Title :
Wear prediction of metro wheels based on the ARIMA model
Author :
Ling Wang ; Wenjie Zhao ; Hong Xu ; Changjun Chen ; Xiai Chen ; Wenbo Na
Author_Institution :
Coll. of Mech. & Electr. Eng., China Jiliang Univ., Hangzhou, China
Abstract :
Based on the analysis of the measured wear data and the wear characteristics in terms the wheels of Guangzhou Metro Line 1, the accumulative wear prediction method of metro wheels based on the ARIMA(p,d,q) model is proposed in this paper. According to the time series modeling method of the ARIMA(p,d,q) model, the stationarity analysis and transformation of the metro wheel wear data are described at first. Then, with the application of the AIC criterion and the Maximum Likelihood Estimation method, the model order is determined and the model parameters are derived for the ARIMA(p,d,q) model. Finally, the flange thickness and the diameter of the metro wheels could be predicted by this developed ARIMA(p,d,q) model. The results show that the proposed prediction method is simple and effective for the short-term prediction of the metro wheel wear.
Keywords :
autoregressive moving average processes; flanges; maximum likelihood estimation; railway rolling stock; time series; wear; wheels; AIC criterion; ARIMA model; ARIMA(p,d,q) model; Guangzhou Metro Line 1; autoregressive integrated moving average model; flange thickness; maximum likelihood estimation method; metro wheel wear data stationarity analysis; metro wheel wear data transformation; metro wheel wear prediction; time series modeling method; wear characteristics; Analytical models; Data models; Flanges; Predictive models; Rails; Time series analysis; Wheels; AIC Criterion; ARIMA Model; Metro Rolling Stocks; Prediction of Wheel Wear; Short-term Prediction;
Conference_Titel :
Control and Decision Conference (CCDC), 2015 27th Chinese
Conference_Location :
Qingdao
Print_ISBN :
978-1-4799-7016-2
DOI :
10.1109/CCDC.2015.7162388
