Accurate vibration time series modeling can mine the internal law of data and provide valuable references for reliability assessment. To improve the prediction accuracy, this study proposes a hybrid model – called the AR–SVR–CPSO hybrid model – that combines the auto regression (AR) and support vector regression (SVR) models, with the weights optimized by the chaotic particle swarm optimization (CPSO) algorithm. First, the auto regression model with the difference method is employed to model the vibration time series. Second, the support vector regression model with the phase space reconstruction is constructed for predicting the vibration time series once more. Finally, the predictions of the AR and SVR models are weighted and summed together, with the weights being optimized by the CPSO. In addition, the data collected from the reliability test platform of high-speed train transmission systems and the “NASA prognostics data repository” are used to validate the hybrid model. The experimental results demonstrate that the hybrid model proposed in this study outperforms the traditional AR and SVR models.
Li X. High-speed train tracking test and researches on time series model of condition monitoring. Dissertation, Southwest Jiaotong University, China, 2011.
2.
Niu BN. Design and analysis the model of reliability testing equipment of high-speed train transmission system. Dissertation, Jilin University, China, 2012.
3.
Liu T, Liu K and Fang P. A hybrid model of AR and PNN method for building thermal load forecasting. In: Asian simulation conference, 2016, pp.146–155. Singapore: Springer.
4.
CheJWangJ. Short-term electricity prices forecasting based on support vector regression and auto-regressive integrated moving average modeling. Energy Convers Manage2010; 51: 1911–1917.
5.
ZhangYLiCLiL. Electricity price forecasting by a hybrid model, combining wavelet transform, ARMA and kernel-based extreme learning machine methods. Appl Energy2017; 190: 291–305.
6.
ZhangQWangBDHeB. Singular spectrum analysis and ARIMA hybrid model for annual runoff forecasting. Water Resour Manage2011; 25: 2683–2703.
7.
Adnane M and Belouchrani A. On the use of auto-regressive modeling for arrhythmia detection. In: Signal processing conference, 2016. New York: IEEE.
8.
Wang Z, Yang Y and Yang WX. Application study of EMD-AR and SVM in the fault diagnosis. In: Prognostics and system health management conference, (PHM-2014 Hunan), 2014, pp.93–96. New York: IEEE.
9.
ZhaoXHChenX. Auto regressive and ensemble empirical mode decomposition hybrid model for annual runoff forecasting. Water Resour Manage2015; 29: 2913–2926.
10.
Zurita-MillánDDelgado-PrietoMSaucedo-DorantesJJ. Vibration signal forecasting on rotating machinery by means of signal decomposition and neurofuzzy modeling. Shock Vib2016; 2016: 1–13.
11.
LiWXChenGMZhuL. A hybrid prediction model based on empirical mode decomposition and RBF neural network. J Vib Meas Diag2012; 32: 817–822.
12.
GaoYLiY. Fault prediction model based on phase space reconstruction and least squares support vector machines. IEEE Comput Soc2009; 3: 464–467.
13.
SilvaIDDMouraMDCLinsID. Non-stationary demand forecasting based on empirical mode decomposition and support vector machines. IEEE Latin Am Trans2017; 15: 1785–1792.
14.
ZhangJLZhangYJZhangL. A novel hybrid method for crude oil price forecasting. Energy Econ2015; 49: 649–659.
15.
ZhuMTLiuJ. Research on method of determining order of AR model for road roughness reconstruction. Signal Process2010; 91: 25–28.
16.
KallasMHoneinePFrancisC. Kernel autoregressive models using Yule–Walker equations. Signal Process2013; 93: 3053–3061.
17.
LiuSHuYLiC. Machinery condition prediction based on wavelet and support vector machine. Intell Manuf2017; 28: 1045–1055.
18.
XuGTianWQianL. EMD- and SVM-based temperature drift modeling and compensation for a dynamically tuned gyroscope (DTG). Mech Syst Signal Process2007; 21: 3182–3188.
19.
WuHY. Analysis and synthesis of time series, Wuhan: Wuhan University Press, 2004.
20.
WangWCXuDMChauKW. Improved annual rainfall-runoff forecasting using PSO-SVM model based on EEMD. Hydroinform2013; 15: 1377–1390.
21.
PradeepkumarDRaviV. Forecasting financial time series volatility using particle swarm optimization trained quantile regression neural network. Appl Soft Comput2017; 58: 35–52.
22.
TianGZhouMLiP. Disassembly sequence planning considering fuzzy component quality and varying operational cost. IEEE Trans Autom Sci Eng2018; 15: 748–760.
ZhouYWangNXiangW. Clustering hierarchy protocol in wireless sensor networks using an improved PSO algorithm. IEEE Access2017; 5: 2241–2253.
25.
ChengYChengWMZhengY. Structural reliability optimal design based on chaos particle swarm optimization. J Central South Univ2011; 42: 671–676.
26.
Su J. Gearbox reliability test bench for transmission system of high-speed train. Patent 2015207847502, 2016.
27.
Gong HB. Research on reliability test method of high-speed train drive system. Dissertation, Jilin University, China, 2013.
28.
Yang Q. Research on monitoring and early-warning system of reliability testing equipment of high-speed train transmission system. Dissertation, Jilin University, China, 2016.
29.
Wang F, Li X and Zhang B. A study on time series models and criterion rules based on condition monitoring of the tracking test system. In: Proceedings 2013 international conference on mechatronic sciences, electric engineering and computer (MEC), Shengyang, China, 20–22 December 2014, pp.663–667. New York: IEEE.
30.
AwadM. Forecasting of chaotic time series using RBF neural networks optimized by genetic algorithms. Int Arab J Inform Technol2017; 14: 826–883.
31.
LiuXJ. The interval estimation of wind power fluctuation based on the beta distribution. Electric Power Autom Equip2014; 34: 26–30.
32.
TianGZhouMLiP. Disassembly sequence planning considering fuzzy component quality and varying operational cost. IEEE Trans Autom Sci Eng2018; 15: 748–760.
33.
Lee J, Qiu H, Yu G, et al. Rexnord Technical Services, “Bearing Data Set'', IMS, University of Cincinnati #NASA Ames Prognostics Data Repository, NASA Ames, Moffett Field, CA. http://ti.arc.nasa.gov/project/prognostic-data-repository (accessed 9 May 2009).
34.
DongSLuoT. Bearing degradation process prediction based on the PCA and optimized LS-SVM model. Measurement2013; 46: 3143–3152.
35.
LiuYMZhaoCCXiongMY. Assessment of bearing performance degradation via extension and EEMD combined approach. J Central South Univ2017; 24: 1155–1163.
36.
JiangXZhangLChenX. Short-term forecasting of high-speed rail demand: a hybrid approach combining ensemble empirical mode decomposition and gray support vector machine with real-world applications in China. Transp Res C-Emerg Technol2014; 44: 110–127.
37.
ZhaoZChenWWuX. LSTM network: a deep learning approach for short-term traffic forecast. IET Intell Transp Syst2017; 11: 68–75.
38.
Yu W, Su J and Zhang W. Research on short-term traffic flow prediction based on wavelet de-noising preprocessing. In: 9th international conference natural computation, 2014, pp.252–256. New York: IEEE.
39.
WenYWuJYuanY. Multiple-phase modeling of degradation signal for condition monitoring and remaining useful life prediction. IEEE Trans Reliab2017; 66: 924–938.
40.
ChehadeABonkSLiuK. Sensory-based failure threshold estimation for remaining useful life prediction. IEEE Trans Reliab2017; 66: 939–949.
41.
ZhengJPanHChengJ. Rolling bearing fault detection and diagnosis based on composite multiscale fuzzy entropy and ensemble support vector machines. Mech Syst Signal Process2017; 85: 746–759.
42.
MárquezFPGRobertsCTobiasAM. Railway point mechanisms: condition monitoring and fault detection. Proc IMechE, Part F: J Rail and Rapid Transit2010; 224: 35–44.
