-
摘要:
针对传统相似性匹配方法易引入伪相似发动机导致预测精度不高的问题，提出一种结合发动机性能退化特点与相似性匹配特点的区间划分改进相似性剩余寿命(RUL)预测方法。利用堆栈自编码器基于选择的参数构建健康指数；根据测试发动机的已知运行循环进行区间划分，利用传统相似性匹配方法进行初步筛选；对于不同区间下的测试发动机，分别使用不确定性修正和退化一致性检验去除初选参考发动机中的异常发动机，得到最终的剩余寿命预测结果。利用NASA的C-MAPSS数据集进行验证，结果表明：所提方法的预测精度比当前相似性匹配方法提升了34%，证明了方法的有效性。
Abstract:The traditional similarity matching method is liable to introduce pseudo-similar engines, which leads to low prediction accuracy. To address this issue, a remaining useful life (RUL) prediction method of improved similarity with interval partition was proposed, which combined the engine performance degradation characteristics and similarity matching characteristics. Firstly, the health index was constructed based on the selected parameters by using the stacked autoencoder. Then, based on the known running cycles of the test engine, the interval was divided, and the traditional similarity matching method was used for preliminary screening. For the test engines in different intervals, the uncertainty correction and degradation consistency test were performed to remove the abnormal engine from the preliminarily selected reference engines, and the final remaining life prediction results were obtained. The C-MAPSS dataset of NASA was used for verification, and the results show that the prediction accuracy is 34% higher than the current similarity matching method, which proves the effectiveness of the proposed method.
-
图 3 区间划分改进相似性方法流程
Figure 3. Flow of improved similarity method with interval partition
表 1 堆栈自编码器网络参数
Table 1. Stacked autoencoder network parameters
层数 节点数 激活函数 1 4 Sigmoid 2 2 Sigmoid 3 1 Sigmoid 
下载: 导出CSV
表 2 区间划分改进相似性参数
Table 2. Improved similarity parameters with interval partition
参数 数值 时间窗口长度$ W $ 60 距离放缩因子$ \lambda $ 0.002 相似滤波系数$ \gamma $ 0.9 多模型个数$ N $ 20 时间窗口平移系数$ \beta $ 1 伪相似度百分比$ \theta $ 80
下载: 导出CSV
表 3 第89台测试发动机及对应的伪相似度
Table 3. The 89th test engine and its pseudo-similarity
第89台测试发动机
RUL真实值初选参考
发动机序号初选参考发动机
RUL预测值传统相似性匹配方法 退化一致性检验 相似度 相似度阈值 RUL预测值 误差 伪相似度 伪相似度阈值 RUL预测值 误差 28 20 27 0.985 0.906 42 14 1.018 1.234 32 4 54 20 0.988 1.100 59 26 0.997 1.217 80 62 0.991 1.389 91 62 0.992 1.388 52 26 0.992 1.202 49 29 0.987 1.485 95 61 0.994 1.476 $ \vdots $ $\vdots $ $\vdots $ $\vdots $
下载: 导出CSV
-
[1] 中国民用航空局. 中国民用航空发展第十三个五年规划[EB/OL]. (2017-02-16)[2022-08-18]. http://www.gov.cn/xinwen/2017-02/16/content_5168506.htm.Civil Aviation Administration of China. The Thirteenth Five-Year Plan for the development of China’s civil aviation[EB/OL]. (2017-02-16)[2022-08-18]. http://www.gov.cn/xinwen/201702/16/content_5168506.htm(in Chinese). [2] 尉询楷, 杨立, 刘芳, 等. 航空发动机预测与健康管理[M]. 北京: 国防工业出版社, 2014: 132-133.WEI X K, YANG L, LIU F, et al. Aero-engine prediction and health management[M]. Beijing: National Defense Industry Press, 2014: 132-133(in Chinese). [3] ELSHEIKH A, YACOUT S, OUALI M S. Bidirectional handshaking LSTM for remaining useful life prediction[J]. Neurocomputing, 2019, 323: 148-156. doi: 10.1016/j.neucom.2018.09.076 [4] 周俊. 数据驱动的航空发动机剩余使用寿命预测方法研究[D]. 南京: 南京航空航天大学, 2017.ZHOU J. Research on data-driven prediction methods for remaining useful life of aero-engine[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2017(in Chinese). [5] PECHT M. Prognostics and health management of electronics[M]. New York: Wiley, 2008. [6] CHEN X, JIN G, QIU S Q, et al. Direct remaining useful life estimation based on random forest regression[C]//Proceedings of the Global Reliability and Prognostics and Health Management. Piscataway: IEEE Press, 2020: 1-7. [7] 车畅畅, 王华伟, 倪晓梅, 等. 基于1D-CNN和Bi-LSTM的航空发动机剩余寿命预测[J]. 机械工程学报, 2021, 57(14): 304-312.CHE C C, WANG H W, NI X M, et al. Residual life prediction of aeroengine based on 1D-CNN and Bi-LSTM[J]. Journal of Mechanical Engineering, 2021, 57(14): 304-312(in Chinese). [8] 郭庆, 李印龙, 郑天翔. 基于非线性 Wiener 过程航空发动机性能退化预测[J]. 推进技术, 2021, 42(9): 1956-1963.GUO Q, LI Y L, ZHENG T X. Performance degradation prediction of aero-engine based on nonlinear Wiener process[J]. Journal of Propulsion Technology, 2021, 42(9): 1956-1963(in Chinese). [9] ZHANG H, ZHANG Q, SHAO S Y, et al. Attention-based LSTM network for rotatory machine remaining useful life prediction[J]. IEEE Access, 2020, 8: 132188-132199. doi: 10.1109/ACCESS.2020.3010066 [10] WANT T. Trajectory similarity based prediction for remaining useful life estimation [D]. Cincinnati: University of Cincinnati, 2010. [11] 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 Y, WANG C S, LU N Y, et al. Remaining useful life prediction for aero-engine based on the similarity of degradation characteristics[J]. Systems Engineering and Electronics, 2019, 41(6): 1414-1421(in Chinese). doi: 10.3969/j.issn.1001506X.2019.06.32 [12] CAI H S, JIA X D, FENG J S, et al. A similarity based methodology for machine prognostics by using kernel two sample test[J]. ISA Transactions, 2020, 103: 112-121. doi: 10.1016/j.isatra.2020.03.007 [13] YU W N, KIM I Y, MECHEFSKE C. An improved similarity-based prognostic algorithm for RUL estimation using an RNN autoencoder scheme[J]. Reliability Engineering & System Safety, 2020, 199: 106926. [14] XUE B, XU F M, HUANG X, et al. Improved similarity based prognostics method for turbine engine degradation with degradation consistency test[J]. Applied Intelligence, 2022, 52(9): 10181-10201. doi: 10.1007/s10489-021-03034-6 [15] TKACZ E, KOZANECKA D, KOZAN-ECKI Z. Investigations of oil-free support systems to improve the reliability of ORC hermetic high-speed turbomachinery[J]. Mechanics and Mechanical Engineering, 2011, 15(3): 355-365. [16] 王昆, 郭迎清, 赵万里, 等. 基于SSAE和相似性匹配的航空发动机剩余寿命预测[J]. 北京航空航天大学学报, 2023, 49(10): 2817-2825.WANG K, GUO Y Q, ZHAO W L, et al. Remaining useful life prediction of aeroengine based on SSAE and similarity matching[J]. Journal of Beijing University of Aeronautics and Astronautics, 2023, 49(10): 2817-2825(in Chinese). [17] 李浩, 王卓健, 李哲, 等. 基于堆栈自编码器和DeepAR的航空发动机剩余寿命预测[J]. 推进技术, 2022, 43(11): 72-80.LI H, WANG Z J, LI Z, et al. Prediction of remaining useful life of aero-engine based on stacked autoencoder and DeepAR[J]. Journal of Propulsion Technology, 2022, 43(11): 72-80(in Chinese). [18] KHELIF R, MALINOWSKI S, CHEBEL-MORELLO B, et al. RUL prediction based on a new similarity-instance based approach[C]//Proceedings of the IEEE 23rd International Symposium on Industrial Electronics. Piscataway: IEEE Press, 2014: 2463-2468. [19] RAMASSO E. Investigating computational geometry for failure prognostics in presence of imprecise health indicator: results and comparisons on C-MAPSS datasets[C]//Proceedings of the 2nd European Conference of the Prognostics and Health Management Society. Piscataway: IEEE Press, 2014: 1-13. [20] MALHOTRA P, TV V, RAMAKRISHNAN A, et al. Multi-sensor prognostics using an unsupervised health index based on LSTM encoder-decoder[EB/OL]. (2016-08-22)[2022-08-18]. http://arxiv.org/abs/1608.06154. [21] SAXENA A, GOEBEL K, SIMON D, et al. Damage propagation modeling for aircraft engine run-to-failure simulation[C]//Proceedings of the International Conference on Prognostics and Health Management. Piscataway: IEEE Press, 2008: 1-9. [22] 李杰, 贾渊杰, 张志新, 等. 基于融合神经网络的航空发动机剩余寿命预测[J]. 推进技术, 2021, 42(8): 1725-1734.LI J, JIA Y J, ZHANG Z X, et al. Remaining useful life prediction of aeroengine based on fusion neural network[J]. Journal of Propulsion Technology, 2021, 42(8): 1725-1734(in Chinese). [23] SUN W J, ZHAO R, YAN R Q, et al. Convolutional discriminative feature learning for induction motor fault diagnosis[J]. IEEE Transactions on Industrial Informatics, 2017, 13(3): 1350-1359. doi: 10.1109/TII.2017.2672988 [24] WU J, HU K, CHENG Y W, et al. Data-driven remaining useful life prediction via multiple sensor signals and deep long short-term memory neural network[J]. ISA Transactions, 2020, 97: 241-250. doi: 10.1016/j.isatra.2019.07.004 -
下载:
点击查看大图
计量
- 文章访问数: 224
- HTML全文浏览量: 111
- PDF下载量: 23
- 被引次数: 0
