MA Xiao-jun, REN Shu-hong, ZUO Hong-fu, WEN Zhen-hua. Prediction method of aero-engine life on wing based on LS-SVM algorithm and performance reliability[J]. Journal of Traffic and Transportation Engineering, 2015, 15(3): 92-100. doi: 10.19818/j.cnki.1671-1637.2015.03.011
Citation: MA Xiao-jun, REN Shu-hong, ZUO Hong-fu, WEN Zhen-hua. Prediction method of aero-engine life on wing based on LS-SVM algorithm and performance reliability[J]. Journal of Traffic and Transportation Engineering, 2015, 15(3): 92-100. doi: 10.19818/j.cnki.1671-1637.2015.03.011

Prediction method of aero-engine life on wing based on LS-SVM algorithm and performance reliability

doi: 10.19818/j.cnki.1671-1637.2015.03.011
More Information
  • Author Bio:

    MA Xiao-jun(1962-), male, doctoral student, +86-25-84890647, maxiaojun@comac.cc

    ZUO Hong-fu(1959-), male, professor, +86-25-84891098, rms@nuaa.edu.cn

  • Received Date: 2015-01-17
  • Publish Date: 2015-06-20
  • Based on the monitoring data of practical performance for aero-engine, the degradation model of time-varying performance was established, and the performance trend was predicted. According to the much information related to aero-engine life on wing in the monitoring data, the relation between the performance degradation process and the failure distribution function was analyzed, and the aero-engine life on wing under a given reliability was obtained. Based on the practical life data on wing for aero-engine, the distribution model of life on wing was tested by using K-S fitting test method, and the model parameters was determined by using least squares- support vector machine (LS-SVM). Combined with the performance degradation trend, the revised life on wing for aero-engine was calculated, and example verification on six PW4000 aero-engines was carried out. Analysis result shows that when regularization parameters are 25, 37, 28, 40, 27 and 35 respectively, the practical lives on wing for six PW4000 aero-engines are 6 921, 7 160, 7 820, 8 490, 8 498, 6 921 cycles in order, while the corresponding prediction values are 6 534, 6 726, 7 378, 7 940, 9 103, 6 534 cycles in order. The maximum relative error is 0. 071 190, the minimum relative error is 0. 055 917, and the mean relative error is 0.060 824. The practical engineering requirement can be commendably satisfied by using the proposed method.

     

  • loading
  • [1]
    RONG Xiang. Methods of life prediction and maintenance decision making in civil aero engine health management[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2008. (in Chinese)
    [2]
    WANG Ye, ZUO Hong-fu, 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, 2009, 37(6): 955-959. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-NJLG201306031.htm
    [3]
    HOU Jia-dong, YAN Feng, LI Ming. Reliability analysis to life limit parts of general aviation engine based on Minitab[J]. Aviation Maintenance and Engineering, 2013(2): 74-76. (in Chinese) doi: 10.3969/j.issn.1672-0989.2013.02.036
    [4]
    CHEN Guo-dong, XING Lei, ZHAO Ming. Analysis of life deciding and life extending for aeroengine critical parts[J]. Aeroengine, 2013, 39(5): 60-65. (in Chinese) doi: 10.3969/j.issn.1672-3147.2013.05.014
    [5]
    BAI Fang, ZUO Hong-fu, REN Shu-hong, et al. Research on a smoothing method for engine removal rate[J]. Jouranl of Aerospace Power, 2008, 23(10): 1821-1828. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HKDI200810011.htm
    [6]
    ZHANG Ying, ZUO Hong-fu, REN Shu-hong, et al. Research on an algorithm for prediction of aeroengine performance deterioration[J]. China Mechanical Engineering, 2011, 22(9): 1009-1013. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGJX201109001.htm
    [7]
    SUN Shao-hui, WANG Hua-wei, LI Wei. Predict residual useful life of aircraft engine during stage of potential failure[J]. Aeronautical Computing Technique, 2012, 42(1): 8-11. (in Chinese) doi: 10.3969/j.issn.1671-654X.2012.01.003
    [8]
    PENG Hong-bo, LIU Meng-meng, WANG Yue-ge. Life prediction of engine based on take-off EGTM[J]. Science Technology and Engineering, 2014, 14(16): 160-164. (in Chinese) doi: 10.3969/j.issn.1671-1815.2014.16.029
    [9]
    WANG Hua-wei, GAO Jun, WU Hai-qiao. Residual remaining life prediction based on competing failures for aircraft engines[J]. Journal of Mechanical Engineering, 2014, 50(6): 197-205. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JXXB201406032.htm
    [10]
    LU C J, MEEKER W Q. Using degradation measures to estimate a time-to-failure distribution[J]. Technometrics, 1993, 35(2): 16l-174.
    [11]
    PLACE C S, SRURTT J E, ALLSOPP K, et al. Reliability prediction of helicopter transmission system using stress-strength interference with underlying damage accumulation[J]. Quality and Reliability Engineering International, 1999, 15(2): 69-78. doi: 10.1002/(SICI)1099-1638(199903/04)15:2<69::AID-QRE232>3.0.CO;2-#
    [12]
    TAKED A, SUZUKI N. An empirical model for device degradation due to hot-carrier injection[J]. IEEE Electron Device Letters, 1983, 4(4): 111-113. doi: 10.1109/EDL.1983.25667
    [13]
    MEEKER W Q, LUVALLE M J. An accelerated life test model based on reliability kinetics[J]. Technometrics, 1995, 37(2): 133-146. doi: 10.1080/00401706.1995.10484298
    [14]
    ZHAO Jian-yin. Study on reliability modeling and applications based on performance degradation[D]. Changsha: National University of Defense Technology, 2005. (in Chinese)
    [15]
    SUYKENS J A K, VANDEWALLE J. Least squares support vector machine classifiers[J]. Neural Processing Letters, 1999, 9(3): 293-300. doi: 10.1023/A:1018628609742
    [16]
    LIU Qi. Research of time series prediction method and application based on LS-SVM[D]. Harbin: Harbin Institute of Technology, 2011. (in Chinese)
    [17]
    ZOU Xin-yao, YAO Ruo-he. Application of support vector machine to reliability evaluation of small-sample IC[J]. Journal of South China University of Technology: Natural Science Edition, 2009, 37(1): 22-26. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HNLG200901008.htm
    [18]
    REN Shu-hong. Research on methods of performance reliability assessments and life on wing prediction for civil aero engine[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2010. (in Chinese)
    [19]
    WANG Ye. Research on technique of reliability assessments for civil aircraft products[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2009. (in Chinese)
    [20]
    BAI Fang. Methods of scheduling and condition based maintenance decision making in civil aero engine fleet[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2009. (in Chinese)
    [21]
    GUO Xin-chen. Study on least square support vector machine algorithms and their applications[D]. Changchun: Jilin University, 2008. (in Chinese)
    [22]
    CHENG Li. Application of the Kuhn-Tucker conditions[J]. Journal of Yichun University: Natural Science, 2005, 27(4): 22-23. (in Chinese) doi: 10.3969/j.issn.1671-380X.2005.04.009
    [23]
    GUAN Shan, YAN Li-hong, PENG Chang. Application of regression algorithm of LS-SVM in tool wear prediction[J]. China Mechanical Engineering, 2015, 26(2): 217-222. (in Chinese) doi: 10.3969/j.issn.1004-132X.2015.02.016
    [24]
    WANG Guo-feng, LI Qi-ming, QIN Xu-da, et al. Application of support-vector machine in tool wear of multi-stage monitoring[J]. Journal of Tianjin University, 2011, 44(1): 35-39. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-TJDX201101008.htm
    [25]
    BAO Wen-lei. Research of efficient computing method on GPU-based LS-SVM[D]. Harbin: Harbin Institute of Technology, 2012. (in Chinese)

Catalog

    Article Metrics

    Article views (853) PDF downloads(818) Cited by()
    Related

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return