LASSO based variable selection for similarity remaining useful life prediction of aero-engine
-
摘要:
由于航空发动机监测变量众多,传统方法直接选取性能退化趋势明显的变量进行寿命预测,所以提出一种基于LASSO(least absolute shrinkage and selection operator)的变量选取方法,结合相似性寿命预测方法有效提高了预测精度。基于K-means聚类区分不同工况,对航空发动机多个监测变量根据聚类结果进行变量转换。基于LASSO方法选取最优传感器变量。基于相似性方法进行航空发动机剩余寿命预测。将基于LASSO的变量选取方法与传统的根据退化趋势大小进行选择的方法进行剩余使用寿命预测的结果进行了对比研究。结果表明:基于LASSO选取变量的相似性寿命预测误差的标准差在3种运行周期下分别减少了约1.84、3.46、4.23。
Abstract:Due to the large number of aero-engine monitoring variables, the variables with obvious performance degradation trend were directly selected by traditional method for the life prediction, so a variable selection method based on LASSO (least absolute shrinkage and selection operator) was proposed, which combined with the similarity life prediction method to effectively improve the prediction accuracy. Based on K-means clustering, different working conditions were distinguished, and multiple monitoring variables of aero-engine were transformed according to the clustering results. The optimal sensor variables were selected based on the LASSO method. The remaining useful life of aero-engine was predicted based on similarity method. The results of remaining useful life prediction based on the variable selection method by LASSO and the traditional selection method by the degradation trend were compared. The results showed that the standard deviation of the similarity life prediction error based on the variables selected by LASSO decreased by about 1.84, 3.46 and 4.23 under three operating cycles.
-
Key words:
- prognostics health management /
- K-means clustering /
- LASSO method /
- similarity /
- remaining useful life
-
表 1 6个聚类中心参数表
Table 1. Parameter table of 6 cluster centers
聚类中心 Opsetting_1 Opsetting_2 Opsetting_3 1 10.0030 0.2505 20 2 25.0030 0.6205 80 3 20.0029 0.7005 0 4 35.0030 0.8405 60 5 0.0015 0.0005 100 6 42.0030 0.8405 40 表 2 两种方法预测误差的标准差结果
Table 2. Standard deviation results of prediction errors of the two methods
运行时间 方法 预测误差的标准差 50%周期 LASSO选取方法 24.6605 传统方法 26.4961 70%周期 LASSO选取方法 16.6148 传统方法 20.0720 90%周期 LASSO选取方法 13.8048 传统方法 18.0313 -
[1] LEI Y,LI N,GUO L,et al. Machinery health prognostics: a systematic review from data acquisition to RUL prediction[J]. Mechanical Systems and Signal Processing,2018,104(5): 799-834. [2] 李琪,高占宝,李善营,等. 变工况条件下基于相似性的剩余使用寿命预测方法[J]. 北京航空航天大学学报,2016,42(6): 1236-1243. doi: 10.13700/j.bh.1001-5965.2015.0396LI Qi,GAO Zhanbao,LI Shanying,et al. Similarity-based remaining useful life prediction method under varying operational conditions[J]. Journal of Beijing University of Aeronautics and Astronautics,2016,42(6): 1236-1243. (in Chinese) doi: 10.13700/j.bh.1001-5965.2015.0396 [3] 孟光,尤明懿. 基于状态监测的设备寿命预测与预防维护规划研究进展[J]. 振动与冲击,2011,30(8): 1-11. doi: 10.3969/j.issn.1000-3835.2011.08.001MENG Guang,YOU Mingyi. Review on condition-based equipment residual life prediction and preventive maintenance scheduling[J]. Journal of Vibration and Shock,2011,30(8): 1-11. (in Chinese) doi: 10.3969/j.issn.1000-3835.2011.08.001 [4] WANG Tianyi, YU Jianbo, SIEGEL D, et al. A similarity-based prognostics approach for Remaining Useful Life estimation of engineered systems[C]//2008 International Conference on Prognostics and Health Management. Denver, US: IEEE, 2008: 1-6. [5] YOU M Y,MENG G. A generalized similarity measure for similarity-based residual life prediction[J]. Proceedings of the Institution of Mechanical Engineers Part E: Journal of Process Mechanical Engineering,2011,225(3): 151-160. doi: 10.1177/0954408911399832 [6] WU Y,YUAN M,DONG S,et al. Remaining useful life estimation of engineered systems using vanilla LSTM neural networks[J]. Neurocomputing,2018,275(1): 167-179. [7] 任子强,司小胜,胡昌华,等. 融合多传感器数据的发动机剩余寿命预测方法[J]. 航空学报,2019,40(12): 134-145. doi: 10.7527/S1000-6893.2019.23312REN Ziqiang,SI Xiaosheng,HU Changhua,et al. Remaining useful life prediction method for engine combining multi-sensors data[J]. Acta Aeronautica et Astronautica Sinica,2019,40(12): 134-145. (in Chinese) doi: 10.7527/S1000-6893.2019.23312 [8] LI X,DING Q,SUN J. Remaining useful life estimation in prognostics using deep convolution neural networks[J]. Reliability Engineering and System Safety,2018,172(4): 1-11. [9] WEN Pengfei,ZHAO Shuai,CHEN Shaowei,et al. A generalized remaining useful life prediction method for complex systems based on composite health indicator[J]. Reliability Engineering and System Safety,2021,205(1): 107241.1-107241.15. doi: 10.1016/j.ress.2020.107241 [10] LISTOU E A,EMIL B,VILMAR A,et al. Remaining useful life predictions for turbofan engine degradation using semi-supervised deep architecture[J]. Reliability Engineering and System Safety,2019,183(3): 240-251. [11] CELESTINO O,FERNANDO S L,JAVIER R P,et al. A hybrid ARIMA–SVM model for the study of the remaining useful life of aircraft engines[J]. Journal of Computational and Applied Mathematics,2019,346(1): 184-191. [12] LU F,WU J,HUANG J,et al. Aircraft engine degradation prognostics based on logistic regression and novel OS-ELM algorithm[J]. Aerospace Science and Technology,2019,84: 661-671. doi: 10.1016/j.ast.2018.09.044 [13] 张妍,王村松,陆宁云,等. 基于退化特征相似性的航空发动机寿命预测[J]. 系统工程与电子技术,2019,41(6): 1414-1421. doi: 10.3969/j.issn.1001506X.2019.06.32ZHANG Yan,WANG Cunsong,LU Ningyun,et al. Remaining useful life prediction for aero-engine based on the similarity of degradation characteristics[J]. Journal of Systems Engineering and Electronics,2019,41(6): 1414-1421. (in Chinese) doi: 10.3969/j.issn.1001506X.2019.06.32 [14] LASHERAS F,NIETO P,DE COS JUEZ F,et al. A hybrid PCA-CART-MARS-based prognostic approach of the remaining useful life for aircraft engines[J]. Sensors,2015,15(3): 7062-7083. doi: 10.3390/s150307062 [15] MANGALATHU S,JEON J S,DESROCHES R. Critical uncertainty parameters influencing seismic performance of bridges using LASSO regression[J]. Earthquake Engineering and Structural Dynamics,2018,47(3): 784-801. [16] ZHAO Z,WU S,QIAO B,et al. Enhanced sparse period-group LASSO for bearing fault diagnosis[J]. IEEE Transactions on Industrial Electronics,2019,66(3): 2143-2153. doi: 10.1109/TIE.2018.2838070 [17] 张靖,胡学钢,李培培,等. 基于迭代LASSO的肿瘤分类信息基因选择方法研究[J]. 模式识别与人工智能,2014,27(1): 49-59. doi: 10.3969/j.issn.1003-6059.2014.01.006ZHANG Jing,HU Xuegang,LI Peipei,et al. Informative gene selection for tumor classification based on iterative lasso[J]. Pattern Recognition & Artificial Intelligence,2014,27(1): 49-59. (in Chinese) doi: 10.3969/j.issn.1003-6059.2014.01.006 [18] 曾津,周建军. 高维数据变量选择方法综述[J]. 数理统计与管理,2017,36(4): 678-692. doi: 10.13860/j.cnki.sltj.20170329-001ZENG Jin,ZHOU Jianjun. Variable selection for high-dimensional data model: a survey[J]. Journal of Applied Statistics and Management,2017,36(4): 678-692. (in Chinese) doi: 10.13860/j.cnki.sltj.20170329-001 [19] 宋瑞琪,朱永忠,王新军. 高维数据中变量选择研究[J]. 统计与决策,2019,35(2): 13-16. doi: 10.13546/j.cnki.tjyjc.2019.02.003SONG Ruiqi,ZHU Yongzhong,WANG Xinjun. Research on variable selection in high-dimensional data[J]. Statistics and Decision,2019,35(2): 13-16. (in Chinese) doi: 10.13546/j.cnki.tjyjc.2019.02.003 [20] 雷从英,夏良华,林智崧. 基于相似性的装备部件剩余寿命预测研究[J]. 火力与指挥控制,2014,39(4): 91-94. doi: 10.3969/j.issn.1002-0640.2014.04.022LEI Congying,XIA Lianghua,LIN Zhisong. Research on similarity-based remaining life prediction of equipment components[J]. Fire Control and Command Control,2014,39(4): 91-94. (in Chinese) doi: 10.3969/j.issn.1002-0640.2014.04.022 [21] SAXENA A, GOEBEL K. PHM08 challenge data set[EB/OL]. [2022-05-22]. https://ti.arc.nasa.gov/tech/dash/groups/pcoe/prognostic-data-repository/