Metabolism and difference iterative forecasting model based on long-range dependent and grey for gearbox reliability

被引:9
作者
Liu, He [1 ]
Song, Wanqing [1 ]
Zio, Enrico [2 ,3 ,4 ]
机构
[1] Shanghai Univ Engn Sci, Sch Elect & Elect Engn, Shanghai, Peoples R China
[2] Politecn Milan, Energy Dept, Via La Masa 34-3, I-20156 Milan, Italy
[3] PSL Res Univ, CRC, MINES ParisTech, Sophia Antipolis, France
[4] Kyung Hee Univ, Dept Nucl Engn, Coll Engn, Seoul, South Korea
关键词
Fractional Levy stable motion; Grey model; Metabolism method; Difference iterative form; Gear degradation; REMAINING USEFUL LIFE; PREDICTION;
D O I
10.1016/j.isatra.2021.05.002
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The reliability prediction of gearbox is a complex and challenging topic. The purpose of this research is to propose a hybrid difference iterative forecasting model to forecast reliability of the gearbox. On this score, a hybrid model based on the fractional Levy stable motion (fLsm), the Grey Model (GM) and the metabolism method is proposed. To solve the problem of insensitivity to weak faults inside the gearbox, we use feature extraction method to reveal the gearbox degradation. Then, the least square theory is used to separate the degradation sequence in the gearbox into a deterministic term with monotonicity and a stochastic term with Long-Range Dependence (LRD). Next, the fLsm with LRD and non-Gaussian is used to forecast the stochastic term, the deterministic term is simulated by the GM, and the hybrid forecasting model is used to modify the prediction results. The metabolism method is used to update the degradation sequence and to forecast longer-term trend. Finally, a case demonstrated that superiority and generality of the hybrid forecasting model. (c) 2021 ISA. Published by Elsevier Ltd. All rights reserved.
引用
收藏
页码:486 / 500
页数:15
相关论文
共 49 条
[1]   Attention-based recurrent neural network for multistep-ahead prediction of process performance [J].
Aliabadi, Majid Moradi ;
Emami, Hajar ;
Dong, Ming ;
Huang, Yinlun .
COMPUTERS & CHEMICAL ENGINEERING, 2020, 140
[2]   Reliability prediction of an offshore wind turbine gearbox [J].
Bhardwaj, U. ;
Teixeira, A. P. ;
Guedes Soares, C. .
RENEWABLE ENERGY, 2019, 141 :693-706
[3]   Characteristic function based parameter estimation of skewed alpha-stable distribution: An analytical approach [J].
Bibalan, Mohammadreza Hassannejad ;
Amindavar, Hamidreza ;
Amirmazlaghani, Maryam .
SIGNAL PROCESSING, 2017, 130 :323-336
[4]   Forecasting of time data with using fractional Brownian motion [J].
Bondarenko, Valeria ;
Bondarenko, Victor ;
Truskovskyi, Kyryl .
CHAOS SOLITONS & FRACTALS, 2017, 97 :44-50
[5]   A data-driven ground fault detection and isolation method for main circuit in railway electrical traction system [J].
Chen, Zhiwen ;
Li, Xueming ;
Yang, Chao ;
Peng, Tao ;
Yang, Chunhua ;
Karimi, H. R. ;
Gui, Weihua .
ISA TRANSACTIONS, 2019, 87 :264-271
[6]   Enhanced Particle Filtering for Bearing Remaining Useful Life Prediction of Wind Turbine Drivetrain Gearboxes [J].
Cheng, Fangzhou ;
Qu, Liyan ;
Qiao, Wei ;
Hao, Liwei .
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2019, 66 (06) :4738-4748
[7]   Fault Prognosis and Remaining Useful Life Prediction of Wind Turbine Gearboxes Using Current Signal Analysis [J].
Cheng, Fangzhou ;
Qu, Liyan ;
Qiao, Wei .
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY, 2018, 9 (01) :157-167
[8]   Reliability prediction of machinery with multiple degradation characteristics using double-Wiener process and Monte Carlo algorithm [J].
Cheng, Yiwei ;
Zhu, Haiping ;
Hu, Kui ;
Wu, Jun ;
Shao, Xinyu ;
Wang, Yuanhang .
MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2019, 134
[9]   A novel Switching Unscented Kalman Filter method for remaining useful life prediction of rolling bearing [J].
Cui, Lingli ;
Wang, Xin ;
Xu, Yonggang ;
Jiang, Hong ;
Zhou, Jianping .
MEASUREMENT, 2019, 135 :678-684
[10]   Fault diagnosis and prognosis based on physical knowledge and reliability data: Application to MOS Field-Effect Transistor [J].
Djeziri, M. A. ;
Benmoussa, S. ;
Mouchaweh, M. Sayed ;
Lughofer, E. .
MICROELECTRONICS RELIABILITY, 2020, 110