89. WANG Y.Applied time series analysis[M].Beijing:China Renmin University Press, 2008:88-89(in Chinese).
[20]	张晓峒.计量经济学基础[M].3版.天津:南开大学出版社, 2001:274-277. ZHANG X T. Basic econometrics[M].3rd ed. Tianjin:Nankai University Press, 2001:274-277(in Chinese).
[21]	MACKINNON J G. Approximate asymptotic distribution functions for unit-root and cointegration tests[J]. Journal of Business & Economic Statistics, 1994, 12(2):167-176.
[22]	AGENOR P R, MCDERMOTT C J, PRASAD E S. Macroeconomic fluctuations in developing countries:Some stylized facts[J]. The World Bank Economic Review, 2000, 14(2):251-285. doi: 10.1093/wber/14.2.251
[23]	刘大同, 周建宝, 郭力萌, 等.锂离子电池健康评估和寿命预测综述[J].仪器仪表学报, 2015, 36(1):1-16. http://www.cnki.com.cn/Article/CJFDTOTAL-YQXB201501003.htm LIU D T, ZHOU J B, GUO L M, et al. Survey on lithium-ion battery health assessment and cycle life estimation[J]. Chinese Journal of Scientific Instrument, 2015, 36(1):1-16(in Chinese). http://www.cnki.com.cn/Article/CJFDTOTAL-YQXB201501003.htm